Get PPC & CRO Knowledge Right To Your Inbox
Covering topics to help your business make more money.

19 Reasons Why Single Keyword Ad Groups (SKAGs)
Always Win [2.0 UPDATE]

UPDATE: This post has received an upgrade. Now we’re showing you how we create and manage single keyword ad groups (SKAGs) since tons of questions have been asked. Scroll down to the “How To” sections and enjoy! 🙂

You don’t have to be a bread lover, like me, to understand the value of a perfectly sliced loaf.

Slicing bread makes it easier to handle, easier to enjoy, and way easier to rub on your face.



This is me…. – GIF source


But what does bread have to do with your Google AdWords account?

Well, you also don’t have to be an AdWords lover to understand the value of slicing up your AdWords account in a more granular way, to make it more manageable, and, to have it perform much better.

If you embrace that, then your AdWords campaigns will start looking like this: 



Just like this.


That’s what makes single keyword ad groups just like sliced bread, and the best thing in AdWords, since, you guessed it,
sliced bread (you were waiting for that one, weren’t ya?).

See, your AdWords account structure is vital to your PPC success.

And because of that, the Single Keyword Ad Group (SKAG) approach is one of the fastest ways to elevate your click-through-rates, quality scores, and most importantly, the money that you’re making.

I have over 350 comments to back up that claim, from when I originally wrote about this tactic back in 2014. Not to mention, hundreds of AdWords accounts across different verticals that are all benefitting from this strategy.

But the best thing? 

This Single Keyword Ad Group approach isn’t just for AdWords, it works for social (Facebook, Twitter, LinkedIn, etc) and display networks as well, and it’s insanely easy to take advantage of.


Otherwise, keep reading…


What Single Keyword Ad Groups Are & Why You Should Care

To put it simply, SKAGs are a way for you to gain more control and achieve higher performance from your AdWords account.

And as the name implies, Single Keyword Ad Groups are ad groups with just one keyword in them. Here’s why:

By pairing your keywords into their own unique ad groups, you can make sure that the keywords you’re bidding on, match the search terms you’re paying for.

If you don’t take care of this, you end up with something I call “The Iceberg Effect” (urge you to read that article). And if you’ve ever watched the Titanic, you know that icebergs are not a good thing.

Like I mentioned earlier, Single Keyword Ad Groups are for one keyword per ad group. But (and this is a big BUT), you’re allowed (and should) include multiple match types of that one keyword in that specific Single Keyword Ad Group.

Let’s take an example.

If you sell custom made Speedos (as many people do), here’s what the keywords inside one of your SKAGs should look like:

+custom +made +speedos
“custom made speedos”
[custom made speedos]

This SKAG will then be called “Custom Made Speedos” (same name as the keywords in the ad group) and will act as the root keyword ad group for those search terms (notice how I didn’t say keywords. Again, read The Iceberg Effect if you’re not sure about the difference).



These are the Speedos you’re selling – image source


If you find that your search terms (and Speedos) are longer tail than the keywords in that SKAG, then you simply extract those search terms and create a new SKAG, following the same formula (this is what we call “keyword refinement”).

Now, once that new SKAG (let’s call it – “Custom Made Glitter Speedos”) is created from the search term report of the “Custom Made Speedos” SKAG, you’ll want to make sure that ‘glitter’ is an ad group level negative keyword to your original “Custom Made Speedos” SKAG.

This will make sure that your shorter tail keywords don’t steal away impressions from your longer tail, more specific keywords.

Now here comes the good part…

Once you’re controlling your search terms and keywords, you’ll have the ability to make your ad copy insanely specific to your keywords.

In my original SKAG article from 2014, I used Nutella as an example to showcase how your headlines should attempt (hoping that you don’t run out of character space) to include the keyword in the headline and display URL, and if not that, somewhere in the description lines.


negative keywords

The keyword for this SKAG is “nutella crepe recipes” – image source

Effectively with SKAGs, you’re reducing your discrepancy ratios from search to keyword, and from keyword to ad (which is the opposite of The Iceberg Effect). 
adwords help

Notice how the Search terms never match the Keyword. That’s bad.


When that happens and your Search terms start matching your Keywords, your click-through-rates (CTRs) can skyrocket because you don’t have 10-50 keywords in one ad group all competing to be relevant to one ad.

And when your CTRs improve, a few other nice benefits happen as well, which we’ll discuss in the Pros and Cons section here in a few.

Here’s what some other AdWords accounts look like when using the SKAG approach:


AdWords skags

Not bad, right?


How The SKAG Approach Works For Social & Display

Not only is the SKAGs approach a solid way of improving managed PPC performance, but it immediately allows you to have a much more stronger foundation than your competitors.

And it’s not just for the Search networks. The same focus of granularity works wonders for social marketing ads and display SEM agency campaigns as well.

If you’ve ever broken down your Automatic placements within an AdWords Display campaign or segmented your audiences and interests in your Facebook campaign, then you can quickly start to see what’s performing well and what isn’t.



Are your Automatic placements all over the place?


You can then break out individual placements and interests to create micro-segments of targeting that produce predictable results week over week.

When making things more granular in your agency PPC account (with SKAGs and other types of micro-segmenting), your next big PPC wins will be on the landing page side, allowing you to stop finagling with silly bid modifications and pointless ad tests that don’t move the needle.

This is because you’ve created such a strong account structure that many of your competitors are too lazy to even attempt it.

They’re still stuck trying to slice their bread with a door stop. N00bs.


adwords guide

Total n00bs – GIF source


How To Create Single Keyword Ad Groups (SKAGs)

There are two things you’ll want before you create your single keyword ad groups: root keywords and AdWords Editor.
Root Keywords: Root keywords are the shortest tail of keywords that you’re willing to bid on.
Let’s say that you’re an online college and you want to use SKAGs to attract potential students. Your root keywords could look like this:
– online college
– online university
– online degree
– online school
In the list of keywords above, you’ll notice that these are root keywords because you wouldn’t want to bid on “online” by itself or “school” by itself. They’re the shortest of short tail keywords that you want to use.
Now, take it one step further.
Take your root keywords and multiply them with other synonyms that you might find. Your original root keyword list might look like this:
– online college
– online university
– online degree
– online school
– online academy
– online seminar
– online classes
– online courses
Then, copy and paste and replace “online” with “web”:
– web college
– web university
– web degree
– web school
– web academy
– web seminar
– web classes
– web courses
If you have more synonyms, then keep going:
– digital college
– digital university
– digital degree
– digital school
– digital academy
– digital seminar
– digital classes
– digital courses
Here’s the best part: you don’t need to do any keyword research around average CPCs or competition levels, because a SKAG set up allows you to get out of the gates with a high quality score. So research here is a waste of time. Only research the synonyms and the like to expand your root keyword list.
Now that we have those done, next step is to use AdWords Editor.
If you don’t have AdWords Editor installed yet, then grab it here – you’ll need it for this next step.
Next step is to open up AdWords Editor and download your AdWords account.
Once there, we’ll create our first new SKAG. Here’s how it will be set up:

The Keywords:

You’re going to use modified broad match, phrase match, and exact match around the same keyword (you’re not going to use broad match).
So it would look like this:

+online +college
“online college”
[online college]

The reason why you want three match types is two-fold:
1) Even if someone searches for a longer tail version search term that is triggered by the phrase or modified broad match type keyword, then your ad will still be the most relevant ad possible in your account.

2) If you split up your SKAGs by match type (turning one ad group into three), then you further dilute your data collection and ad testing will be even harder. SKAGs by nature already dilute your ad testing potential, so don’t worsen it.

Next, you’ll want to set all your bids at the same amount. Then adjust the bids later based around performance.

The Ads:
Once you’ve set up your keywords and their match types, it’s time to create your ads.

Here are the only rules you need to follow:
1) Keep the keyword in the headline and path (used to be called Display URL). This will make the ad more relevant to the searcher leading to a better chance of higher CTR, which leads to a higher quality score and many other benefits.

2) Anything you write in headline 2 and the description is up to you. You just want to make sure that the ads are different enough to split test. It’s okay if your headlines and paths are the same for the two ads (could be from your keyword being too long and takes up too many character spaces), just aim for difference between the two ads when it comes to headline 2 and the description.


The Cloning:
Here comes the fun part with AdWords Editor, the cloning.

With AdWords Editor, you can quickly copy and paste ad groups so you don’t have re-write or re-work much.

As you can see in the GIF below, all you manually need to create is the first SKAG, and then you can go crazy with cloning them, which is why you need your list of root keywords first.

In AdWords Editor, you can literally create all your “online” SKAGs, and then copy and paste all of them and just replace “online” with “web” or “digital”.

Pretty easy, right?


How To Optimize Single Keyword Ad Groups (SKAGs)

The beautiful thing about single keyword ad groups (SKAGs) is that improving their performance is the simplicity.

A lot of people worry about creating hundreds of different variations of their keywords and if they should include singular/plural variations.

Like this guy

And the answer is no. All he needs is to have the SKAG of “dry cleaners” and the search term report will take care of the rest.

Because “dry cleaners” is his root keyword, the modified broad match and phrase match versions of that keyword will automatically trigger the longer tail keywords in his list above.

So why create the SKAGs if there’s no search volume for that keyword? The search term report will tell you if there is.

If you look at the example GIF below, you’ll notice that the search terms here are variations of the keyword “electric bike”.


I have a little text sheet for me to write down the search terms that don’t match the keyword (this is a discrepancy we’re intentionally looking for) – these are the new SKAGs I’ll create.

Quick Note: Notice how I skipped the search term “electric bicycle electric bicycles”? That’s because it’s a bot search that usually has very low CTR. I’m not going to create a SKAG for that.

Essentially, you’re extracting search terms to turn them into SKAGs based around the impression volume in descending order.

Then, you’ll simply add those search terms as ad group level negative keywords (to eliminate internal competition) and your performance will continue to improve over time because your new SKAGs have more specific ads which leads to higher CTRs.

That’s it!


The Positives of Single Keyword Ad Groups (SKAGs)


1) Easiest Blueprint For Success

When you create all new campaigns in the Single Keyword Ad Group format, then you make it very easy to choose and judge where your keywords should go.

There are no themes or groupings of keywords to consider. Since you’ll never add new keywords to a SKAG, you’ll only be creating brand new SKAGs.

This makes your campaign creation blueprint super simple to follow and easy to teach.


