Designers have long understood and advocated for user research as a key component of our work. You need to talk to real people in your target audiences to achieve the best results. But when it comes to gathering quantitative data – the hard numbers that enable you to track behaviors and measure success – many designers feel that it’s outside their area of expertise.
It shouldn’t be. Quantitative data, often called analytics, provides valuable insights that most methods of qualitative data can’t provide. The trick is to use both together, to paint a full picture of what’s happening, why it’s happening and what to do about it. What do you need to know about analytics as a designer? Read on to find out.
It’s nice to know the numbers, but what does tracking analytics actually get you? There are a few key reasons to track analytics, whether your product is brand-new or has been in the market for years.
The first reason is to establish baselines. Baselines are numbers that represent what’s happening in your company now, which is the starting point for all your other measurement efforts. Essentially, they’re the “you are here” of your organization.
Without tracking the baseline of where you are now, you’ll never know how close you are to those business goals you’ve set, or know when you’ve met those goals and can set new ones. For example, if your goal is to increase sign-ups by 10%, you need to know how many sign-ups you’re getting now (your baseline), and you need a way to track that number over time – are you getting closer or farther from your goal?
Focus on the Right Things
Analytics also help you know where to focus your time. It can be overwhelming to figure out what to do next in moving your product forward. With good analytics, you can make smart decisions about your future activities.
Coming back to that example goal of increasing sign-ups: by viewing the flow for your sign-up process (the sequence users have to go through to successfully sign up), you can see if there’s one place that a large group of people are dropping off before completing, or if very few people even start the process.
This way you don’t spend time trying to improve a part of your product that will give marginal returns instead of solving the main problem. Once you’ve identified where a problem is happening, you can usability test that flow to find out exactly what the problem is and why it’s causing users to struggle, so you can solve it successfully.
Patterns are important for understanding the details of how people interact with your product over time. Analytics tracking allows you to see the patterns that create long-term impact and success for your brand. You can see – what path do most people take to complete certain tasks? How often do people use certain areas of navigation, or sections of your product?
This type of tracking helps you make smarter design decisions, since you can see what people are doing. Particularly, it can help you understand what parts of your product are currently working, as well as the parts that need fixing – so you don’t break something good in an attempt to fix something else. It also allows you to make smart decisions to maximize traffic, increase visibility with your target personas and close more sales.
For example, do people use primary navigation to access a certain area of your product, or do they use the in-line dashboard links? Does anyone ever access their settings? What do they change or not change there? How findable (and frequently used) is your help information vs people calling customer service? Look at the patterns in your analytics and you’ll know. By tracking and analyzing patterns you can improve success for both your customers and your product over time.
What should you track?
When it comes to analytics tracking, there’s an overwhelming amount of both data and tracking software out there, making it hard to know where to start or what to track. Before choosing your software though, you’ll want to consider what you actually want to track – this way you can choose software that works best for your needs (or alternatively, create your own internal tracking system).
The easiest way to figure out what data you need is to focus exclusively on the goals you’ve created, working from the business goals down – these will give you clear information about what to track.
How? Let’s take an example: if your goal is to get more sign-ups, then track the number of sign-ups, as well as everything related to signing up, including all parts of the sign up process: things like the most commonly visited pages that lead to someone signing up, and how many people who sign up are actually in your target audience or primary personas. Whatever your goals are, there are actions, workflows and data points that you can track to better understand what is happening now and how it changes as you make updates.
At the very least, always track any activity that makes you money. The most obvious of these is any product or service you sell that someone can buy right on your site. You’ll want to track workflows around the checkout process, especially drop-off points and abandoned cart rates; these are crucial to your business.
If you have an app with existing users, you can still track monetary opportunities with those users. Are there features that require an upgrade? How many people try to use those features without the upgrade? Are they converting? If not, what are they doing instead? Any action someone takes that could lead to money for you is a definite on the list of things to track.
You will also want to track feedback information, which can come from social media messages, logs about the frequency of error messages in your product, customer service or help desk questions, feature requests, and other similar sources. This data provides you a direct line into the concerns, struggles, and frustrations that your users have, at scale.
This type of data is excellent, because while it doesn’t necessarily bring in more money, it can help you build customer loyalty and prevent churn. By monitoring this, you can enable yourself and your team to make smarter decisions about where to spend time researching, testing and improving.
Now that you have a good sense of what to track, it’s time to figure out how to track it. We’ve covered some of the most common methods and tools here, but it’s important to find the ones that work best for you.
The mother of all analytics software, Google Analytics works for almost every organization no matter the size, industry or purpose. There are two main reasons everyone uses GA:
- It’s free
- Google has access to data across the majority of the internet
Those are really good reasons to use Google Analytics, and if you use nothing else, set up your organization with a Google Analytics account. Even if you don’t have time to view it now, tracking starts as soon as you set it up, so when you’re ready, your data will be there.
