It seems that “77% of business decision makers consider big data a priority” according to a research on 500 marketers  

As marketers, we gravitate heavily towards quantitative data. We’re constantly knee deep in Google Analytics reports, traffic numbers, conversion rates, and a lot more…

And for good reason, quantitative data paves the way for measurement, analysis, comparison – and consequently, higher conversions. But a bias towards quantitative data results in the neglect of another crucial counterpart of conversion-oriented research: Qualitative data.

In this article, I’m going to show you five ways to use qualitative research to generate insights for your optimization plan.

But first…

Have you seen the movie “The Wolf Of Wall Street”?

There’s a scene in the movie in which protagonist Jordan Belfort, played by Leonardo DiCaprio, asks a friend to sell him a pen.

His friend then tries to sell the pen by mentioning supply and demand.

qualitative research intro

Image source

Well, the real Jordan Belfort (who the movie is based on) had a different view:

The real answer is, before I’m even going to sell a pen to anybody, I need to know about the person, I want to know what their needs are, what kind of pens do they use, do they use a pen?

This powerful scene captures the contrast between qualitative and quantitative data.

All research orbits around the understanding of human behaviour. But different types of research shed light on different types of problems. They also represent data and problems in their own ways.

Quantitative research – like supply and demand numbers – uses numbers, statistics, measurements, and comparison to find out what’s happening. Qualitative research – like interviewing customers – uses observation and inquiry to ferret out the answer to one question:

The Unyielding Power of “Why?”

It places a heavier on emphasis on the motivations behind the human beings generating data. Qualitative research is:

  • Who your customers are
  • What your customers want
  • The language your customers use

Through proper qualitative research, you can skyrocket the ROI of all marketing activities. It empowers you to create winning test hypothesis, accurate personas, and pivotal changes in your website, copy, landing pages, customer service, value proposition etc.

And guess what?

It’s more important now than ever before…

Edelman’s consumer marketing study surveyed over 10,000 people across 8 different countries. The study found that 51% of people felt that brands are under-performing when it comes to asking about their needs; and that only 10% of brands are doing this well.

This means you can cut through the clutter and make a huge leap in conversions; if you know how to truly understand your customers through proper qualitative research.

qualitative research graph

How To Conduct Qualitative Research

So, qualitative research plays a massive role in boosting conversions, but how is it conducted?

The 5 most optimal qualitative research methods for conversion optimization include:

  1. Online surveys with recent customers
  2. User Testing
  3. Web Traffic Surveys
  4. Live Chat Transcripts
  5. Customer Support Insights

1. Online surveys with recent customers

Online customer surveys can be great for hard-hitting qualitative research. They’re more scalable than live chats and individual interviews, and allow you to mine data from more people at a faster rate. Plus, over 24.8% of customers are willing to complete them.

Surveys allow you to peek into the minds of your customers. You get to learn about who your customers are, how they view your product/service, what their pains and problems are, and what language they use.

Andra - qualitative research


As Andra Baragan, Head of CRO Strategy at Omniconvert says:



“When surveying your customers, you don’t want to just survey anyone and everyone.The best way to do it is to survey only recent customers because they won’t hold a biased opinion, and their experiences with you are still fresh. If possible, survey them immediately after they purchase. The faster you survey post-purchase customers, the more accurate your results will be.”

A fatal mistake to avoid when surveying your customers, is not knowing why you’re surveying them. Okay, you’re surveying them for research, but what do you hope to learn from the surveys?

Remember, the answers you get are only as good as the questions you ask; surveying is no different. Which is why your questions should relate to:

  • Who your customer is – You need a general idea of who your customer is before you can market to them. This is essential for creating accurate customer personas.
  • What their burning pains and problems are – What’s the driving factor behind them buying from you?
  • Problems buying from you – What points of friction did they encounter when buying from you? Was there anything that deterred them or made the process harder than usual? This will kill crucial points of friction on your site and enhance user experience for customers.

More Survey Questions You Can Ask:

  • What can you tell us about yourself?
  • What matters to you when buying [product]?  
  • How long did the buying process take?  
  • Did you look at competitor sites?
  • How many/Which ones?  
  • Which hesitations did you have before buying?  
  • What made you buy from us?  
  • What can we improve for next time?

