How to approach your website data and improve your users’ experiences & conversions.

Interpreting website data into actions that get results isn’t a straightforward process. But we at Extellio have developed a way of thinking about websites that can help you draw meaningful and actionable conclusions that contain some real potential for improving overall user experience and conversion.

You’ve probably spent some time staring at your website’s analytics trying to figure out what people are doing on your site and why. You may have asked yourself: “Why does their behavior seem so erratic?” Or “Why don’t they just convert?”

It’s not easy to translate whatever conclusions you draw from analytics into actions or development plans. That’s partly because website analytics only allow you to see some of the picture, leaving you to fill in the blanks with guesswork. But it’s also because we tend to think about our data from one perspective at a time.

To go beyond that guessing game, we at Extellio use a simple methodology that relates the available data to three key aspects of a website:

What visitors want to do on your website
What you want them to do on your website
What they actually do on your website

Looking at things this way will put your data in a more holistic perspective and help you realize what information you lack in order to draw meaningful conclusions.

It’s a slightly more advanced approach than the average “do’s and don’ts” list, but also more powerful. It’s a simple technique to organize your thoughts and start making measurable progress.

The 3 key aspects of a website

Let’s take a brief look at each of the three aspects of a website that you need to consider when analyzing your data.

What visitors want to do

Visitors come to your website for a reason, they have something they want to do on your website - their intent. It might be just browsing around, but for most of them, it’s much more specific. They might want to find a particular piece of information, understand more about a product, use an online service, or read blog content.

What you want visitors to do

As a website owner, on the other hand, you have things you want visitors to do on your website. These are tied to your website goals. You want visitors to buy something, leave their email address, find product information, use self-service tools, and whatnot.

What visitors actually do

How visitors act on your website is a different question. They usually try to fulfill their intent, but they can take indirect routes to get there. Even if they don’t get side-tracked, visitors tend to be a bit all over the place. Smooth and direct user journeys are very rare. We’ve all experienced a “bad website”, and spent too much time looking for something we couldn’t find or trying to do something that didn’t work.

Aspects that don’t perfectly match-up?

We’ve now covered the three essential (yet, different) aspects of a website. So far, so good. But why do theymatter? The point here is that the intent, actions, and goals don’t perfectly match up. Once you realize that, it’ll be much easier to understand what’s happening on your website and why.

Let’s take this step by step not to get things mixed up. We’ll start with visitor intent and website goals.

When you think about it, it’s obvious that visitor intent doesn't always correspond with your website goals. Your visitors will want to do things that don’t always fit with your aims, and you will want them to do things that they haven’t planned on doing. We can illustrate this with the overlapping circles of a simple Venn diagram.

This is your sweet spot, the good conversions. This is where visitors want to convert to your goals and manage to do it without too much hassle.
This is your potential. Here, you share a common goal with your visitors, but they don’t manage to fulfill it on your website. The question is: Why? Is the navigation getting them lost, is information lacking, or is a function broken?
These are the things that don’t drive your business forward, but you let users do on your website anyway.
These are the bad conversions. When you manage to get users to do something you want them to, but they don’t. These are things like well-hidden pre-checked boxes about newsletters that visitors don’t always notice. People can usually only be fooled once by something like that, and the distrust it creates will very likely exceed any momentary gains.
These are your unrealistic goals. You’ll either want to reconsider them or somehow make them appealing to your visitors.
These are visitors’ unmet assumptions about what one should be able to do on your website. You can either embrace them and make them part of your own goals or inform visitors that your website won’t do them.
This is your dead weight, the areas on your website that are neither relevant for your business nor your visitors. Consider just getting rid of them.

Now that you understand all the parts of our diagram, how do you apply it to your website? How do you know what’s dead weight, what is potential, and what is in the sweet spot? Well, you can listen to what your gut has to say. You might get lucky, but it can just as likely be wrong. A safer way is to let real data guide your actions.