2) Compounding Improvements

The AdWords graph screenshots I shared earlier show early improvement from using the SKAG methodology. But eventually, you’ll hit a point of diminishing returns.

But diminishing returns in this case are more like compounding improvements.

As you continue to create SKAGs in descending order of Search term impressions, you’ll find that over time, your overall account health continues to get stronger as all the important metrics in your account, start improving.


3) Improved Click-Through-Rates

When you lower your discrepancy ratio between Search terms to Keywords and Keyword to ad, you immediately start to see that visitors are more likely to click on your ads because you’re using your ad as a “keyword holder”.

skags adwords

Notice Vitacost’s headline and the bolded display URL.


This means that your click-through-rates (CTRs) will improve since no one reads all the ads in their entirety before clicking.

And when you improve your CTRs, other magical things happen:


4) Improved Quality Scores

You hopefully know by now that the biggest part of the AdWords Quality Score is CTR.

Once you’re able to improve your Quality Scores with SKAGs, you’ll start seeing that your average positions, cost per clicks, impression shares, and other great things improve as well.

If you’re not familiar with the AdWords Quality Score, then read this post:
29 AdWords Quality Score Factors – Facts, Myths, and Quick Fixes


5) Improved Average Positions

Once your CTRs and Quality Scores start improving, you’ll notice that Google starts rewarding you with higher average positions.

This is where you’ll start outranking your competitors, even if your keyword bid is lower.

That’s because Google cares a ton about visitor experience. And your improved CTRs are a great indicator that people are liking the ad results that you’ve been working hard to create.


6) Improved Impression Shares

When your Quality Scores and your average position improve, then your impression shares start growing as well.

This means that of the impressions available for certain keywords, your ads are showing more often and more frequently.

Especially now that there are no longer 11 text ads on (but a max of seven), it’s even more important to care about your impression shares and how they may be declining.


7) Lowered Cost Per Clicks

If you’ve ever wanted to lower your average cost per clicks (CPCs), without lowering your bids, then this will make you extremely happy.

One of the great benefits of SKAGs that improve Quality Scores, is that improved Quality Scores give you lowered First Page Bids (FPBs) as you can see in Optmyzr’s calculation below:

keywords help

The lower your Quality Score drops, the higher your costs – image source


This is because Google will discount your CPCs when they see that your Quality Scores are higher than your competitor, with your relevancy being better too.


8) Lowered Cost Per Conversion

When you lower your cost per clicks, you ultimately lower your cost per conversions. No more explaining needed here.

But here are two cats that enjoy bread as much as my wife and I do.

lower cost per conversion

Acquire bread at all costs – GIF source


9) Potentially Higher Conversion Rates

When you expose Search terms as new Keywords and decrease your Search term to Keyword discrepancy ratio, then you immediately increase your control and are able to bid differently for different keywords that have different conversion rates and intents.

This also helps you pause any wasteful Keywords or lower bids on lower performing Keywords that then increases your conversion rates because more of your clicks have a higher likelihood to convert.


10) Increased Conversion Volume

Higher CTRs mean more clicks, and more clicks mean more conversions.

That’s so simple that you can high-five your mom after coming home from the grocery store with the biggest loaf of bread.

increased conversions

Winning – GIF source


Side note: I’m seriously impressed with how many amazing bread GIFs there are.


11) Less Ad Spend Waste

When all your important AdWords metrics are moving in the right direction and your Search terms are being exposed, then you’ll find that you’ll have to add less and less negative keywords over time.

Having your Search terms as keywords also helps you understand the closing and sales rates of your keywords (not conversion rates) if you’re in the SaaS or lead gen space.


12) Increased Levels of Control

Single Keyword Ad Groups and their level of granularity mean that you’re giving Google less control and taking over yourself.

Since Google recommends that you use 10-20 keywords per ad group, you’re immediately able to make sure that all your keywords have less Search term variance.


Bad advice from Google, but that’s how they make more money.


13) Troubleshooting Improvements

With Single Keyword Ad Groups, it will be much easier and more black and white to figure out why certain keywords or ad groups are stealing away impressions or not showing up on Google at all.

This is because you’ll know that every single ad group will only have one keyword in it, making it quick and easy for you to run a keyword diagnosis to see what might be wrong.


14) More Money Being Made

Now, with all the great positives we just covered, you and I both know that there’s no better metric than the ultimate one – money being made.

When you lower your cost per conversions, you immediately increase your profit margins.

When you increase your conversion volume, you can immediately increase your revenue as well.


The Negatives of Single Keyword Ad Groups (SKAGs)


1) Data Dilution

There are a few down sides to Single Keyword Ad Groups. But trust me, the positives severely outweigh the negatives.

One of the bigger negatives around SKAGs is data dilution.

Because you’re creating more and more ad groups, your data is spread more thin, which means it can take you longer to test ads.

But don’t worry, SKAGs is a scientifically more accurate way of testing anyways (to isolate one variable at a time).


2) Takes More Time

It’s no mystery that creating Single Keyword Ad Groups takes much more time than just adding regular keywords to an existing ad group from the Search term report.

And this is what has many people questioning to begin with creating SKAGs in the first place.

The fastest way to remove this fear however, is to test the SKAG approach in your top five keywords and compare their metrics (apples to apples) against the same keywords who are not set up in SKAGs.

If performance does improve, then you have the proof to continue.


3) Takes More Effort

To get the biggest benefit from SKAGs, then they do require you to keep a closer eye to Search terms that don’t match the keywords and to use ad group level negative keywords to make sure your shorter tail SKAGs aren’t stealing away impressions, like the cats were stealing away bread earlier.


4) Ad Testing Can Take Longer

As I mentioned in the Data Dilution section, your actual ad testing can take longer since you’re not getting a condensed level of clicks from multiple keywords in the same ad group.

If you find that you’re constantly able to test ads with continual improvements (which is rare), then you could take your high traffic volume keywords as SKAGs and keep the lower traffic volume keywords in broader ad groups.


5) Stupid People May Laugh At You

People who are lazy will always be hating on people who get things done. Just look at Jon here:

single keyword adgroups

Don’t be hatin’


Moral of the story? Don’t worry about Jons. Worry about your account and only doing things that actually make you more money.

I’ll bet that SKAGs will definitely help with that endeavor 😉


In Closing…

At the end of the day, you want to make more money. And the SKAGs approach (as proven above) will definitely help you get there.

Because just like myself and Oprah, you gotta eat.

So here’s to getting more bread…

skags advice

You and me both, Oprah – GIF source

PPC Management
That Continues To Perform

Klientboost Blog Author <br>Johnathan Dane

Johnathan Dane


  • Natalie Goguen

    You have multiple ads per SKAG, correct? Thanks so much, great article! I can’t wait to give it a try.

    • Always at least two 🙂

      Otherwise, you’re never testing ads.

  • Michael

    Any benefit to have 1 keyword match only in 1 adgroup? For example 1 adgroup with [fresh apples] in exact match, 1 adgroup with “fresh apples” in phrase match (with exact [fresh apples] negatibe) and 1 adgroup with +fresh +apples broad match (with “fresh apples” negative) ? That setup will force Google to select proper match. What do you think? Any benefit here? Also you can set mobile bid modifier for every match type …

    • Hey Michael 🙂

      It’s overkill, and will unfortunately dilute your data even further.

      If you follow the above approach, then Google does a pretty accurate job at giving the right match type the click credit from the specific search term.

      • Michael

        Hi Johnathan, Thanks for clarifying this. One more question, not related to this post. Do you guys use any “fraud click” software for your clients and if you do, had you ever got any “money back” from Google / Bing ?

        • Haven’t really found any abnormal activity outside of AdWords’ own reporting, so the answer is no 🙂

  • Pingback: SEMówka #23 - Czyli Co Przeczytać W Tym Tygodniu - AdWords | Blog Deva Group Kraków()

  • Pingback: SEMówka #23 – czyli co przeczytać w tym tygodniu | Aktualności SEO, Blogi SEO, Pozycjonowanie()

  • Gaby Martinez

    Hi Jonathan,

    Great read! I tried SKAGS after reading the first article you released in 2014, and I must say I am impressed by how much improvement resulted from implementing this strategy. My CTRs skyrocketed, a good chunk of my quality scores are perfect 10s and the best part is that my CPCs decreased while achieving dominant ad rank. SKAGS plus localization are a game A strategy. On another note, that comment Jon made about having hundreds of ad groups is hilarious, because If you know PPC, then you know that’s the best route to take in order to improve performance. I am a proud 200+ ad group implementer, lol, there’s no denying the increased conversions and revenue at the end of the day.

  • Det

    Hi John,
    For SKAG how to handle plural and singular form? For example my keyword is only ‘thermocouple’, should I put the variations: [thermocouple] [thermocouples] “thermocouples” “thermocouple” +thermocouple +thermocouples ? Thank you.

    • Keep them as separate SKAGs 🙂
      Over time, you will start seeing a difference in performance between singular and plural keywords.

      • Ethan

        Hi Johnathan! Thank you for sharing this.
        Further to Det’s question. How do we create negative keywords in the case of 2 SKAG and

  • Paul

    Hi Johnathan. Great advice, thanks. Your blogs are clearer than anything I’ve read before. We have a huge database of products and therefore keywords (millions). From what I read: I still need to go with SKAGs (after google specifically advised us to go with keyword-urls instead) but I am concerned about the feasibility of adding negatives at the Ad Group level unless I do this in an spreadsheet-automated way in the first place – to take care of the bulk – and then refine with my top performing ads. What I am driving at: can I add negatives in advance based on the the other keywords that would, in a non-SKAG world, have been in the same group. Example: Adgroup with the “Chrome Rain Shower Head” (bmm, exact, phrase) and add in as negatives,in advance, all other variations of “Chrome Shower Head” “Shower Head” and “Rain Shower Head”. That Ok? Thanks.

    • Thanks Paul 🙂

      Don’t worry about adding ad group level negatives until you see the discrepancy in your search term report after you’ve created your SKAGs. Otherwise it becomes a super messy spiderweb.

      You don’t need to add in those ad group level negatives to the SKAG of Chrome Rain Shower Head. Because BMM makes sure that you always have to have at least those 4 words in the Search term.

      Instead, you’d be adding these ad group level negatives to your Shower Head SKAG:
      – chrome
      – rain

      But again, no need to do it before the updated Search term report tells you. Internal competition is not a big deal, and it’s easier to fix when you can just click a little checkbox and then add them as ad group level negatives.

