I’ve written previously about the fact that the enterprise UC space being disrupted by consumer technologies. One of the key drivers of this consumerization of the enterprise is the superior user experience (UX) that the consumer tech bring with them. A superior UX leads to greater end user adoption and even grassroots support for an enterprise deployment.
For those of us in the enterprise UC space, this leads to a simple question: how are these consumer guys creating such a superior UX? I’ll answer this question with an example from one of the companies I admire. Last year I saw a presentation from Ken Florance, VP of Content Delivery at Netflix. Netflix is an awesome example of a consumer tech company upending an established market with software that has a great UX. Of the many interesting things covered in that presentation was an insight into how it engineers its UX.
A Deeper Look at Netflix’s UX
The older UI you see on the left relies on a DVD rental store metaphor with DVD boxes and bubble callouts (which I understand were called Bob for “back of box”). On the right is what Netflix calls the “cinematic” UI, which is a more visual, immersive experience. How did Netflix get from the original design to what is in place today? A large part of the answer is that it evolved the UI in a data-driven way. Let me take a moment to explain what that means. The company established key data metrics or KPIs that it was interested in optimizing. As an example, one of the key metrics mentioned was the total amount of time that a Netflix subscriber spent watching streaming video through their service in a given time period. Let’s call this total watch time per user.
Since its service is entirely software based, it actually measured this end user metric while trying different UXs. So they might try a different UX for a subset of its users and see how the total watch time
KPI differed from the baseline end user population. This is what is called an A/B test, where the test group getting the modified UX could be in group A vs. the rest of the users with the unmodified UX in group B. If Netflix found that a UX change increased the total watch time metric, that change was good.
And if it made a change that decreased that metric, that was bad. The amazing thing about this approach is that it is data driven. Netflix did over 1,200 A/B tests last year alone to help define how to evolve their UX.
The thing to note is that this is very different than what I have seen in enterprise software companies.
In those, some design mockups are potentially put together upfront, and maybe there is a debate about which design is better, but then it is handed off to engineering to implement. Many enterprise software companies aren’t measuring the success of these design decisions on an ongoing basis like Netflix. And it’s important to also note that the cloud model makes this sort of data-driven approach much easier than a premise-based software deployment.
Why Enterprise Tech Should Take Note
Making UX a data-driven practice is really one of the key approaches that consumer tech companies have mastered, and it is also what enterprise software companies need to learn if they want to stay competitive.