Apply it on your own website

It’s easier and makes more sense to map out a separate diagram for each of your target groups, e.g., existing customers, potential customers, suppliers, job seekers, employees, etc. Start with the three or four of themost important groups.

The one thing you can control is your own goals. Make sure you have thoroughly thought them out for each target group.

You can measure user behavior (what your visitors actually do) with web analytics, heatmaps, and session recordings. Most websites use Google Analytics, but if you’re concerned with GDPR, you should choose another platform or tool that’s more privacy-focused.

You can measure visitor intent (what your visitors want to do) with surveys. These are good for segmenting your data into relevant user groups as well. Segmentation within analytics data is just not relevant in most cases. Surveys can let you segment in a much more meaningful way since you can ask questions about visitors’ relation to you and the role in which they visit your website.

To be able to create a diagram that leads to worthwhile insights, you need to be able to segment your behavioral data based on your survey data and the other way around.

An example for the target group of potential customers might look like this:

In reality, there will, of course, be some visitors who want to deep dive into your excellent sustainability report, but you will see a clear pattern where certain things are important to a certain target group, while others are not.

So, to do meaningful data segmentation and to learn visitor intent, you need to use surveys. To find out what each target group does, you should then use analytics data filtered for each target group.

Investigate your potential

The next step is to investigate your potential (area 2 of the diagram). In our example above, there’s only one issue listed as potential: “Find an online reseller with product A in stock.” But in real life there will be severalissues, especially when you create diagrams for more target groups.

Investigating your potential is detective work. For our example diagram, we’d start analyzingeverything we know about the subset of potential customers that want to find an online reseller.

First, you have to try to isolate the problem. Does it apply to all potential customers or only to some of them, like visitors from specific counties? Does it apply only to visitors looking for a specific product?

Once the problem is isolated, you can then investigate why it occurs. At this level of the analysis, you get granular and look at every piece of data you have. The combination of surveys, feedback, heatmaps, and screen recordings will give you more than enough to come up with a couple of well-founded hypotheses.

Did visitors find the online resellers page? If not, where did they take a wrong turn? If they found the right page, why did they fail? Is it that resellers don’t have the product in stock? What can screen recordings tell us about the functionality on the page? Have users commented on their problems in surveys and feedback?

When you’re done, you summarize all your findings into testable hypotheses. You’ll find that they are much fewer and much stronger than you’d ever get them by using the traditional analytics-only approach.

That was the fun part. Now, you need to apply the changes that you think will make a difference to your website. You can track your changes using a KPI that applies to your problem, generally a user success score or a conversion rate.

If you want to track progress more closely, you set up a multidimensional funnel, which is a funnel combining analytics and survey data.In any case, if you’ve done a decent analysis job, you should see an effect from your changes quite soon, and you can move on to other issues.

Reconsider your goals and services

We are not quite done yet. There are more areas on the map that deserve our attention.

Area 5 for example, the things you want users to do but that don’t interest them. In general, you should consider if you want to have those things on your goal list and even if you want them on your website at all. If you don’t want to reconsider, you’ll have to put some effort into trying to build demand for them

It’s also worthwhile to take a minute and think about area 6: things visitors would like to do, but you don’t really care for and maybe don’t even offer them. There are bound to be some things there that at least could improve user experience if you offered them. They could even serve as one of your own goals, or at least as something that could get your visitors closer to one of your conversion points.

Final words

This method came about from Extellio’s need for a systemized way to lead discussions with clients around their different kinds of UX data.

We discovered that it’s a great tool for organizing team workshops to generate, discuss and prioritize hypotheses for updating websites. It’s also a simple way to communicate with the rest of the organization regarding priorities and resource allocation.

This method is not a universal solution to all website problems and not the only approach or advice you should consider or apply. You’ll still benefit from the dos and don’ts, UX writing rules, CTA guides, and so on. Our point is that it’s not always about the details (although they do matter). Sometimes, it pays off to look at the big picture and work your way down.