Dashboard Confessional: How to (Not) Build the Perfect User Dashboard
Think about the dashboard of your car. It tells you enough information and gives you enough control to operate the vehicle. It doesn’t surface every last detail about the current status of the engine. It gives you alerts when something goes wrong so you can then take action to address the problem.
Now think about the dashboards you’ve encountered in various digital products. The very best are like your car’s dashboard. They know what information and alerts you need to get your job done.
Most digital dashboards are nothing like that. They’re full of too much information, fancy but irrelevant graphics, and alerts that don’t lead to action.
Why is that? Because, like a lot of things in life, dashboards that look and feel easy are not easy to create.
Here are four things to consider when your business or product leaders start talking about building flashy and eye-catching user dashboards.
1. Know thyself and thy audience.
The most important thing about building a dashboard? Knowing what your user is trying to do.
That sounds easy, but it takes time, dedication, and persistence to uncover. That’s because what people say and what they actually do are often at complete odds with each other.
To build actionable and useful dashboards you have to know what each user type is trying to do. There are many types of dashboards: key performance indicator (KPI) dashboards, analytic dashboards, business intelligence (BI) dashboards, and productized application dashboards, to name a few. Each has its strengths and weaknesses.
Your users all have different roles and responsibilities. It’s not their job to tell you what those are. You have to go out and find it with thorough user research. You need to see for yourself the end goal your users are trying to accomplish. Once you understand their goals, and how they get to those goals, you can draw backwards to the dashboard that will help them get there.
But it’s not enough to know your users. You also need to know yourself. Are you capable of providing what your users want and need? Will it break the product or otherwise compromise its functionality?
Be realistic about the time it will take to build a dashboard that will actually meet user needs. You might find you can only build a dashboard that’s useful to one out of five of your user types. Great! Take the resources you would have spent on four dashboards and build something else. Trying to build a dashboard without uncovering your user stories is like trying to design the cover of a book before you’ve written it. You might make it look pretty, but it’s still empty on the inside.
Trying to build a dashboard without uncovering your user stories is like trying to design the cover of a book before you’ve written it.
2. Keep it Simple
The problem with most dashboards we’ve helped redesign is they don’t have the right balance of information and action.
Part of knowing your users is knowing what alerts will help them in their work. They expect the dashboard to help them do something in as few clicks as possible. Too many dashboards look pretty but they don’t offer a path to take action.
Go back to your car’s dashboard for a second. The core purpose of the dashboard is to help you operate the vehicle. Videos about how to drive a car or updates about every historical marker you pass might be interesting, but they don’t help you adjust your speed and know when to get gas so you can get to your destination.
Too often, dashboards become a catch-all for everything leadership thinks should be there and an alternative to making tough product decisions.
Most users come to a dashboard to do something, even if it’s to export data. A sure sign that a team hasn’t had the time, resources, or desire to listen to users is when the dashboard has pretty images and more information than a user can manage.
An informational dashboard is a passive experience that looks good on marketing websites. An actionable dashboard helps you get things done.
3. Avoid the Customization Trap
How many times have you actually customized a dashboard? Customization is often shorthand for, “We’ll let the user figure it out.”
Your user’s job is to do what they came to your product to do, not to finish creating their own dashboard. One dashboard is almost never right for everyone, but it can be right for 80% of your users and use cases. Find the right type of dashboard for each user, don’t try to cram all your users into one dashboard.
It’s tempting to throw as many features and options into the dashboard as possible and then stick a customization option on it. We’ve only seen this work in rare cases where the user is doing technical day-to-day work.
Build your dashboard for each user type so they don’t even have to think about it. If you launch a dashboard and get bombarded with questions or requests, then you need to fix it, not add a customization feature that most users will ignore.
4. Focus on Iteration, Not Perfection
It’s time for some brutal honesty. We’ve never deployed the perfect dashboard. We’re pretty sure it’s like Sasquatch and the Loch Ness Monster - rumored to exist, but unlikely to be real.
It’s difficult to navigate the resource constraints, user issues, and technical challenges of building any dashboard. That’s why you should never underestimate the effort it takes to build one.
Products are a complex combination of features and functionality. They're changing and evolving all the time. Your dashboard will have to change and evolve with your product, adding to any resource constraints.
It takes a certain amount of will and courage to stop and take a step back to ask if a dashboard is really what you need. Because, while a perfect dashboard may be a unicorn, a useful and actionable dashboard is worth more gold than the pot at the end of a leprechaun’s rainbow. It’s possible to build, but it will take significant effort and resources to do it.
A dashboard can reveal more than the information that your user wants to access. It also can reveal the dysfunction in your product and your team. If you’re not grounding your decisions in solid user research and building something that users will actually use, you’re probably wasting time and resources that would be more valuable elsewhere.
Need a Dashboard Confessional?
We’ve been down this road many times across different industries and can help you identify the potential risks along the way. We’ll also help you build the closest thing we know to a perfect dashboard if that ends up being the right fit for your users. Let’s talk about how we can help you get clarity around what your dashboard should be and if you should build it.