      You can then continue to use the Search term report to further build out your SKAGs by focusing on the descending order of impressions.

      Hope that helps! 🙂

  • Kasia Subieta

    AMAZING READ! I’ve learned so much and your writing ability made it so easy. Quick question, how to do you handle word order? If I’m understanding this correctly, I would do 2 separate ad groups like this:

    +CPA +firm +Chicago
    “CPA firm Chicago”
    [CPA firm Chicago]

    +Chicago +CPA +firm
    “Chicago CPA firm”
    [Chicago CPA firm]

    Is that correct?

    • Thanks so much Kasia 🙂

      Your first broad match modified keyword of +CPA +firm +Chicago would trigger both search terms of:

      CPA firm Chicago
      Chicago CPA firm

      Now, if your Search term report that people actively search for both versions, then do create them as separate SKAGs like you were asking 🙂 No need to create the second one if your Search term report doesn’t show that people are searching that way.

      Hope that helps!

  • If your taking over an account with over 2,500 keywords is the best approach to re-structure the account by creating Single Keyword Ad Groups from all the keywords that have conversions over the history of the account? Then would you add Single Keyword Ad Groups from the Search Term Report each week or each month after that?

    Also, just wondering when running an international campaign, do you find it best practice to have 1 adwords account per country?

    • That depends on how quickly you can deem SKAGs valuable or not 🙂 I would recommend starting off with a few of them and compare apples to apples. Then you decide whether or not it’s worth the time.

      I’ve heard people say one AdWords account per country, but logically it makes no sense to me as to why that’s a good idea 🙂 Do you know of any?

      • Thanks. Starting with a few as a test is a good idea. Some of the reasons I came up with for doing 1 account per country were:

        >> Allocate AdWords marketing budget to each country in it’s own currency
        >> Set a CPA goal for each country in the local currency and optimise for it
        >> Optimise bids in the local currency for keywords in specific countries and regions where competition is different
        >> Report to country level marketing managers in their own currency

        • If the main marketing hub is in one country, then I would imagine all the goals are in your currency, no?
          I’m guessing that the currency situation is the only upside for individual accounts 🙂

          Thanks for sharing, Chris!

          • Yeah, I think 1 account should be no problem at the start with small ad budgets. But it is good to have the currency right so you don’t need to worry about month-to-month changes in exchange rates affecting bigger marketing budgets. Also, just from an international expansion perspective there could be a point where each individual country is in charge of managing their own adwords, so would want their own account.

  • Great tactic and a bit intimidating to apply at first.

    Does it make sense to group branded keywords into SKAGs? I mean, if you don’t gather high CTRs and Conversions for your branded terms, then you are probably being outbid by your competitors.

    So, my idea is to have standard ad group structure on branded terms and SKAGs on non-branded terms.

    • Why not?

      If you are trying to eventually track sales (not just leads), then it’s a must. Most of the time you’ll quickly figure it out by testing it and see what happens.

      • Makes sense to test-run it. Would you go about tagging each landing page/final destination URL with unique tracking parameters for even more granular attribution capabilities?

  • When you compare apples to apples (SKAG vs keyword not set up in SKAG), does this ever cause your ads to compete with one another for impressions? Have you ever come across a scenario where you compared apples to apples and the SKAG lost?

    • Not when it lost, but ads compete internally in an ad group on purpose, and then you have internal competition between ad groups where short tail keywords steal away impressions from long tail keywords.

      Ad group level negatives help remedy this 🙂

      • Thanks. So when you test the keyword in the Single Keyword Ad Group, should you pause that same keyword in the other ad group it is in, so there is no internal competition while you run the test?

        • Use a negative keyword list to add to your old campaigns with the broad match keywords from your new campaigns 🙂

          • Just to clarify… if your keyword is +custom +made +speedo and the search term report shows conversions for the plural “custom made speedos” so you make a new ad group for that. What ad group level negative keyword and match type would you add to the Custom Made Speedo ad group to stop impression stealing?

  • Jonathan, what’s your reasoning behind using the 3 match types in 1 ad group (instead of just BMM where the + sign allows the term to be read in a different order as well as in Exact or Phrase match)?

    • They all perform differently when exposed and targeted 🙂

      • Thanks, that makes good sense, I’m guessing quality score could be very different for different match types as well. When you add a long tail SKAG from the search term report, do you add the full long tail keyword as an exact match negative keyword at the ad group level of the short tail SKAG? Or do you have an example of what ad group level negative keyword AND match type you would use?

  • Rami Janoudi


    After reading your old article some time ago, i have been using this strategy for a while. However, recently I ran into a situation which made me think twice. Let me know what you think:

    Search term: cheap fitness class in chicago

    SKAG #1: +cheap +fitness +class “cheap fitness class” [cheap fitness class]
    SKAG #2: +cheap +fitness +class +in +chicago “cheap fitness class in chicago” [cheap fitness class in chicago]

    “cheap fitness class”: max CPC = $2.00, QS = 8 –> Max AdRank = 8 x 2 = 16
    [cheap fitness class in chicago]: max CPC = $2.20, QS = 6 –> Max AdRank = 6 x 2.20 = 13.2

    If I don’t have any Adgroup level negatives, Adwords selects “cheap fitness class” rather than the long tail exact match keyword. In this case, the phrase match with a Max AdRank 16 enters into the auction.

    Now, if I were to make “chicago” an adgroup level negative for the “cheap fitness class” adgroup, then I would indeed end up with an exact match keyword being selected to go into the auction. However, that exact match keyword would have a lower AdRank: 13.2. Therefore if I understand correctly, I wouldn’t be as competitive in the auction as if I had entered with AdRank of 16 with the phrase match head term.

    In other words, it seems having an exact match keyword enter the auction isn’t always the most competitive, and I may lose impressions if I use this method (or have a lower position).

    Let me know if I’m missing something or if I’m on the right track!



    • Looks like your math is good, but you’re missing the ad extension portion of the ad rank calculation. That being said, will your QS stay steady once your CTR potentially improves because your Chicago ad is more specific for the search term? Probably.

      So you’ll have to try it out no matter what to see what happens 🙂

      • Rami Janoudi

        Forgot to mention this is for a Call Only campaign…so no Ad Extensions in this case. Do you recommend using just +Chicago as the negative keyword in the head term SKAG? Or +fitness +class +Chicago? etc… Thanks again!

  • Greg


    What are the pros/cons of your method vs. using the method that this Altus company teaches where you only use the exact match in your SKAG?

    • Hey Greg, exact match only SKAGs don’t allow your modified broad match and phrase match keywords to mine for longer tail keywords for more conversion volume that you can further extract and improve your CTRs with even more.

  • Hi Jonathan! Great read. I’m already doing SKAG on most of my accounts. One question has been bothering me and I’m wondering what you think: How do you deal with keywords with low volume?

    These keywords are big enough not to show “Low Search Volume” but the volume on them is still relatively small. Would you still create a single ad group for each of them? Or try grouping them into the “shorter” tail keyword?

    How would you avoid having accounts with hundreds of Ad Groups making it very difficult to manage overlapping keywords and ad group negatives?

    • Hey Pierre 🙂

      Usually we wouldn’t create SKAGs with low volume impressions. So the next ones we create are all dictated by the volumes from the search term report.

      If I do turn them into SKAGs, then I would never group them together. That only helps you save time, but doesn’t give it the best chance of getting that click.

      For your last question, just follow the search term report and check off the box to search terms and keywords that don’t match to create ad group level negatives and create new SKAGs from those search terms. That’s the easiest way to avoid to negative keyword spider web 🙂

  • Luke Austin

    Enjoyed reading this. Care to briefly discuss the recommended campaign layout for using SKAGS? Do you recommend a new campaign for each “core term” or just one campaign named whatever and X amount of ad groups with the “core term” as the ad group name?

    • Yep, campaign structure is completely up to you and has no impact on the ad groups.

  • With the SKAG approach does that mean you would have to create 100’s of campaigns? There could be 1000’s of combinations of keywords for a particular subject.

    • Follow the search term report 🙂 No need to create for the sake of creating as you’ll eventually hit a point of diminishing return.

      • Thanks Johnathan. So you start with the highest volume search terms and create a campaign for each one.

        • It’s not Single Keyword Campaigns, it’s ad groups 🙂

  • How can you implement a specific ad copy in your ad campaign successfully? The biggest challenge faced in the single keyword ad group management is writing, testing its creativity and then successfully implementing it in your campaign.

    • You start off with creating at least 2 ads that are competing against each other and can apply to all SKAGs in the campaign. Then you fine-tune at the individual ad group level once data starts rolling in 🙂

      • fritalci

        Curious to see your response to what I wrote above.

    • fritalci

      This is an area where I have loved using Adwords Editor to upload spreadsheets full of ad copy. I create one sheet where column A has the campaign name and column b has the ad group name. The remaining columns are for each ad copy field. I then just copy and paste my original ad into each row by highlighting all of the rows. Then I just upload. After a few weeks go by, I export the data. For the ad groups with a lot of data, I study the data in each ad group and make decisions about which of the two ads to pause. For the ad groups that didn’t have enough data to make me confident of a decision, I create a pivot table to see how each ad copy did on average. I did this a few years ago for an account that had over 100,000 long tail SKAGs (yes, you read that correctly) and we were able to make big improvements.

  • Integra Crane

    Hello Jason,
    Best adwords strategy post I have ever read! although I kinda late to found it 🙁
    My question jason, what your suggestion when my keyword consist only one word? its like the head of a product category and a lot of people search using this keyword. Thanks, sorry for my bad english

  • Pingback: The Ultimate List of 300+ Top Growth Hacking Tips and Tricks – Standuply Blog()

  • Vlad B

    Hey Johnathan,
    Really great article. Overall quality with precious information included. Good work.
    I have a question, though.
    Should I make two different SKAGs for <> and <> or should I go for a different tactic?
    Hope you can help.

    • You should make two different SKAGs if both of those search terms have enough volume to warrant it.

  • Billy Robinson

    Hi Jonathan, I just read your ppc automation article which was great. I notice you mention you can have adwords scripts automatically create SKAGS with starting ad copy. Have you covered this anywhere? How do I go about learning this? Thanks!

