Dovetailing nicely with one of our previous posts on “humble pie” we’re continuing to build our data model(s), our backend(s), our infrastructure as a whole as well as the design(s) that we’re showcasing our early testers.
One of the biggest challenges when working with a ton of data is how to best show the data to an end-user in a way that’s truly intuitive. In other words, data is great but if the customer doesn’t know what they are looking at and if they can’t extract immediate value out of what they see then you’ve failed terribly.
So what we’ve been gathering feedback on from our early customer interviews and conversations around the UI/UX is simply this question, in gist:
Do you understand what you see and can you do anything with it (immediately)?
This is clearly an imperative because that is essentially our mission as we seek to help organizations become high-performance engineering powerhouses (#EngOps): We want folks to have actionable insights via the data quickly and easily.
In fact, there should be very little to interpret, especially in the beginning.
Here’s what our early Alpha Customers are looking at when they jump into the system:
It’s a bunch of data, some horizontal graphs / pills, and a lot of text. From our standpoint, much of this is actionable, besides interesting and educational.
But, that’s because we already know what we’re looking at and we already know what we can do with this data. That is not the context in which a new user finds themselves and it can be overwhelming and confusing.
The point, again, is that data is good (great, even) but if you can’t showcase or display the data in a way that makes sense or that is intuitive then you’ve lost the game before you even got started.
As a result, we’re experimenting with different UIs so we can get better UX. Here’s an example, for instance:
Essentially the same data but with a clearly different angle and approach to showcasing the data. Note, this new iteration (among many additional ones), doesn’t guarantee greater success but the point of these exercises is to get more “hits” than “misses” with our data.
And when you’re an early-stage product company you have to experiment with different views and data visualizations so you can land on one that works for most folks (i.e. not 100%). It’s all part of the process of building something that works for your customers.
You see, what I’ve come to realize is that “building something that people want” is a combination of not just providing the product in raw form but you have to build it in such a way that is approachable, palatable, and overwhelmingly useful.
This means that in a design-centric universe, sometimes the function follows (closely) the form and unless we get the form right then we might never get the user to appreciate the function (data) that lies right beneath.
High-performance engineering isn’t just bytes and bits and data all-the-things, it’s product and user-centric in equal parts as well.
[Make sure to subscribe to our new High-Performance Newsletter for more!]
Also published on Medium.