You know when you have a gut feeling about something and you’re usually right?
If only it was that easy when it came to conversion rate optimization (CRO).
Because when it comes to CRO research, there’s a very good chance that you’re clouded with various cognitive biases and assumptions that could lead you down the wrong path.
Playing detective and doing your due diligence when it comes to split testing can make or break your CRO goals.
More importantly, it can help you get results, faster.
So why not be smart about it and run a calculated methodical research plan so you have more accurate intel on your visitors?
With the right insights you can reach the right conclusions about your visitors, which will allow you to chase after your CRO and revenue goals with the right tools in place.
We’ve teamed up with Conversion Sciences to bring you the following tips on how to stop following your gut and start following accurate CRO research methods, so you can start growing your business even quicker.
What is Conversion Research?
We defined it in our conversion research guest blog post:
“Conversion research is a strategic approach focused on identifying and interpreting relevant data to find possible points of friction in a sales funnel – and, ultimately, to increase the overall conversion rate.”
Without the right clues about your audiences, you’ll be chasing the wrong leads and heading down the wrong path. And by that I mean you likely won’t find the conversion rates that you’re looking for.
You might even be designing your landing pages with unknown friction points that may be preventing your visitors from clicking your call-to-action (CTA) buttons.
CRO research can help you uncover useful insights about your target audiences, which will not only help you reach your visitors but also help you to convert them on your landing pages.
Quantitative Research – Numbers To Crunch
Quantitative research is the numbers crunching kind of research, where you gather and analyze data that’s based on stats, formulas, charts and graphs.
You can use these types of reports to see where exactly your problem points are, where you can improve and where you should invest more time and money.
There are several formats that quantitative research can take on. Here are some of the reporting formats that can help you analyze your visitors’ data:
- Mouse behaviors: Heat mapping, scroll mapping, click mapping, confetti mapping
- Conversion stats: click-through rates, opt-in rates, impression shares
- Device stats: mobile, desktop, tablet
- Time-based stats: load times, time spent on page, time before CTA is clicked
- Traffic reporting: referrer, number of people, pageviews, visited site
Kissmetrics found that:
“A 1-second delay in page response can result in a 7% reduction in conversions.”
What this quantitative research can mean for an ecommerce site:
“If an ecommerce site is making $100k per day, a 1-second page delay could potentially cost $2.5m in lost sales per year.”
Here’s another example from Kissmetrics on quantitative research:
By analyzing quantitative stats on where people abandon your landing pages, you can find out where your visitors are experiencing hang-ups in your conversion funnel.
Address those specific friction points on your landing page to improve your conversion rates.
Qualitative Research – Feedback To Digest
There’s more to it than just the numbers. Qualitative research can help you identify the humanistic stuff, like how your visitors are feeling, and what types of questions, thoughts and opinions they have.
Feedback tools like surveys, chat boxes, questionnaires and forms can help you gather this type of quantitative analysis.
Here’s an example of a Qualaroo exit survey, which can help you better understand visitors’ thoughts on their way out:
Another example of qualitative reasoning is StubHub’s case study. By using UserTesting as a platform, StubHub was able to collect qualitative data that helped them understand their user experience.
Here’s what they learned:
“By listening to and watching UserTesting.com panels on their site, StubHub discovered that a link, labeled ‘See Details,’ was causing confusion and in need of some improvement to boost conversion rates.”
With this qualitative info, they were able to implement some changes that included updating their button copy from ‘See Details’ to ‘Go.’
The result? StubHub increased their rate by 2.6%.
Other methods that you can use for pulling in qualitative data are:
- Focus groups
- In-person interviews
- Direct observation
- Frequently Asked Questions
User & Usability Testing – Do Visitors Do What They Say?
Even with all the quantitative and qualitative analysis, there could be more that you’re missing. Usability testing is another way to seek the truth about how your visitors behave.
By conducting user testing you can find out by tasking, recording and observing real users interacting with your site.
We may have good intentions when answering usability test questions, but oftentimes what we say and what we do are completely different things.
If you’re human, you’re likely to fall for cognitive biases and answer with what you think moderators want to hear or with what you think you’re supposed to answer.
Here are five category questions to help you structure your usability tests, provided by Jakob Nielsen at Nielsen Norman Group:
- Learnability – how easy is it for users to do basic tasks the first time they see your site?
- Efficiency – how quickly are users performing the tasks?
- Memorability – when users return, are they still up to speed?
- Errors – how many and how severe are the errors, and are the users able to recover?
- Satisfaction – how pleasant is the experience?
You can conduct testing in three main format – moderated in-person, moderated remote and unmoderated remote. Whichever format you choose, be sure to create an environment where the user has the freedom to act as natural as possible for real-life results.
Daunted by the recruiting process?
According to Jakob Nielsen, you really only need about five users per test. Here’s why:
The amount of new findings caps out after about five users. Spread your recruited test subjects over a multitude of usability tests and spend your time uncovering more insights.
Segmented Analytics – Easier Access To The Right Insights
By slicing up your analytics into smaller pieces, you can take a deep dive into a certain subject and uncover some real truths about your visitors.
Sure default dashboards are helpful from a quick ‘n dirty overview perspective, but the more you know about a specific area of your audience the more impactful you can tailor your landing page strategies.
Use segmented analytics to uncover findings below the surface. You can break down your analytics into subsets based on things like:
- Geographic location: city, state, country
- Product or service type: visitors to different category sections of your site
- Subsets of users: repeat visitors, first-time visitors
- Subsets of sessions: sessions originating from specific campaigns, which sessions turned into conversions
- Subsets of hits: interactions during a session (pageview, event, transaction), segment conversion levels (i.e. those that created revenue over a certain dollar amount)
Here’s an example of what conversion segment setup looks like in Google Analytics:
Once your reporting parameters are all set up to the way you want it, you can save your reports for quick and easy access in the future.
Comparing similar reports across different time periods can be useful in checking for significant changes month-over-month or year-over-year.
The Priority List – What To Test Next
So now that you know the importance of doing your homework, research, testing and analysis… confused about what to test first?
Fret no more.
Create a priority list that’s ranked by certain parameters so you can put your first efforts toward things that need to most attention.
There are a few different prioritization frameworks out there.
Here’s one called the PIE framework, which stands for Potential, Importance and Ease:
According to WiderFunnel, some questions you can ask about PIE are:
- Potential: How much improvement can be made on the pages?
- Importance: How valuable is the traffic to the pages?
- Ease: How complicated will the test be to implement on the page or template?
ConversionXL has their own framework dubbed the PXL framework, and it looks like this:
The PXL framework focuses on making certain ratings more objective and fosters a more data-informed mindset.
There’s a couple more frameworks called ICE – one stands for Impact, Confidence, Ease; the other for Impact, Cost, Effort.
Whichever framework you choose to use, prioritizing what to test next can help you allocate your time, money and effort in a more pragmatic way.
Laying the groundwork for proper CRO research can help you to uncover important insights about your visitors.
More importantly, insights that are based on facts and figures vs. an inkling will help you to discover the tried and true results.
Garbage in, garbage out – right?
The more accurate the picture of your audience is, the better you can strategize and tailor to each of your visitor’s needs on your landing pages, which will in turn, lead to higher conversion rates.