    • Might come in our 3.0 roll out of this article.
      Stay tuned 😉

  • Pingback: Product Listing Ads [Gifographic]: How to Increase ROAS & Sales | NEWZE()

  • Pingback: Product Listing Ads [Gifographic]: How to Increase ROAS & Sales()

  • Pingback: 5 Little-Known Tips for Improving the Performance of Your Campaigns [Infographic] | Tech Reviews by Anna()

  • Pingback: 5 Little-Known Tips for Improving the Performance of Your Campaigns [Infographic]()

  • Pingback: 5 Little-Known Tips for Improving the Performance of Your Campaigns [Infographic] – Lucid Biz Intelligence()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC | Digital Marketing Tips()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC -

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC | Digital Reading List()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC - Real Blackhat Methods()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC – Writingcastle()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC | Internet Marketing Ideas for Businesses()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC | Magnetic Affiliate: Tips, Tools, & Training()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC | Social Media Marketing News()

  • Pingback: 6 Ways to Target Early-Stage Buyers in PPC - LEARN MUCH BETTER()

  • fritalci

    This is both one of the most spot-on informative posts and one of the funniest posts I’ve ever read. I LOVE LOVE LOVE SKAGS. I cannot overstate how much they have improved my accounts.

  • Btype

    Hi Jonathan, you are saying that you should create separate SKAGs for plural/single key words unless the volume is low.
    What is considered by low?

    For example in my situation google keyword planner shows these numbers:
    san francisco tours – 1400, san francisco tour – 500
    san francisco bike tours – 1200, san francisco bike tour – 400
    san francisco walking tours – 120, san francisco walking tour – 330
    san francisco segway tours – 200, san francisco segway tour – 50

    Should i create SKAGs for all of them?

    • Hey Btype 🙂

      If you read the post, you don’t have to do any research.
      Your SKAG of ‘san francisco tours’ would trigger ads for bike, walking, and segway tours as well.

      So all you need to do is follow your search term report, sort by descending impressions, and if you think there’s enough search volume, then you can create a SKAG from your own data right there.

      • Btype

        Johnathan hi, thank you for your reply.

        i did read the full post and found it very helpful! Seems like my question was not clear 😉

        What i am asking is if i should create separate SKAG for prular and single versions.
        Example: one SKAG for “+san +francisco +walking +tours” another SKAG for “+san +francisco +walking +tour” and add negative keywords so they don’t overlay.

        Or should i create just one SKAG for that pair “+san +francisco +walking +tours” and that’s it.

        • Just pick one (plural or singular), and the search term will tell you if there’s search volume for the other.
          Regardless of plural or singular, they’ll trigger ads for each other.

          • Mitch

            So just to be clear on this one. If I’ve already got the report and the singular exercise has 21k impressions and exercises 11k then I would create a SKAG for each?

  • Gabe Lee

    I would love to get a video tutorial walkthrough of how you manage the excel bulksheets. Also, how do you do the bidding side of things? “Enhanced CPC” or manual bids?

    • It says so in the article 🙂 All same bids, then adjust after data comes in.
      What it didn’t say is what type of bidding strategy, so use manual bids 🙂

  • Hi Jonathan,

    Great post, as always!

    There is a big downside of having such a granular approach with keywords. It happened to me many times that I add a specific keyword based on the search query report, but it gets the “Low Search Volume” status. The bad thing is that Google temporarily make these keywords inactive so that they don’t trigger my ads anymore (because the query was also added as a negative keyword in the more generic ad group).

    See how Google manage “Low Search Volume” keywords :

    How do you deal with that?


    • If the search term report shows those and you create SKAGs from them, then you may have hit a point of diminishing returns.

      • So it would be better not to create SKAGs from them?

  • Torben Ni

    Awesome, I love this approach. I never even started a PPC-Campaign but that strategy seems just great.

    My mom owns a local gastronomy in a smaller village in germany, where she organizes birthday parties, weddings and other sorts of events. I thought about helping her getting new clients and thought about PPC. Therefore thanks for your helpful content.

    • Awesome, can you have her send a cake our way after she gets her first sale from PPC? 😉

      • Torben Ni

        Of course! I will give you a line when she gets her first client 🙂 I’m sure my mom will love that!

        • I’m so frikin’ excited haha! I’m gonna check in on you in a few 😉

  • I’ve subscribed to this philosophy since your original post, love it! Actual implementation to a t is a different story!

    So you:
    1. Start with seed keyword “male speedos”
    2. SKAG out all the variations that have impressions and clicks
    3. Add negatives of new SKAG groups back to original group
    4. But then do you also SKAG out that original seed keyword “male speedos” into it’s own group? If so, then the original group would just consist of the cruft and left to keep running to pull in new terms? Or would the original group be +male +speedos with a negative exact match of [male speedos] ?

    Secondly, how do you control for costs upfront? Do you just limit the budget and know you’ll be wasting some money for a bit before SKAGs are built?

    • 1. Yes, but it’s a seed keyword that you turn into a SKAG
      2. No, if “male speedos” is a seed keyword, then you don’t create variations of that, you allow the search term report to do that
      3. No, you do this from the search term report only
      4. No, because “male speedos” is your original SKAG, so you keep all 3 match types in it

      You control the budget and bids however you’d like 🙂

      • Boris_Mayer

        … a question that came into my mind: do you also create new SKAGs if instead of “male speedos” ppl are looking for “speedos male”?

        • Your search term report would tell you if they are, and that’s what you need to follow.
          If your keyword is +male +speedos, then it would trigger your ad to show if the search terms are:

          male speedo
          male speedos
          speedos male
          speedo males

  • Zakaria Desai

    Awesome update to an epic post Jon!

    I’ve been using SKAG’s for about a year now. And let me tell everyone out there – you’re crazy to not use it! Quality scores are up, CTR and most of all – relevant clicks!

    Anyhoo – just wanna know more about the negative keywords between ad groups. So you’re saying if you have a search term that you specifically want to be allocated to an ad group – then add this term as a negative in all the other ad groups?

    Thanks, Zak

    • Haha awesome, Zakaria! 😀

      If a search term doesn’t match the keyword 100%, then you need to add that search term as an ad group level negative by checking off the box next to it and then create a new SKAG for that search term.

      This is why the 3.0 update will be great, because it’s ridiculously simple, but I guess I suck at explaining it haha!

  • Shuai Yang

    I totally agree with this post and subscribed to the blog!
    Is there any tips for ad testing? or any article like this about ads testing you wrote? As SKAG will create some inconvenience in ads testing right?

    • You test ads like you normally would in any ad group 🙂
      Nothing fancy there.

  • Pingback: 19 Reasons Why Single Keyword Ad Groups (SKAGs)...()

  • Felipe Tinoco

    Great post and explanation mate.
    Actually I have a lot of campaigns clusterized by demography and objective and Im willing to migrate to SKADs, but i have 3 questions:

    1- I have 2 types of groups, 1 directing to my site (checkout) and 1 directing to a landing page to capture leads. Is there a problem if I use the same keywords for both groups but directing to a different link? Am I going to compete to myself this way?

    2- I have different links to specific locations (e.g and, if i use the same keywords and ads, but changing the link , im going to compete? Whats the best way to structure my campaigns?

  • Hi
    Skag is somethung new that i read here, so than you for writing such a great articles,
    I have few points which you like to note
    1- creating SKAG with editor is more easier and faster if we use csv import and export function. With excel we can create complete adgroup with keywords and can import with few clicks and same for ads in each adgroup.

    2-(curious) about internal competiton, i am agree with you but have also a doubt. Adwords have some preference to choose keywords for a search query in below order:
    Match type

    So I think internal competition will be more if we add duplicate keywords (may be by mistake) but for slight variations, we should not worry about this..
    I would like to know your feedback on my points.


    • Hey Anil 🙂

      1) Agreed
      2) The only internal competition you need to worry about is short tail keywords stealing away impressions from longer tail keywords.

  • Ateendra Sharma

    Hey Johnathan,

    Amazing article! I have been using this for some time and I see good results pouring in. I have read elsewhere that broad match modified should have a 10-15% lower bid than phrase match which itself should have a 10-15% lower bid than exact match. Would you recommend following this when deploying SKAGs or is it not needed since AdWords always goes by relevancy first?

    P.S. Would love to see this answered/covered in the 3.0 update. It might help quite a few PPC managers who have this confusion.

    • Hey Ateendra 🙂

      I’m not sure of the logic as to why you’d lower those bids by a percentage? That would mean that you MUST know that those other match types make you less money or have lower conversion intent? I find that hard to believe and probably just a random idea with no proof.

      I always recommend to keep all your bids the same, and then adjust them once you see the data. It’s insanely simple and nothing that needs to made complex 🙂

      • Ateendra Sharma

        Phew, such a relief to know that! Thanks for the reply, Johnathan! 🙂

  • jbogs101

    So as you build out your SKAGs, are you also building out different landing pages for each SKAG? Or are you using 1 landing page with individual copy related to each specific SKAG somewhere on the page (to increase QS and advert to landing page match)?
    Increased CTR is great but if the landing pages aren’t converting your sunk. Do you have a suggestion on tracking landing pages or is it simply looking at CTR vs actual sale? Another words if you find high CTR but poor conversion it’s time to test a different landing page?

    • Not sure why we’re just seeing this comment now – but the answer is no 🙂 Because going after QS improvements on an LP can be a deathblow to conversion rates.

      That’s what you share care about when it comes to LPs.

  • Múčka

    Incredible! Thank you so much for this extremely useful article.

  • vlad

    Hi there :
    Thanks a million for this … I’ve decided to jump right now to adwords just after reading this post 🙂
    Nevertheless as I’m using adwords editor, I got 2 questions :
    – do we have to “rotate ads evenly” or let that option as “optimize for clicks (or conversions)” ?
    – do we have to include or exclude “search partners” ?
    Many thanks !!!

    • You should rotate them evenly and then the search partners depends on the performance.
      I would recommend targeting them to begin with.

  • Yes, you always want to have the keywords in the ads too 🙂
    Otherwise you won’t get that CTR lift.

    • Steve Woltmann

      I do have keywords in the ad. Once in the headline then again in the text below.

  • Thanks for the excellent write up – I’m going to start testing this on some of the accounts I manage. Quick question: I noticed you didn’t mention anything about campaign negatives. Are they superfluous here, or is it worth adding them if there are words that you for sure don’t want triggering keywords. e.g. in your “electric bicycle electric bicycle” example would it be worth adding “electric bicycle electric bicycle” as a campaign negative if it had enough search volume in the search queries report?

