Shopify Qualitative Data Analysis: Three Steps To Decode Your Customers DNA
Learn three qualitative data analysis methods for conversion rate optimization.
Turn customer understanding into a landslide of leads and sales for your business.
A step by step guide on how to decode your customers DNA.
When it comes to increasing ecommerce conversion rates and sales, most people jump straight to digital analytics and a/b testing.
Using software like Google Analytics and Optimizely.
But if you really want to improve your conversion rate and drives more leads and sales to your business, you need more than just these tools.
Actually, what truly affects your conversion rates and profits is customer understanding.
Deep customer understanding comes from combining your quantitative analytics data (like from Google Analytics) with qualitative data.
Qualitative data is collected through an exploration of customer behaviour, processes like customer development interviews.
You can decode your customers behaviour using the three qualitative data analysis methods you’ll learn in this post.
What I call your “Customers DNA”.
Once you know your target customer persona inside out, you can leverage that understanding to create a landslide of leads, sales and profits for your ecommerce business.
Because conversion optimization is about relevance, and once your messaging is aligned with your customer persona, you’re conversion rates will go through the roof.
Three Steps To Decode Your Customers DNA
When it comes to analysing your qualitative data in conversion optimization there are three important steps you must follow to get the most from your results.
This three step process is what I call, “Decoding your customers DNA.”
Let’s first learn what the three steps represent.
Qualitative data can be hard to sum up, with data from customer interviews for example there are lots of different opinions, expressions, pain points and concerns.
If you do 15 interviews in a customer development process you’ll end up with a lot of different things people said about your problem space, and that can be overwhelming.
Taking all your data and depicting it using a word cloud can help you to find patterns and recurring language customer use, more on that later.
The truth is when it comes to qualitative data, analysing it takes time. Step two is to run through all your data and note the big takeaways. What one big pain point or one true goal were people focussed on the most for each interview or survey response. The goal here is to summarize each data point into one or two sentences.
Step three is all about quantifying your qualitative data. Arrange your data into categories. This allows you to measure how many people were focussed on the same one big pain point or one true goal.
Now let’s learn this process in detail using example data.
Step By Step Qualitative Data Analysis
Step 1: Depict
There are a number of types of qualitative data you can collect to improve your ecommerce store, for example to name just a few:
- Customer development interviews
- Website & exit intent polls
- Customer surveys
- Live chat
The one things all these data sources have in common is that they allow you to capture information about your customers behaviours and emotions in their words.
With interviews you have the audio recording of what the person said, with polling and surveys you have the answers they typed in and with live chat your have the transcripts of the conversations.
These data sources are a huge untapped resource for insights into your customer, their one big pain point and one true goal.
Step 1.1: Gather all your customers words
For different data sources this will require different tasks, let’s cover the four data sources we discussed earlier.
1. Customer development interviews – Use a service like Speechpad to transcribe your interview audio
2. Website & exit intent polls – Export the customers answers from your software, whether it be Qualaroo or my favourite, Hotjar.
3. Customer surveys – Export your survey responses from your software.
4. Live chat – Export the call transcripts from your software, like Olark.
Step 1.2: Create your word cloud
Next up you want to take only your customers words, so for interviews and live chat, ignore the interviewers half of the conversation.
Then paste them into a word cloud creator app, like this one.
What you now have is a cheat sheet of the words your customers use to describe their pain and desires.
Step 1.3: Create a list of customer vocabulary
Next you want to study your word cloud for recurring words and phrases.
For example we surveyed jewellery designer By Charlotte customers and asked them what they valued in a piece of jewellery.
You can see the results in the word cloud above:
We can now use these words to create and test a new value proposition that should be more aligned with the customer and what they desire from a piece of jewellery.
Step 2: Note
Now if you were looking for a quick fix to qualitative data analysis, I’ve got some bad news.
The truth is, to really get the customer understanding you need to skyrocket your conversion rates, you have to digest all the data line by line.
Step 2.1: Digest every interview, survey response and transcript line by line
This means listening to every interview recording, reading every live chat transcript and…you get the point.
So when you do take the time to delve into this data, you need to note the big takeaways from each data point.
That way, when you return in the future you’ll be able to skim read your notes rather than wade through the data again.
It also makes the data much more useless and actionable across teams.
Step 2.2: Note down the big takeaways into two sentences
The secret to great notation as an analysis method is to try to summarize the whole data point, the whole interview for example, in two sentences.
Sentence 1: What was the customers one big pain point or one true goal.
Sentence 2: What was the biggest learning from the data, product feature request, customer service complaint etc
Step 3: Arrange
Finally, to quantify your data you need to arrange it into categories.
Step 3.1: Categorize
The first step is to look at your biggest learning from Step 2, the second sentence you wrote with the big takeaway or focus from the data.
This could have been a bug that people mentioned or in the case of the By Charlotte survey earlier the idea of quality as a core desire when shopping for jewellery.
Now as with any data source you will have recurring patterns.
So tally up the different learnings from each source and see which feature or benefit is most prevalent.
This is often your core differentiator.
For example,we completed ten interviews when improving the conversion rate of jewellery store By Charlotte.
Here are how we arranged them.
We can then incorporate this into the marketing messaging and ads and testing for conversion improvements.As you can see, by coding the data we can see that most people were focussed on quality as their main feature when buying jewellery.
Qualitative Data Analysis Methods For Higher Conversions
When it comes to getting more leads and sales for you ecommerce store, don’t forget, there is more to CRO than just Google Analytics.
Analytics software is only half the story, as well as quantitative data there is also a second source of data, qualitative data.
Quantitative data tells you what happened, where and when.
But without qualitative data you won’t understand why.
So if you learn from your analytics that people leave your sales funnel on your cart page, you still need qualitative data to learn why.
Ask them, poll them, survey them, dig deeper and plug the leaks in your sales funnel.
With this post I’ve given you a process to analyse your qualitative data and decode your customers DNA.
For those of you who are serious about taking your ecommerce conversion rates, I’ve put together a bonus download area.
You’ll get a free cheat sheet detailing the process above, so you can find it and reference it anytime you need to.
That way, next time you try to get more leads and sales, you’ll have the right tools kit at hand.
Now tell me in the comments, what other types of data do you use to improve your conversion rates?
Just say, “Hey Giles, we collect…”