The Process

  1. Be clear about what you are looking for, define your goals  
  2. Understand the data through an initial review  
  3. Organize the data in a way that makes sense to you and is manageable  
  4. Interpret the data and identify patterns  
  5. Summary report of findings – next steps for testing

Case study

To see how survey findings create actionable insights, let’s review survey findings for an online restaurant supplies shop…

  1. 41% of buyers are buying for personal use, only 20% for an actual restaurant  
  2. 43% of buyers nearly did not purchase because of high shipping costs
  3. 68% bought thanks to the competitive price  
  4. 48% did not trust the site because of its outdated look
  5. 75% of the customers compared the site and its offers with other sites ( 2-3 or even 6 in some cases)

What to do with the insights

Based on the survey insights above, the following changes can potentially lead to increased conversions:

  1. Any CRO worth his salt knows that: The more relevant your site is to visitors, the higher your conversions. So if the majority of people are buying for personal use; the design, image, and copy of the site can be made more relevant to boost conversions.
  2. Shipping costs are a general barrier in the ecommerce world. If customers are complaining about them, the site can run calculations to see if they can afford free shipping.
  3. If price was a major factor for customers, the site can reinforce their competitive pricing in their copy or landing pages to create a stronger appeal.

Track Certain Responses And Words

qualitative research word out

Asking participants to answer in their “own words” will give you better insights. You’ll bump into frequent words and phrases that your customers use to describe their experience with you. By doing this, you’ll weed out sly sources of friction such as:

  • shipping costs too high
  • return policy
  • poor filtering on site
  • incorrect prices until the actual checkout

And you’ll expose yourself to the language of your customers, helping you craft copy that’s more resonant.

2. User Testing

A site that is difficult to understand and navigate is a site with poor conversions.

This is where user testing comes in.

Like most qualitative research methods, user testing shows how people interact with your site and highlights points of friction…but with a unique twist: you observe real users – in real time, completing a set of tasks on your site to spot behavioural patterns or trends.

Jakob Nielsen (usability expert) showed that user testing is likely to have a strong impact on conversions. His research – which included over 66 studies over a period of years –  revealed that “the average business metrics improvement after a usability redesign is now 83%”.

Types Of User Tests You Can Run

  1. Over the shoulder testing  

This is where you observe the test in-person and watch your users. Users will be commenting on all the tasks they complete and will notify you about any problems. It’s a great method for accurate results, the only problem being that people might act differently when watched.

  1. Unmoderated remote testing

Similar to over the shoulder testing, except that unmoderated remote testing lacks a physical observer. The tasks are recorded via audio and screen capture technology. This might be more time convenient, but sometimes unrelated technical issues can occur, leading to skewed results. There’s also 20-30 minute time limit for the task to be completed.

Toby Biddle, Loop11


Toby Biddle, CEO at Loop11 emphasizes the pros of using unmoderated user testing:


“When it comes to conversion rate optimisation and finding where the real drop off points are within your product, one of the most insightful tasks you can undertake is unmoderated online user testing. Being able to set real world tasks and follow-up questions for users to complete is key. But then, to be able to run large numbers of participants through the tests is where things get exciting. Unmoderated user testing is the only way to quickly and cost effectively get statistically relevant data about the efficacy of your funnels. You’ll then have incontrovertible evidence pointing to where you need to focus your CRO design efforts. Since unmoderated user testing is cheap and fast you can repeat this process regularly, fitting in with rapid development methodologies. “

  1. Moderated remote testing  

Remote moderated usability testing works by letting you connect in real-time to test subjects. It take more time and planning, but you get a live view of your tester’s facial expressions and their screen.

This can be extremely helpful for probing and questioning testers to find out where they might be getting frustrated.

User Testing Best Practices

  • Declare what you’re testing and why. For example “We’re testing the checkout page because it has enough traffic, but few conversions.”
  • Use only 5 to 10 users
  • Target audience users. Usually, user testers aren’t buyers. But the closer the gap between your user testers and actual buyers, the more concrete your results
  • Track tasks that affect KPIs. Imagine someone testing “how does the blue background make you feel”, it’s an incredibly silly thing to test, and pointless. Make sure you’re testing against KPIs like conversions, click-through, and time-taken to complete tasks
  • Create scenarios for the users. Depending on what you’re testing, ask them to fill out forms, browse your product page and look for certain products
  • Put them through the whole checkout process

Questions to ask in your user tests:

  • Tell us what this website is about?  
  • Shop around for a [specific product] you like  Find a red shirt, in size 10, between $10 and $20?  
  • Find 3 suitable microwave ovens. Evaluate them and pick the best one for you.  Find the product “New Aviator Style Sunnies, Gold”  
  • Start and complete the checkout process – provide them with a dummy credit card if possible