    • Hey Benji, and thanks 🙂

      You’re right, but I’d also say that’s obvious. You of course would want to have regular negative keywords like:

      In your example, it looks like a bot search, but if it has conversions, then you wouldn’t exclude.

      • Thanks, Jonathan!

        Regarding ad group level negative keywords, is it really enough to only add the exact match versions of the negatives? Or do you have to add the phrase match version of the negative as well to reduce cross competition?

        Also, is it safe to say that this strategy is best for accounts with bigger budgets and therefore, less data dilution? How can smaller budget accounts benefit from this strategy outside of waiting longer for data to accumulate?

        • henry

          Hi benji, i too have this issue. Tons of internal competition. This article would really benefit from johnathan covering how to handle this. In my case “blue swim shorts” triggers another ad group bmm +swim +shorts +blue. Johnathan mentioned i should add a phrase match negative also. But i have tons of this going on, so i need to know if its better to add exact & phrase negative at the same time off the bat from the search term report. Instead of waiting for this mismatch to happen.

          It seems adding the negative is not enough to perform this process. Which in my situation is the only downside to this awesome article. To truly make this article the go to cornerstone conent this needs addressing.

          I also have issues with CTR. The bmm keyword always has poor ctr. Does this impact on account QS? Should we lower bid or something?

          Im sure all people here will see these issues. It would be good to see update 3.0 cover this above.


        • No, you just check off the box next to the search term that doesn’t match your keyword, and then exclude it as the default exact match negative. To see this, you’ll want to have your keyword column visible in the search term report.

          We use this strategy for tiny accounts too, since we know that after SKAGs, the next performance lift won’t come from ad testing, but CRO initiatives 🙂

      • henry

        So we make these bot searches exact match negatives? They have 0 conversions.

  • jack tan

    jon i really appreciate this youve done something great for the adwords community. Ive a few questions if u can help pls as im stuck a lil.

    q1) you mentioned as a sample root kws “online seminar” & “online classes”. And i understand we should create new SKAG’s when we find new search terms, and thereafter add negatives to counter each other. But off the bat, what if someone searched “online seminar classes”. Seems that both root kws would be triggered. So just based on this example, am i right to say that we should be doing counter negatives even for our root kws, even before we analyse our search term report? Im asking cos i find alot of my root kws have cross triggers so am thinking if i should create negatives off the bat?

    q2) my keyword is too long for path 1 in adwords editor. Can i split it into path 2? for eg kw “handsome elephants”: is that fine?

    q3) Is it ok if different ad groups have similar ad [descriptions] and [headline 2’s]? Mainly only kw i have made different in [headline 1] and [url path]

    q4) how smart is google’s plural check for broad match. For eg: i did all 3 match types for “renovation companies”, so i can leave out “renovation company”?

    txx very much Jon

    • Hey Jack 🙂

      1) Correct, but you wouldn’t add them off the bat. You’d just wait until the search term report shows you and then add the ad group level negative there

      2) Yes 🙂

      3) Yes, you need ad variety to split test

      4) It allows plural/singular search terms to come through. So all you need to do is follow the search term report.

      • jack tan

        Thanks V much man. Appreciate the quick help! Keep rocking Jon.

  • henry

    Hi Johnathan, I have read the article many times, and implemented SKAGs. But i face an issue where i have 2 skags that contain the same keywords, in a different order.


    root keyword: “blue swim shorts”

    Search Report shows the following have high impressions:

    “Mens Blue Swim Shorts”
    “Blue Mens Swim Shorts”

    I tick both as an adgroup negative, and then create the two new skags.

    Ad Preview & Diagnosis shows a problem of mismatch – when i search for “Mens Blue Swim Shorts”, it sometimes pulls from the +Blue +Mens +Swim +Shorts. This is a different ad group to the Mens Blue Swim Shorts.

    Following your steps, the negative kw is entered only into the root keyword ad group (blue swim shorts) making the above issue prevalent.

    I guess the above can be solved by entering a negative kw into these ad groups like below:

    – Mens Blue Swim Shorts AD GROUP – enter negative of Blue Mens Swim Shorts
    – Blue Mens Swim Shorts AD GROUP – enter negative of Mens Blue Swim Shorts

    Q1 – Is this correct solution?
    Q2 – assuming it is, what match type do i make the negative. In your steps, you are adding exact match negative only, because you click the tick box from the search term report.

    Please can you clarify the above. I look forward to hearing back from you.


    • That’s correct, Henry 🙂 You just want to use at least a phrase match negative keyword in this situation, since a traditional broad match (like you normally would) would prevent both from showing.

      You can even add broad match negatives, but you might as well follow the exact match negative that’s default when going through your search term report 🙂

      • henry

        Hi Jonathan, I always add the exact match negative from the search term report as per your steps.

        So to clarify you are saying when 2 skags contain exactly the “same keywords in different order”, we add BOTH exact & phrase match negatives. Exact match negative from the search term report & manually going back and adding in a phrase match negative.

        If you can confirm.


        • You would only really need to add the phrase match negative.

          • henry

            Sorry to drill it home but its pivotal to executing this accurately.

            Is there no need for exact match negatives then in your skag methodology?

            In your article it shows you adding the negatives from the search terms report by clicking the box. By default are these being added as exact match or phrase match? In my adeords they auto create as exact match negative if i add them from the search term report. Does that mean you yourself only add the exact match negative?

            In 3.0 can you cover the negative match types you add so we can all see.


          • They’re being default added as exact match, but you could also do broad match as you’d add most negatives to your root keyword SKAG, so you’re only preventing longer tail versions of that keyword from appearing as search terms.

            In your example, which isn’t ordinary, you can do all match types except broad match. Doesn’t matter.

          • henry

            Awesome thank you.

          • Evan

            Hi John, i have 3 keywords on my Group Ad called “Pabrik Tas”
            1. +pabrik +tas = Modifier broad match
            2. [pabrik tas] = Exact match
            3. “pabrik tas” = Phrase match

            three of it eligble but i got warning “Another creative in the ad group was selected over this one.”
            please tell me ho to fix it..

            (see attachment below)
            thank you

          • Mark Lilley

            Hi Jonathan
            Super useful article, only question I have at the minute is the same as Evan – do you have any advice re this?
            Many Thanks

          • Most likely internal competition between ad groups that can be resolved with ad group level negative keywords.

          • Mark Lilley

            Thanks Jonathan, I’ve been looking at it and consistently in all of my adgroups 1 of my keywords (each adgroup has exact, phrase and BMM keyword) has the status as above i.e. “Another creative in the ad group was selected over this one.”. Checking all keywords with the status in the Ad Preview tool they are all showing ads? Match type getting the status is also different across adgroups, although mostly it is the phrase match keyword.

          • Jake Freedman

            I have added the corresponding negative keywords to each ad group from my other ad groups, and I am still getting this error. Any way to fix?

          • I’d have to see it to know what’s going on. Have you tried calling Google support?

          • Jake Freedman

            Not yet. Now it appears there was some delay after adding the negative krywords and that about half are showing “yes”, and the others are still showing the same message. I’ll try calling

          • Jake Freedman

            For example: Lets say my keyword is “doctor”. ads are showing for BMM and exact match, but not phrase match. I have negative keywords added from my other SKAGs with keyword such as “walk in doctor”.

  • John Smith

    Hi Jon, thanks for the great strategy!
    Would you be kind to tell a bit detailed about ad copies strategy?
    Ok, I start with minimum 2, but is there some maximum? Some of my ad groups have 25 ad copies.
    When I do new SKAG, do you recommend to remove ad copies with low share (of course with adding new copies to keep at least 2)?
    Is it worth to include {KeyWord} into titles into some of the ad copies?
    Thanks in advance!

  • henry

    Hi Johnathan, SKAGS have been running for a month now. I am seeing a lot of bot searches with massive impressions & 0-1 clicks. I have tried to make these ad group negatives, but they are very lomg in length & i get an error out keyword length. Can you let me know how you handle this as i can see this hurting performance badly. Thanks

    • Nothing you can do about it unfortunately, other than adjust the keyword itself.
      Those are bot searches and usually run over 10 words in length. 10 is the max word count to use for a negative keyword.

      • henry

        Ok but

        – Do i just reduce the bmm or phrase?

        – lets say the bot search is 12 words in length, can i emter the first ten words as a phrase negative or something, would that stop it showing up.

        At this point i am glad to say your method seems to work nicely.

        The only other painful issue i get is on long kws not fitting headline or url –> non perfect QS 😉

  • Sharnid Ahmed

    Hi John,

    Great article. I just needed a bit of clarification on some aspects.

    Let’s say I take out all the top 5 keywords from each of my ad groups, and create SKAGs for each, how long should I wait to evaluate performance? And after that I see the SKAGs are performing so much better, do you recommend just pausing the other keywords in the account, well I guess I could just pause those original ad groups altogether right? Also after evaluating match type performance of the root keyword, do you just select the most converting match type and go with that and pause the other 2?
    Would you recommend creating SKAGs for lower performing keywords (worse than top 5), or should I just stop bidding on them? Also in order to find out which top 5 keywords to build the SKAGs on, do you look at all time data, or a specific range?

    I am a bit confused by some of the discussion about which keywords to negative, in terms of shorter tail keywords taking impressions from longer tail. I was wondering if you could explain this process a bit more.

    Thank you!

  • bari

    Hi Johnathan, thanks a lot for the rather detailed post. It seems hard to come by detailed & practical posts these days. It is like every other blogger went on a crazy publishing marathon and not really caring about providing actual useful content.

    Iv have two questions one of which you have been asked already but I would appreciate it if you can provide more insight on:

    1- Why usemore than one match type ?? why use [blue hats] if +blue +hats is going to capture the query and show your ad anyways ? You have mentioned that different match types perform differently. Well I think logically they shouldn’t. But lets say they did, lets say I ended up with [blue hats] with a QS of 8/10 and +blue +hats with QS of 9/10, how should this effect my next step ? should I “remove” the less performing match type or what ?? In other words, what action should I take after getting the datA and performance difference between the two match types ?

