Discussing data types and how to apply them can get real complicated–real quick. But if you stick with it, data can become your best friend. Knowing how to use it to better understand your customers and improve your website can be a major benefit to your business.
And when it comes to data, one area where many fall short is the collection and analysis of the subjective, qualitative end of the spectrum. Let’s take a look at how qualitative intelligence can work together with objective, quantitative data to create a customer insight powerhouse.
What is qualitative data?
Qualitative data is collected through observation and is non-numerical in nature. In website analysis, qualitative data is used to categorize properties of a person’s behavior.
This type of data helps us recognize trends that may otherwise be missed should we solely rely on numerical, quantitative data. The best option, then, is to combine them to paint the clearest possible picture of what’s really happening.
In practice, this means pairing something like average time on page (quantitative) with session recording notes showing which content blocks are read the closest (qualitative) on each page. This sets us up well for creating meaningful optimization goals that incorporate both sides of the data spectrum.
What makes qualitative research difficult
Qualitative data is not going to show up in an automated spreadsheet once a month.
It takes time and energy to become a student of visitor behavior and to gain the experience to know what you’re looking at. This last part is the most important. Just remember, it takes time to recognize behavior trends and themes and several rounds of testing to optimize with that new knowledge. You can always take advantage of various file-sharing platforms, like StuDocu, where students share with their precious gained knowledge regarding different types of spheres.
As intimidating as this may sound, modern conversion rate optimization tools make things a whole lot easier. With Lucky Orange, you’re able to quickly see the effective fold, which traffic sources are driving quality traffic and which product descriptions are connecting with visitors.
How to gather and use qualitative data
Consider your questions
The first step in great qualitative research is deciding what you’re chasing. Asking questions like the ones below of yourself or your team will shape the research process and your key performance indicators.
- What do I want to know about my website visitors? (How they navigate content-heavy pages, which page features cause the most action, etc.)
- How will knowing this help the business?
- How will I collect data about this?
- Will I need qualitative data, quantitative data or both in combination?
- Are there any possible confounding variables at play that need to be considered?
- How much data do I need to come to a conclusion?
Then, start with quantitative data
If you have Google Analytics placed and your site is getting even a small amount of traffic, you have quantitative website data. Let’s start there.
Begin by making note of the metrics listed below and any conversion goals you’ve set up. This core data, while potentially considered vanity metrics, is a great place to start when wrapping your head around what’s happening on the site.
For illustration purposes, consider these made-up metrics from an ecommerce store.
Date range: Jan 1, 2021, through March 15, 2021
Avg. time on site: 0:32
Avg. pages/session: 1.23
% new visitors: 87%
% mobile: 15%
% Add-to-cart: 23%
Cart abandonment rate: 12%
What out of this even matters?
In this data, we see a low average time on site paired with a majority of visitors entering the site for the first time and on a desktop browser. In this instance, a good goal might be to increase time on site, with hopes this will increase conversions due to people being exposed to more CTAs and valuable content.
How can we use qualitative data to better understand why these metrics are where they are? What can we do to make better optimization tests in our first attempt?
Dive deeper with qualitative analysis
Qualitative data, in this case, helps us better understand why people might be spending less time on the site than we’d hope. It can also show us specific examples of people following certain behavior patterns that we care about.
For example, if we see a specific traffic source engaging with a CTA at a lower rate, it might be worth filtering a dynamic heatmap to only show that data. In the example below, we can see that traffic from this source is, on average, not scrolling far enough down the page to even see the CTA in question. This means we need to either move the CTA up the page or do a better job of using page content and layout to get them to scroll further.
What about the time-on-page issue? For an ecommerce store, we need to get visitors to view individual products. So, while this layout is clean, it might make sense to prioritize a few top-selling products or at least calling attention to a few more subcategories to attract different types of people further into the store.
Circle back to complete the process
Now that we have identified a quantifiable issue (time on site with new website visitors) and can pair qualitative notes (average visitor not scrolling to our interactive jeans finder tool; not giving enough opportunities to engage with individual products), we can start to optimize.
It’s worth noting that this type of research will only take a business so far. You may miss a confounding variable or be considering your qualitative findings from the wrong angle. The best way to figure out if you’re on to something is by testing aggressively, both in frequency and prominence of the test. If you’re at this point and are stuck deciding which tests to run, here’s a testing prioritization rubric you might find helpful.
Documenting and communicating your findings
No matter if you’re evaluating high-level trends on a heatmap or watching session recordings of individual visitors, you’ll need to figure out how to document your findings. A simple way to start is a spreadsheet.
Here’s what this might look like in practice. Take note of the columns included that help the researcher think through what they’re looking at and the future reader understand what was happening without having to dive in themselves.