Products that people love and incorporate into their everyday lives tend to have two things in common: They’re both useful and usable. In other words, they’re things people want to use because they fill a particular need. They’re also things that people can use because they’re not prohibitively complicated.
Unfortunately, a lot of products fail one or both of these tests pretty badly. Either nobody wants to use them because they don’t see the value, or people try using them and give up in frustration. If you’re struggling with an embedded dashboard that isn’t getting the sort of business intelligence adoption you’d like, it’s worth conducting some research in order to find out if you’ve shipped something useless or unusable. Or both!
Making Embedded Dashboards Useful
To build a truly useful analytics dashboard, you have to understand what your prospective user is trying to do or learn. What do they want? How might a product like yours improve their lives? Does it help them track their fitness progress so they can run that marathon? Does it help them optimize their sales funnel so they can increase their commission?
Fortunately, there are several great techniques for learning these sorts of things about your customers, and plenty of resources to help.
If you have direct access to your users, probably the easiest thing to do is observational research. You can learn a lot by simply watching your users perform tasks with your product or having them walk you through a typical day.
Customer Development Interviews
Sometimes you want to learn whether certain users are going to be your potential customers. In that case, you may want to try customer development interviews. Cindy Alvarez’s excellent book Lean Customer Development gives very clear instructions on how to recruit participants and conduct useful interviews.
More General User Interviews
Customer development is generally focused on figuring out whether or not people are likely to buy your product, but sometimes you just want to learn a little bit more about people’s lives and needs. A great all-purpose book for learning how to conduct better interviews is Steve Portigal’s Interviewing Users. As a bonus, this book will make you better at listening to people in general.
Identifying and Categorizing Assumptions
Even if you’ve done your research, you may still be harboring certain not-yet-validated assumptions about your users. These can be extremely dangerous. My video from the Lean Startup Conference can help you find and categorize the types of risks that are most likely to derail your product if not discovered early.
Analytics + Interviews
You had to know this one was coming. After all, if you’re creating analytics dashboards, you probably have pretty strong feelings about the benefits of using analytics to understand your own product. Obviously, you should be looking at key metrics. But you need to combine these metrics with user interviews in order to understand them fully. Numbers can tell you what is happening with your users. Only people can tell you why it’s happening.
Making Embedded Dashboards Usable
Once you’ve made sure that you’re building something people want to use, you need to test whether you’re making it impossible for people to use it. And be careful! If it’s your product, you’re probably the worst person in the world to evaluate whether it’s easy to use or not. After all, you spend most of your time thinking about your product. Your user may only think about it for a few minutes a day.
Ideally, analytics dashboards should be easy enough to use that a typical customer could figure it out quickly, and if they can’t they should be given simple instructions to get them to the next step. To understand whether you’re delivering on this, you need to test your product with both real customers and potential customers.
The single most powerful tool for this is good old usability testing. There are a million resources to teach you how to conduct usability tests, but one of my favorites is Rocket Surgery Made Easy by Steve Krug. The goal of usability testing is simply to make sure that users are able to perform specific tasks using what you’ve built. It won’t tell you if people will use your dashboards, but it will tell you if they can.
Usability testing has gotten so much more popular over the last several years that a lot of tools have cropped up to help you do it better. Products like Validately can help you recruit participants, conduct research remotely, and capture and share insights more quickly. It means that a process that used to take a week or two can now be done in a day.
There’s a really common type of usability problem that isn’t mentioned very often. In order for people to use your dashboards, they have to understand what they do and what they’re supposed to do. This seems obvious, but over and over again, I see people create these lovely, minimalist interfaces that look wonderful but leave users completely confused as to what it all means.
A great resource for finding out if your product is immediately understandable is a five-second test. You can do it yourself by taking a quick mockup or screenshot of your dashboard, showing it to people for five seconds, then taking it away and asking questions like, “What do you think that page does?” and “Who do you think it’s for?” If you want to do this without leaving your desk, you can use Five Second Test and do it entirely online.
Why Does This All Matter?
Why should you go to all this trouble? I mean, it all seems like a lot of work.
Usefulness and usability are the absolutely minimum things required for successful embedded dashboards, and they work together. It doesn’t matter how easy to use something is if nobody wants what it does. Similarly, something could be the most useful product in the world, but if nobody can figure out how to make it work, it’s completely useless.
You don’t have to use all of the techniques listed above. But if you want your analytics dashboard—or really any part of your product—to be successful, you need to make sure that people want to use it and that they’re able to use it.