    2- I have created a new fresh account for a new fresh website, I didn’t even start running the campaign and I got these two crazy cases:

    A- keyword has 1/10 QS and google tells me KW not showing because of low ad rank. The catch ? the KW has 0 impressions, 0 clicks, 0 whatsoever, I haven’t even started running the campaign. I also tried bidding 20USD on those keywords just to check which changed nothing.

    B- keyword doesn’t have QS “because obviously I didn’t start the campaign yet and google says there aren’t enough data to calculate QS score”. Yet again, the ad isn’t showing because of low ad rank ???? again, the keywords have 0 impressions, 0 clicks, 0 whatsoever and I tried bidding up to 20USD on it just to check. and the issue persists.

    It’s like google gone crazy, how can google determine the AD rank if it can’t even determine the QS because of not enough data because the keyword never actually ran ?

    I would really appreciate it if you can answer me on those two questions, thanks a lot.

    • Thanks 🙂

      1) Because they all perform with a different CPA. So if you don’t target all 3 match types, then you can only adjust one bid. Has nothing to do with capturing or impression volumes.

      2A) Have you tried calling Google? There might not be enough search volume for that keyword.
      2B) Same thing.

      • bari

        Thanks for getting back to me. Much appreciated

  • Jordan

    Hey Johnathan!

    Thanks for this guide, it’s been so SO helpful!

    I’m curious as to your thoughts. For industries with higher than normal CPCs ($15+) would it be more ideal to start new campaigns with keywords that you want to go after rather than having modified broad match seed keywords find search terms for you? It seems like it might have more potential for wasted spend no? Or would it be ideal to do both; have your seed keyword find search terms and start with ones you already want to go after?

    The reason being I’m about to run a new campaign for a large client and I’ve done keyword research to find search terms already. Thus, I’ve already set up 300+ SKAG in a spreadsheet to simply enter and start running with. Any help is much appreciated! Thanks in advance! — Jordan

  • Jacob

    Super article. Looking forward to the updated 3.0 version with video 🙂

    A thing which is not clear for me.

    For the campaign structure do you:

    1) Have 1 campaign with all you SKAGs within that campaign?
    2) Do you create 1 campaign with 1 SKAG in it. And then create a new campaign with another SKAG in it. And then create another new campaign with 1 SKAG in it. And then…



  • Pingback: Crush Your Adwords Campaigns With This Small Tweak()

  • henry

    Hi, what is an acceptable QS? I am experiencing

    80% 10/10
    10% 9/10
    5% 8/10
    5% 7/10

    The longer words seem to cause the lower QS as i dont have enough length in my ads for all words.

    For ecommerce would you think this could be improved with dynamic insertion of KW onto the landing page?


  • In your example you added one ad group for “e bike” and another one for “e bikes”. Will the difference between those two ads be so important to make it a good idea to separate those keywords into different ad groups?

    • Most singular/plural keywords perform differently, so the answer is yes 🙂
      But again, follow the search term report and don’t pre-create them if there’s no volume.

      • Understood, cool. About what you mention, do you usually start with seed keywords only and then add long tails continually using the search term report, or do you always add long tails from day one when you´re sure they have search volume?

  • David B

    Great article Johnathan,
    After starting to use SKAG I’ve noticed a difference in just a short time. I was wondering though, in the search terms I get terms for example like “bike shops that are near me” or “bikes shops that are in (city name)” Would those be added to the list of SKAGs as well? Looked in the comments and could not find someone asking this question.

    • They would if they have enough search volume from your point of view 🙂

  • Shuffles03


    Firstly, thanks a million for creating this SKAGS post. We recently launched an eCommerce store and I’ve been following your guidelines.
    However, I’m thinking of completely starting my AdWords again. I originally started by using one search term, created my AdGroup with the various matches including two ads for the group and then grew my groups each day based on the search term report of that AdGroup using negative words along the way. Basically, the search term report is the one that is growing the number of AdGroups for me.

    The brand we sell ‘Teng Tools’ is really specific so creating variations like you have in your blog post example;

    1. web college
    2. web university
    3. web degree

    etc. is not really applicable for us.

    What I’m now thinking of doing is this:

    Use the Keyword Planner Tool, search for new keywords using the phrase option, type in ‘Teng Tools’, target Ireland, english etc. Then under keyword options I’ve selected ‘only show ideas closely related to my search terms’.

    When this page loads up it shows me 98 terms related to ‘Teng Tools’. As it’s a niche brand, ‘Teng Tools’ itself gets an average of 880 hits per month whereas the other 97 results are around 10-20 hits a month.

    As a starting point I’m going to create 98 separate AdGroups (or thereabouts – some of the results are pointless) so out the gate I’ll have over 90+ AdGroups with 2 ads per group.

    Each day I’ll check and mark negative keywords using the search term report and create additional AdGroups if warranted.
    How does this sound?

    Do I need to apply negative keywords to each of these 90+ AdGroups in the beginning?
    Here is an example of a few of the keywords/phrases associated with the ‘Teng Tools’ search:

    1. teng tools online
    2. teng tools offers
    3. teng tools 1 2 socket set

    Do I need to make the word ‘online’ negative in the ‘teng tools offers’ and ‘teng tools 1 2 socket set’ AdGroups and vice versa or will I leave them all alone as they are in their own separate AdGroups and only create negative words when using the search term report?

    Basically, I’m starting out the gate with loads of AdGroups as opposed to waiting for the daily search term report to dsuggest based on teh previous 24 hours of user searches.

    • We wouldn’t it make sense to be specific? If your search term report is sorted by descending impressions and they’re less than 10 words, then it’s possible.

      No you don’t need to add negatives in the beginning. You add negatives from your search term report and just follow that.
      You also don’t need to do it daily. Just do it once a week to begin with.

      Don’t over-complicate it, just follow the search term report once a week for adding ad group level negatives where the search terms don’t match the keyword (you need to add in the keyword column) and then create new SKAGs from those ad group level negatives 🙂

      • Brian Gildea

        Do you have a threshold for which ones you choose to make a skag out of? Obviously you’re not going to do it for every search term.

        Also, how do you remember if the keyword might be competeing with another? Do you flip back and forth between your account structure to see what skags might be competing?

        • Correct, you sort the search terms in descending order of impressions for example 🙂

          For your second question, you just use the keyword column in your search term report and if there’s a mismatch between that and the search term, then you know they’re competing.

    • Michael Tierney

      Great article, thank you, cleared some things up for me – I have a question! – I guess it boils down to this – should we vary our bids for each match type or keep them the same?

      There’s questions behind this but maybe I’ll just leave it there for clarity, thanks in advance

      • What would that accomplish? You don’t know how they’ll perform prior to launch.
        So keep them the same until you get data that you can adjust bids with.

  • Ridwan Ibrahim

    If the objective of SKAGs is to match the Ad copy to search terms, wouldn’t it be far easier to use Dynamic Keyword Insertions?

    Admittedly I’ve never used Dynamic Keywords so I’m sure I’m entirely wrong, but it’d be great if you could help me understand why.

    • If it was called Dynamic Search Term Insertion, then yes 😉 But it’s not. And that’s why you can’t use DKI in this situation.

      • Anton Goncharuk

        Hi @johnathandane:disqus ,
        1. But with SKAG you are also using keyword not search query, what is the difference then?
        2. If you do your keyword research properly and accurate enough then you basically can match your keywords with search query pretty accurate (maybe in 8-9 out of 10 times). So in this situation DKI would work the same way as SKAG does, doesn’t it? Maybe only character limit would be problem in some cases for longer keyword/queries. Is there any other advantages of using SKAG over DKI? I just don’t understand why should I invest my implementation and optimization time using SKAG if the DKI would perform pretty much the same way but with much less afford?
        Thanks so much for your time and knowledge.

        • 1) Google doesn’t dynamically insert the search term into the ad. It only does the keyword – they’re not the same.
          2) It has nothing to do with research. Even exact match keywords have more search terms to the one keyword. It’s all about lowering your ratio as close as 1:1 with search terms to keywords. That’s it 🙂

          Do DKI, and you’ll be left behind the competitors that use SKAGs.

          • Anton Goncharuk

            @johnathandane:disqus yes, DKI does not insert user’s search term into the ad copy, but SKAG doesn’t either, it is your ad text being shown to user, not his search query.

            Let me give you an example to make my point be clear:
            Search term: rental cabin in Destin Florida

            SKAG ad: Rental Cabin in Destin Florida
            DKI ad: Rental Cabin in Destin Florida (as I have this keyword in my account, this keyword triggered search term and been inserted into the ad).

            As you see at the end of the day result is the same, so what is the point of using SKAG over DKI?

          • Correct, never said that SKAG does DKI 🙂

            If you do the DKI set up, then are you not going to refine your search term report by creating new SKAGs? If not, then DKI won’t solve your core problem – if you do, then it’ll be the same as SKAGs, with the exception of the keyword in the path not being able to work as neatly with DKI as it does with the SKAGs.

          • George Hedley

            Hey Johnathan – fantastic article.

            I am slightly confused with this question posed by Anton and whether your theory can apply to Keyword Insertion (not Dynamic).

            So. What about using standard Keyword Insertion i.e. {KeyWord: Default Text} as Headline 1. As the Keyword inserted will be the Search Term that triggers that ad and the Default Text would also be the same Search Term (ideally). Would this not fair better in Google’s eyes that having just the Search Term as Headline 1 (which would happen to be the keyword)? Ignoring DKI in this instance.

            Would love to hear your thoughts on whether that’s a waste of time or something to further enhance and would include a bold headline 1.


          • George Hedley

            And as an extra note. What should an advertiser do if their Long Tail Keywords are too long for Headline 1 and therefore the search term won’t be properly activated?

  • Claus Hillerup Emborg

    Following the SKAG philosophy:
    Should you make an adgroup for compound words as a root… Example “Drycleaners” in addition to “Dry Cleaners”…. This might not be relevant in English – but in my Language Danish – many search words can be / and are / written in both ways…

    • Kun hvis din Search Term report viser at folk søger på begge måder.

      Did you know I was Danish? Born and raised in Hillerød 😉

  • Kyle

    Great article! My question is how do you know when a search term is from spam or not? I have a large majority of search terms with the & sign in it and I don’t think that many normal searches would include a special character

    EDIT: I think I figured it out.This is how it is showing broad match types and the previous campaigns built before only had broad matches. I think I’ll have to start with some basic SKAGs and then I’ll start seeing phrase matches

    • You can usually tell is the search term has a lot of words or the same word is used multiple times in the one search term. It’s pretty obvious that a human doesn’t type like that, especially with a lot of impressions too.