Case Study

Here are common insights found after running user tests:

  • Not obvious what the site was about
  • Hard to find specific items – search not visible enough
  • No way to narrow down choices
  • No comparison features
  • Poor error validation on forms  
  • Forced account creation on checkout page  
  • Too many steps to checkout

Remember the following before you test

  1. Testers are testers, not buyers – keep in mind the fact that they are paid to do it
  2. Sample size is too small for statistical significance – don’t change anything without testing!
  3. Always test the insights you find instead of jumping to conclusions
  4. Record testing sessions – otherwise you will forget important details that you may need ocasionally
  5. Be flexible – enter the session with an open mind. Listen, don’t instruct.

3. Web Traffic Surveys

Imagine someone coming into a retail shop for your product. He briefly looks around and makes a few inquiries, but doesn’t seem satisfied.

So he starts making his way out…

Before he’s gone for good though, trained staff amiably approach the potential customer and ask why your products/shop didn’t seem like a good fit, and what would make him likely to stay.

The feedback you’d get from a system like that would be golden, and on-site web surveys allow you to do something similar on your website.

They let you see what’s going on in your prospect’s mind at critical points in the sales process.

Keep in mind that golden, cut-and-paste questions do not exist. To find questions that draw out the most useful responses, you need to segment and test.

Questions to ask:

Unlike customer surveys, on-page or exit surveys usually deal with fresh, cold traffic. Which means your method of inquiry should be different to garner useful insights.

When using on-page and exit surveys, your questions should aim towards:

  • Message Match/Relevance
  • Friction
  • Addressing Needs and Expectations

You can ask:

  • Why they came to the site?
  • Does the site match their needs?  
  • What are the sources of friction?
  • Which site did they come from? (can be great for identifying qualified traffic sources)

4. Live Chat Transcripts

The sales boosting potential of live chat can’t be disputed. Research reveals that:

  • 62% of customers were more inclined to purchase products online if live customer support is available.
  • 38% of customers have said that they have made a purchase due to a good live chat session itself.
  • The availability of live chat assistance has been found to decrease cart abandonment by up to 30%, which resulted in an increase in sales.

qualitative research live chat

Image source

Live chat sounds like the kind of investment that – while it may be worth it in the long run, is going to cost you a pretty penny in the short term. Fortunately, that isn’t true.

Live chat is easy on the budget and convenient. There are no communication fees, and no dreadfully long waits. Communication is done in real time, and you can easily have conversations transcripted and sent via email.

In comparison to other mediums like phone or email, live chat is anonymous and offers quick answers. Both of these things – anonymity and an instant solution – are incredibly appealing to potential guests. Some businesses even report up to a 70% deflection rate from calls to live chat. This boosts productivity, because agents can support multiple chats, but not multiple calls.

Live chat transcripts are great for detecting on-site/page problems. They will reveal:

  • What information is missing on your site
  • What customers don’t understand about your product
  • Why they aren’t converting

Live chat is most effective for identifying:

  1. What isn’t converting on your site
  2. Possible ideas for your next tests

Want to get the most bang for your buck with live chat transcriptions?

Giles Thomas As Giles Adam Thomas says on the Olark blog:

“After reading the whole transcript, write down one or two of the most important user sentences—in their exact words. Don’t read and then summarize or re-interpret what they said. It is very important to record the customer’s actual words because you will be collecting all the data for another analysis exercise at the end of the process. Again, focus on sentences that express a customer pain point, an information gap in the website, or a desire or goal they want to achieve.”

5. Customer Support Insights

qualitative research customer support

You’ll find it extremely beneficial to conduct interviews with your customer support team. Think about it, they’re in the trenches all day and are constantly problem solving and going back and forth with customers, meaning: they have an advanced knowledge of customer frustrations, both, online and offline. And the great thing?

There’s nothing fancy or complex involved, just ask your support reps simple questions like:

  • What are the 1-5 most common questions you get from (potential) customers?
  • How do you respond to those questions?
  • What one thing frustrates customers the most?


After you’ve done the research above…you’re more or less ready to move onto quantitative research. Only then is it time to validate the insights you’ve extracted from your qualitative research. But remember, qualitative research isn’t an option. It’s a must. Especially if you want to stand out among the hordes of websites and businesses constantly vying for your customers attention.

Banner CTA Free Trial