Enterprise Dashboard Design: Help Your Users to Connect the Dots
A typical enterprise dashboard looks something like this:
A few big numbers (people like big numbers, right?), a couple of charts, maybe a table. In theory, all these elements tell a coherent story to the users, which they can act on. However, the reality is a bit different.
This is how most people see the typical enterprise dashboard:
Everyone sees the dots, but few people are able to see the big picture behind it. Why? Data is quite complex and we ask users to connect all the dots themselves to see the full picture.
There’s usually a reasonable explanation for that complexity in enterprise software. Hard things are hard. There’s an inherent complexity in complicated domains like government, healthcare or marketing. But, you shouldn’t need to have a Ph.D. in data science to be able to read a dashboard, right?
How can design help?
In June 2016, when we started re-designing of one of our experience results dashboards, we had the exact same problem. Our behind the scenes tech was great, but our users couldn’t decipher our dashboard.
There are a lot of numbers here and they are all quite important and informative, but it’s rather difficult to understand what’s going on.
A bit of background
Qubit is a customer experience management platform. Companies like Emirates and Topshop use our software to track, analyse and manage interactions with their customers.
How does it work in practice?
Let’s say you have an e-commerce store that sells ocean cruises, but people don’t buy your cruises enough. There are loads of possible reasons but you’ve done your homework. You’ve had a look at the analytics and see that a lot of people who browse a website drop off on the cruise page. They simply don’t click on “buy tour”.
Why? You’ve talked to your strategist, done some more user research and learned that people are quite reluctant to pay £599 for a tour. It’s a pricey purchase so it’s difficult for them to make up their mind right away.
Your strategist suggested you should use social proof because it works well in similar situations. Awesome.
Next step: prove that social proof actually works.
Running a successful experience is like running a successful scientific experiment: you should define the goals you want to use to measure success; and pick the right method. In this case running an A/B is a good idea since you want to get a proof that your new experience is better than the old one.
So if you look at our dashboard again, it makes a lot of sense. Kind of.
Data doesn’t speak for itself
The main problem we discovered during the initial user research is that not all our customers understood what is going on the dashboard alone. So they would ask our in-house strategists to interpret the data for them.
The challenge: how could we re-design our dashboard so it is easier to understand, without losing its scientific rigour?
The first attempt we made was to break down information into smaller bits and structure it a bit differently.
That didn’t help much, though. People liked the new design visually but still struggled with understanding what was happening.
During our 105th session (well, it felt like it) with the data science team, they tried to explain to me why all the numbers are important and what they mean. Then it struck me.
What if we could just say what was going on in a simple human language?
This way we are connecting all the dots for the user, so they can get what they need to know without all the guesswork.
And advanced users can still see detailed information about each goal if they want:
There are a lot of puzzles in enterprise design and there’s no universal recipe for how to solve them. However, there’s one principle, which I borrowed from an artist David Shrigley and constantly remind myself of:
Weak messages create bad situations
In our case, this means that if there’s a risk that people can misinterpret data, they will misinterpret it. That is why it’s so important to always to zoom in and out, to make sure you’re delivering the right message. Sometimes, it’s way too easy to miss the forest for the trees.
If you wonder what’s in that puzzle above, here you go, I’ve connected the dots for you:
Thanks to Suganya Sivaskantharajah, Sarah Benson, Sophie Coleman, Giovanni Luperti, Jen Stott and Pavlo Huk for reading drafts of this. Shout out to Qubit engineering and product team for making this possible.