Once you’ve got Google Analytics set up, it can be intimidating to figure out where to go to get useful information. Here are 3 reports we recommend as a starting point:
Users Flow. This report is located in the Audience section of the GA reports and it is possibly the most valuable non-event-based report available. It’s exactly what it says – users’ flows. What this report shows you is the path people take to and from each page on your site. This is a perfect report for understanding where people are going in your product, where they drop off, and how they arrive. The great thing is that you can select any page or you can follow any audience segment to see their specific flows.
Mobile Overview. This report is also located in the Audience section, in the mobile overview. The most important part of this report isn’t the graph; it’s actually the table below the graph, which shows you the breakdown of which devices people are on when they visit your product. This is especially helpful if you are considering something like creation of a native mobile app – this will tell you if enough people are using mobile to make it worthwhile.
Events. Located in the Behaviors section, the entire events subsection is a wealth of valuable information about the events you’ve set up for your product. You’ll be able to use this area to understand the results of your events – are people starting to fill out your contact form but not hitting Send? Do people forget to save their notification preferences? If you’ve set up the event, this is where you can track and view it.
Google Analytics can seem overwhelming, but it contains a wealth of useful data, if you know where to look. Start small and build up your understanding of the data available and reports that give you useful information as you have time – it’s better to keep focused on useful data rather than trying to use any data you can find.
A/B or Split Testing
This method of data collection is fairly popular at this point, so it’s likely you’ve heard of it (or even run your own tests). An A/B test or split test is a test in which you create two or more variations of a product, most commonly website or sales page designs. The multiple versions are then randomly displayed to people using the product, and results are tracked on which “performed better”. To successfully run an A/B test, you need a few key things:
Multiple variations of the product you are testing. These can be as simple as a button color or heading change or as dramatic as two completely different products.
A measure of success. This is what you will use to determine which variation performed better. For webpages, it is usually related to sign-ups or purchases; with applications, it is usually related to usage, retention or customer service/help requests.
Software or a method for serving variations randomly. You can’t test multiple versions without actually getting eyes on each version – this means that you a way to do that; you can do it manually, by creating multiple ads which drive traffic to a different URL for each version, or you can get your development team to set up some code that delivers different versions. Alternatively, you can use a service that allows you to create variations for testing and automatically serves up different versions.
Be sure to carefully consider the impact to your users when you A/B test, especially when A/B testing an app or parts of an app. I learned this the hard way in January of 2015: I had just decided to use a free trial of a UX project management tool as part of a course I was consulting on, and reached out to the software team to finalize the details. Two days after all students had been set up, the company launched an A/B test on their entire product, and included the students in the test. Suddenly I was trying to troubleshoot problems for software neither me nor my colleague could even see for half of the class. Needless to say we did not use that software for subsequent courses.
Customer service reps (CSRs) are an organization’s most underutilized and valuable source of data. As noted, they have a direct line to the concerns, frustrations and needs of your customers at scale. The trick is to implement a system to capture that data.
If you’re lucky, your CSRs are using a CRM like Salesforce or Zendesk, where you can automate your tracking processes, at least partially. For these systems, set up specific tags or labels to classify feedback based on features, behaviors, or functionality. Then ask your CSRs to apply the tags to each conversation they have; this will help you view all the feedback of one type at a time and look for common patterns.
If your company is not using a CRM, you can implement a manual method that does the same thing. Instead of tags in a software, create a spreadsheet with one tag per column, and ask your CSRs to tally the number of conversations they have about each tag topic. It won’t be as detailed (unless your CSRs are willing to add a lot of info into the spreadsheet), but it will still give you insight into the areas of biggest concern for your customers and visitors.
Internal Logging System
If you’re at a company that has an existing product out in the world, it’s likely that you have access to another group of people who are already thinking about analytics (on some level): systems engineers. Anybody who manages servers or databases is probably already using a logging system to record certain events at the server level that might indicate trouble: things like failed login attempts, or database requests coming from outside your company’s environment.
These might not sound exactly like the events you most want to track, but they can still be a source of valuable information that can influence design. Start a conversation with your engineers, and ask them what types of events they track in their logging system, and how & when they typically review that information. You may be surprised by how much you can learn from the tools they’re already using.
If you build up enough good will you can even ask them to track specific things that support the work you are doing. And if you are starting with a new product, you can make your tracking needs part of the product requirements as it is built.
Much has been written about qualitative vs. quantitative data collection – but the question “which is better?” obscures the fact that a healthy product creation process needs both. Analytics can tell you the what, but usually not the why; usability testing gets you into the mind of your users, but won’t help you measure your success against the hard numbers that define and shape your business.
While you can do testing almost anytime, the lesson about analytics is: start right now. The sooner you begin to build your baseline and understand how things are today, the better you’ll be able to track your progress toward your goals.
If you’re ready to go beyond the basics, Measured UX is the data + design program to get you there. The program features a self-paced 6 module course with 24 lessons to get you started now, coupled with an in-person workshop for hands on learning and individual instruction. The results are powerful:
- Build trust with your stakeholders
- Make better, more usable products
- Gain confidence in the work you’re doing and the impact you’re making
Sign up for Measured UX today and go from UX designer to trusted product leader.
🙏 This article is sponsored by Pixels for Humans.