  • Larry

    How do you best control the campaigns for larger accounts with different product lines. Do you let a campaign sit with hundreds of adgroups or is there a best practice on how to break it down? For instance seed words in one campaign and first level variations from the search report in another? Looking for a little guidance here. Thanks

    • I’d recommend starting with the campaign with the most traffic and then prioritize the search terms in descending order of impressions and keep the new SKAGs in the same campaign.

      Or you can group your campaigns differently now that you have more ad groups 🙂

  • henry

    Hi Johnathan, i have a 700+ skags now set up since starting out. When I started, my original root KWs were 3 words in length as i was a little scared of going any broader.

    I wanted to know how if i should try and introduce a shorter 2 word keyword to expand out my campaign reach.

    Lets say my industry is swim shorts.

    Imagine starting out with “mens swim shorts”, and other 3 word KWs

    How would i now introduce “swim shorts” – would you add a adgroup negative for all 700 of my skags. Would you just go exact match. Or follow the bmm,exact,phrase method seen in all of my other skags.

    Im concerned this will screw my CTR and QS. Have you ever been in this position.

    This 2 word keyword is going to throw up a lot of irrelevant traffic, like “batman swim shorts” or other items that i just don’t stock.

    Would appreciate your input.


    • Since these 2 word SKAGs are experimental for you, I’d recommend creating a brand new campaign with a small test budget. Then, take the names of the 700 SKAGs and download them in a CSV.

      Then take those 700 SKAG names and add them to a negative keyword list and apply that list to your new 2 word experimental SKAG campaign.

      That way your 2 word keywords are not cannibalizing your budget from the other 700 3 worded SKAGs 🙂

      • henry


        If we can put this to one side for a moment, i have another question. This time to do with internal competition from skags in the same campaign. I am trying to stop shorter tail ad groups stealing impression share. You mention above: ” keep a closer eye to Search terms that don’t match the keywords and to use ad group level negative keywords to make sure your shorter tail SKAGs aren’t stealing away impressions, like the cats were stealing away bread earlier.”

        I follow your method of checking over search report, adding negatives where i spot cannibalization.

        Is this the best method – it’s quick, but it’s reactive. I am only placing these ad group level negatives as they show up on the report.

        If i have 800 skags, i could start at skag 1 and enter the 799 other skags as a negative. I guess a macro or script could do this pretty easily.


        • You could, but you’d fry your brain.
          And internal competition isn’t as bad as you might think, your search term and ad are still pretty relevant to each other.

          Plus, the reason you follow the search term report is that it tells you which new SKAGs to build in addition to which ad group level negative keywords to add too.

  • Nick Simpson

    Hi Jonathan, I’m still trying to figure out the best way forward with SKAG. We sell 1000s of products and want an advert for each one. So for instance, the “HP 123 Printer”.

    The way I see it there are two ways to set up SKAG:

    1. Have 1000s of campaigns so one for each product. Then within each campaign have a handful of AdGroups for the various keyword combinations.

    2. Have a more manageable list of campaigns, such as “HP Colour Laser”, “HP Mono Laser”, “Canon Colour Laser”, etc. Then within each campaign have every keyword in a huge list. So for instance within HP you might have “HP 123”, “123 Printer”, “HP 456”, “456 Printer”.

    Either option is pretty unmanageable so I’m not sure whether SKAG works for campaigns that have 1000’s of products to list.

    What do you suggest? There are no real focus products because of the nature of our business.



    • I’d start with root keywords and your search terms dictating what you build and then go from there.
      If the results follow, then you know what to prioritize.

      • Nick Simpson

        Cheers for the quick reply! Purely for administering would you recommend setting up using method 1 or 2? Both seem awkward but which (if any) would be better?

        • 2 🙂

          • Nick Simpson

            That’s what I was thinking but worried there will be an unmanageable list of adgroups. I guess creating even more campaign categories will help. Maybe, “HP A4 Colour sub £500, etc”. If there are say 10 products in a campaign and each have 10 different keywords then 100 adgroups could be managed. You agree?

  • Nick Simpson

    Hi again, okay I think I’ve nearly formulated a plan. Just one question…

    If I have an AdGroup with Exact, Phrase and Modified Broad versions of the same keyword then do I need to put in negative keywords to stop say the phrase or modified broad version being triggered if the search term is actually exactly matched to the keyword? I’m hoping the answer is no because Google checks to see if it’s exact first, then checks whether there’s a phrase match and finally checks if there’s a modified broad match.

    Cheers, Nick

    • No, because that won’t matter – and it usually doesn’t happen when you have the keywords in the same ad group 🙂

  • Lucas

    Why didnt you mention how to build ETAs for this strategy and focused only on STAs that had days counted. Adding the kwds on the path is a little bit harder now. You should update the post.

    • It’s pretty much the same thing, now you might have to split the keyword up before and after the slashes in the path.

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study]()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – WeRank Virginia Beach SEO Agency()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Jodi Hood()

  • Pingback: How „Message Match“ Can Lift Conversion Rates by 212.74% [Case Study] | LocalHome Pros()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Anthony hanley()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Professional Acting Studio()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Site Title()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – meganmatthewsblog()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Roxanne Greer Blog()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Raymond Castleberry Blog()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | bigideamastermindmentoring()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study]()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - Content Hydra()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | Digital Reading List()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Search Engine Optimisation Company()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | brianlichtig()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Amber Perz's Blog()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] - Scott.Services Online Marketing NewsScott.Services Online Marketing News()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] – BLOG NEW WEB NETWORK()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | Xero Media Services()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – London SEO Services()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | SEO Company Surrey()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study]()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] -

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Jake Bennett's Small Business Tips()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | PAGE ONE()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] - סוריקטה()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | SEO Unity()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] -()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] - Evangelist News()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] – IM review, bonus()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | IM Local SEO()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Gana Dinero Colaborando | Wecon Project()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Kevin Marcotte Negocios Mentoring y Coaching()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Russell Beaudoin Blog de Negocios()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Título del sitio()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] | Severo Mondragon Blog()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] - Austin Local Search()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | My Blog()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | Ad Guide()

  • Pingback: How “Message Match” Can Lift Conversion Rates by 212.74% [Case Study] – Ok Huge()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] - Biz Ops()

  • Pingback: PositionONE How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - PositionONE()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - Odysseus Internet Marketing()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | Latest News and RSS Feeds()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] – Extralearn()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - Forestgram Academy()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | SEO SENSEI()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] -

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | AZ-emarketing()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] |()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - ZRaven Consulting()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] – CRI Emails()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] | Search Burst – All The Search News()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | Parere()

  • Brian Hargrove

    I’m confusing myself – please clarify. On the example about “Dry Cleaner”, it was asked if he needed separate SKAGS for the singular & the plural. It said no, but then said search term report would show. SO, if there is volume of searches for “dry cleaner” and “dry cleaners”, would you set up separate SKAG. And then step further, if search volume showed, wold you set up SKAG for the singular and plural with “Miami”?

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] – Website Design +357 97682182()

  • Pingback: How "Message Match" Can Lift Conversion Rates by 212.74% [Case Study] | SILO Test Site 2()

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] - BFG Consulting()

  • ‫סמי סבין (‪sami sabin‬‏)‬‎

    Hey Johnathan!
    Amazing article thanx so much for it! Cheers from Israel 🙂
    Will be great to hear answer for:
    1 – After you start the campaign you keep runing with all kinds of phrases for ever? I mean – will you pause “Exact” and let the others running?

    Thanx a million!

  • Pingback: How &quot;Message Match&quot; Can Lift Conversion Rates by 212.74% [Case Study] – Are you giving your best?()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement ⋆ Easy blog networks()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement | PAGE ONE()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement | Page 1()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement – VIKAS MISHRA()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement | SEO SENSEI()

  • Pingback: How to Personalize Your Ads and Landing Pages Using Dynamic Text Replacement | Brett Scofield()

  • JM116

    Hi Johnathan,

    I’ve set up SKAGS in my account but I’m being told that having the 3 root keyword match types in my ad group isn’t best practice. Will this set up harm my account in ant way?

    Thanks 🙂

    • Who said that lol?
      That makes no logical sense, so the answer is no.

  • thanks for the great write up. i’ll have to read it a few more times to let it sink in. love the humor in writing as well especially the oprah part….she loves more than bread though….as in anything edible.

  • Melanie vitto

    Jonathan, how often do you recommend the search term report is used to create new SKAGS? i.e. do it monthly to get more insight on how popular the new long tail keywords are, in terms of impression/ctr? I wouldn’t want to have low ctr/impression keywords just lingering. Thoughts?

    • Weekly to monthly depending on how much traffic you have.
      The more you have the more frequent you do it.

  • Alan

    Hi, this is a great article. Could you please explain one detail as I’m not sure I’m getting it correctly?

    Suppose my seed keyword is “electric bike” and from the search term report I get a new relevant keyword, say “electric bike for sale”. I then create a new SKAG, i.e. “electric bike for sale”. Do I use “electric bike” as the negative keyword in this new SKAG?

    • Correct 🙂

      • Jack

        wait, should Alan not be using “electric bike for sale” as a negative keyword in the original SKAG which has the seed keyword? Adding “electric bike” as a negative keyword to the new SKAG would prohibit “electric bike for sale” from triggering the new ad. Or am I completely confused? probably

        you show in the GIF that once you pick out the search terms that you are going to create new SKAGs for you then add those search terms as negative keywords at the ad group level. Is that not adding them to the original root keyword ad group negative keyword list?

        Also, when you add the SKAG-worthy search terms to the original campaign’s negative list, do you add them as negatives in all three forms (borad match mod, phrase and exact)?

        • Kenny

          I’m on the same page as Jack. I would add “electric bike for sale” as a negative keyword in the root SKAG “electric bike”. This will prevent your short tail keyword “electric bike” from stealing impressions from your long tail keyword “electric bike for sale”

          I would add “+electric +bike +for +sale” (modified broad match) in the negative keyword list for the root SKAG “electric bike”. What will happen is any new search queries performed would no longer show up in the Search Term Report of the root SKAG “electric bike” and instead show up in newly created SKAG “electric bike for sale”. Bonus: You will have ads more relevant to that search query too.

        • Lom Amber

          Original SKAG “electric bike”
          Adding new search terms:
          New SKAG1 “electric bike for sale”
          New SKAG2 ““electric bike for lease”
          New SKAG3 ““electric bike for adults”

          New Setup:
          Original SKAG “electric bike” (negatives in pharse match on ad group level: “for sale”, “for lease” and “for adults”)
          New SKAG1 “electric bike for sale” (negatives in pharse match on ad group level: “for lease” and “for adults”)
          New SKAG2 ““electric bike for lease” (negatives in pharse match on ad group level: “for sale”, and “for adults”)
          New SKAG3 ““electric bike for adults” (negatives in pharse match on ad group level: “for sale” and “for lease”)

          this way you “force” Google to use proper SKAG

  • Berit V

    Hey Johnathan,

    I stumbled over this when googling myself insane to find answers to my endless open end questions about how to get the Adwords animal under my control. I hope you can perhaps answer some of my wonderings (regardless I am trying this practice because I am sick of being robbed by Google and really want more control over who sees our ads):

    1. Search Ads: I need to give you this brief background: MICE events, b2b and b2c participants. I am looking for keywords that the potential audience is googling whilst working in their field. Now, these keywords are still connected to the landing page, but just not as direct. Example:

    directly linked keywords that I am already using: finance conferences in England
    what I am looking to capture: financial leadership technologies

    My difficulties when undertaking this:
    – suggested bids start from 5€ and up. Firstly the budget is no where near this PPC and anyhow, I am not looking to pay more than 0.6€ PC just as a principle. Or am I wrong?
    – Do you think this is a waste of time and energy to try to achieve this with Search Ads and I am better off focusing on the direct keywords here?

    2. Display Ads: same background same goal. I am currently using Topics and refining by using Keywords. Not impressed with the results. Too high CPC, too low CTR. Moreover, when going through the placement report, the placements are total bullshit, in my opinion, nothing close to what the topics I have added are about. Ah and I’m using the responsive ad and topics I’ve added are of course sub-topics not the main topic.

    – How can I improve the DSN campaign? I feel like I’ve tried everything, but clearly not because I have not beaten this yet.

    I really hope you have time for this & I appreciate!


    • Sorry for the delay in reply, Berit!

      1) You’re going to have a very hard time paying that little on Search. Even if there are no competitors for those keywords, Google has a minimum bid floor. You gotta pay to play or not use AdWords Search at all.

      2) You gotta know the temperature vs CTA scale for getting Display to work. Doesn’t matter how refined your targeting is. Read this:

      3) What’s a DSN campaign?

      • Berit V

        Hey Johnathan,

        not to worry, happy to get an answer at all. 🙂

        1) thanks

        2) Hmm okey, I get your point. However, do you think only this would make the ads work better? Would you use topics and define by keywords as well or is there a better way?

        3) DSN is for ‘display’

  • Nick

    Hi Jonathan,

    Great article. I’m about to set up some SKAGs for the first time and was wondering about organisation. We currently have about 100 products and thinking of adding more, would you recommend 1x Campaign, with 100 “root” SKAGs in it, or split them out into seperate campaigns?


    • Campaigns are more for themes. So if those products are in different categories, then create campaigns for them.
      That’ll just help you have better oversight in my opinion 🙂

  • Simon

    So I implemented this about a month ago for and it’s a great concept – thanks Johnathan.

    I made a slight amendment though. I created separate campaigns depending on how many keywords there are. So I’ve got a campaign with 2 SKAGS keywords and a campaign with 3 SKAGS keywords. I’ll add 4,5 etc once I have a reason to do so.

    This way I can split the budget between the slightly more broad 2 keyword terms and the slightly longer-tailed 3 words campaigns.

    Not sure if that’s a thing but it’s something I thought I’d try out – it helps a bit with organising the account.

    It’s probably worth pointing out that this is a SAAS business rather than an e-commerce business, so I needed to keep a closer eye on negative terms in the earlier part of the campaign.

    For example, +google +business +reviews is a valid modified broad search term for someone potentially looking to collect reviews for their business, but it does attract searches like “[company name] business reviews in google” or similar. So for non e-commerce applications this is something to be aware of. It’s mitigated using a list of popular business names as a global negative keyword list, which gets added to over time.

    CTR and QS has gone up since implementing this strategy.

    • Awesome to hear! Never heard of the campaign structure though 🙂

  • Janine Hosey

    Great article! After reading, I can’t think of a reason not to use SKAGs. Quick question though, regarding proving this strategy. You said “The fastest way to remove this fear however, is to test the SKAG approach in your top five keywords and compare their metrics (apples to apples) against the same keywords who are not set up in SKAGs.” — I assume this means a before/after analysis of these keywords?

    • Exactly 🙂 In fact, you’re going to have to compare keywords to search terms.

      • Janine Hosey

        Thanks for the quick response, Johnathan! Now that I’m implementing this strategy, I have another question. As I’m adding the individual keyword in 3 match types (exact, phrase, bmm), my question is regarding word order in broad match modified keywords. Since word order doesn’t matter, would these 2 keywords be considered duplicates? +red +nike +shoes and +nike +shoes +red? It’s my understanding they would be dupes & would compete against each other in auction & drive down qs, imp share, etc, therefore should only have one of the keywords.

  • Ankit

    This is one of the best adwords post I have read till date!

    Instead of putting all the match types into the same ad group, what do you think about splitting the match types into 2 ad groups – First with only Exact match keywords & the second with Modified Broad Match + Phrase Match Keywords.

    The idea here is to separate your Keywords by – Keywords that you have already discovered & an Ad Group to Discover new Keywords.

    Could this be a more efficient approach?

  • Steve Woltmann

    I just tried the SKAG method and as I sit here, my CPC are going up and my position is dropping. What I did was close an existing campaign, then setup all SKA’s. Once that was done, I paused the original. Glad I didn’t restructure it and made a copy as the SKAG is not doing well at all.

  • henry

    Hi Johnathan, ive created around 1500 skags now, but i want to know what threshold you use to ensure you dont create a skag with low search vol.

    Currently i sort my search term report by impressions over “last 30 days” – i then set up relevant skags with 10 impressions or more in that period.

    Do you think there is value in lowering this threshold, to say 5 impressions a month or more? I would certainly get a ton more skags this way, but just dont want to have them all be low impressions as this would reduce my traffic.


    • Are you finding value in creating SKAGs from 10 impressions? If so, then I’d recommend going further down the list.
      1,500 SKAGs is a lot to create if you don’t know that it works.

      • henry

        Yeh, it’s working well. This 10 impressions was my threshold as i was trying to avoid the trap of “low search volume” and keyword being redundant.

        But there are 3000+ more skags i am missing out on by not lookinf from 9-1 impressions in last 30days.

        How low do you go? I could easily create all these 3000+ terms, but surely some would just result in low search volume. And the reaulting negative kw that gets added to the root skag would mean no ad would show –> less traffic.

        So is there a pointer you use?

        • Completely up to you, but there is a point of diminishing return as you continue down the list 🙂

  • Dale Powell

    Hey man just a quick question I need some help with. My root keyword is Rat Removal but I want to target Rat Control and Rat Exterminator. Would these two alternatives be their own SKAG?

    • Yea different SKAGs since they’re different root keywords.

      • Dale Powell

        Thanks man!

  • Sudip Samaddar

    You suggest that the modifier search is there to give an idea of new keywords. Now suppose I have a root keyword as “predictive dialer” .. and i get from search terms another positive keyword idea as ” outbound dialer” . Should I create a new campaign with “outbound dialer” or a new ad group within “predictive dialing”..

    • You won’t get the search term ‘outbound dialer’ from the keyword +predictive +dialer – but you should build out new SKAGs from the search terms as you see them mismatching with the search terms.

      • Sudip Samaddar

        One more help..please please. What bid strategy is most recommend while implementing SKAG. I mean the most recommended bid strategy when we first time start it.

        • Manual CPC for big changes like this at least to begin with 🙂

  • Sudip Samaddar

    I have an adword group of crm dialer. I am getting search of dialer crm. Should I create new ad group. If yes, what should be the negative keyword to the original ad group.

    • Yes you should, and this should be the negative: [dialer crm]

  • Jason King

    Oh yes, this works nicely. I manage a dozen adwords accounts, all of them for charities with Google Grants, getting up to $10,000 worth of ads per month. Back in June, Adwords imposed a harsh but fair quality filter on all Grants accounts, causing some (generally the least well structured, or with the poorest website content) to suffer a drastic drop in impressions.

    What to do? I’d already weeded out poor quality keywords and ads, and improved the landing pages. Aha, let’s try SKAGs. I went through half my accounts, identified the best performing keywords and put them into SKAGs. Also, identified the least well performing ad groups and turned them into multiple SKAGs, to see if they could be rescued.

    Admittedly I didn’t always interpret the S in SKAG as meaning “single”. Sometimes I put “several” keywords in a group, but only where they could ALL be fitted into the headline of the ad. Didn’t seem to make a difference, and saved some time.

    Result! An immediate rise in impressions. Every SKAG group is performing well. Some ad groups jumped up to 16% CTR.

    So thanks for the nudge to try SKAGS, I’d been putting it off because of the time involved, but it was well worth it.

    • Awesome to hear, Jason! 🙂

      It does take time, but well worth it. Hope you use this foundation for future campaigns first.

  • Seb Abecasis

    Hi Jonathan,

    Firstly thanks so much for this, using SKAGs I doubled my conversion rate and halved my CPC, giving me tons more conversions and causing my CPA to plummet!!

    One thing I still don’t quite understand is why we need phrase match in SKAGs. Surely, broad match modifier would capture all those search queries?



    • Reason why is that all match types perform differently. So with phrase match, you might actually control 10% of what broad match modifier would capture – which gives you control over a CPA and conversion rate.

We help our clients,
get more clients


What's In The Proposal?

Improvement Tips
Improvement Tips
Competitor Intel
Competitor Intel
Our Pricing
Our Pricing
... and much much more!
Last Step
Previous step

When it comes to PPC, the first person I turn to is Johnathan Dane. He and his team cut through the bullshit and get straight to the point with the goal of making you more money. Work with him.

Peep Laja
Peep Laja
Founder @ ConversionXL
Please enter a valid URL (e.g.
Send my free proposal
Previous step

See Our Case Studies Too



Please look out for an email from us.

Close window

problem here.

try again