It’s the End of the User Interface As We Know It

Rob Bernshteyn
Rob Bernshteyn
CEO & Chairman, Coupa Software

Rob Bernshteyn previously served as part of the executive management team at SuccessFactors (now part of SAP), running product marketing & management for this leader in HCM SaaS software. He also worked in product management at Siebel Systems, management consulting at McKinsey, and SAP systems integration consulting at Accenture. Bernshteyn holds an MBA from Harvard Business School and a Bachelor of Science Degree in Information Systems from the University of New York at Albany.

Read time: 11 mins
Laptop with spend management software.

Laptop with spend management softwareIt’s been proven beyond a shadow of a doubt: Usability in enterprise software matters. As proof, one need look no further than the recent rush by legacy enterprise software vendors to get to market with new user interfaces.

But what if processes could happen without the user having to interact with the machine at all? What if the user interface could just disappear? It’s the logical evolution of usability and an idea whose time has come. This is the way people want to work today, and the future of enterprise software. We got a glimpse of that future at Coupa Inspire last week.

Usability is the fundamental hypothesis behind Coupa. With older procurement systems, people stayed away in droves despite mandates to use them. By creating a frictionless user interface that shifted procurement from push to pull, people naturally gravitated toward the system as the easiest way to get what they needed to do their jobs.

Extending the same user interface to invoice and expense management helped companies to capture a much higher percentage of spending transactions. But an interface is still an interface. A great interface saves the user a lot of time and energy. If you take that to its extreme, it would require no time and no energy. And no interface.


Behind the two-way mirror

Cloud technology is making this possible. Using the analytics of our cloud platform, we’ve spent six years monitoring how people are using our products, and optimizing every single screen to make it easier for humans to drive the software.

Looking at a simple example, a few years ago the average amount of time spent on the screen to approve or reject a requisition was eight seconds. Effectively, there's one thing that's being asked of the user on that screen, so that seemed to us like a long time.

We hypothesized that people’s eyes weren't easily moving to the approve and reject buttons. So we increased the font sizes and made the approve button green, and the reject button red, cutting the average time on that screen in half.

Watching all of the common use cases in our platform, we began asking ourselves, is there new technology that we can apply such that people only need to interface with the software when they have value to add? Which tasks require human interaction, and which could be performed by the software alone?

Trot, trot to Boston

Let's see how that might play out. Let’s say I'm on a business trip to Boston. I arrive around 7 p.m. and check in to the Marriott on Copley Square. I’m starving so I go over to Haru on Boylston Street for a cup of tea and a bento box and I get a receipt for $34.02. I take a picture of it and put it in my Coupa Wallet.

Back home in San Mateo, I open a new expense report and go into Wallet. I see the receipt, I type in Haru, dinner, $34.02. Then I go onto the next line item, and the next. It takes me 20 minutes to record everything.

This is how people use Coupa Expenses today. It’s much easier than filling out an expense report used to be, but there’s still a lot of human interaction needed.

But what is the real value I'm offering that couldn't be done by the system itself?

My itinerary is on the smartphone in my pocket, and its GPS knows my location. It knows that when I walked into Haru, it was 7:00 p.m. local Boston time – dinnertime. It knows I took a picture on my phone about 45 minutes later and with image recognition software it knows it’s a picture of a receipt. There’s a high likelihood I just incurred a business expense. The software can enter a line item on my expense report: Haru, dinner, $34.02.

Furthermore, it knows I’m the CEO of the company, what codes to apply and where to put it on the chart of accounts, so the work is half done for the person in AP as well. Why should I be bothered to do anything other than say, "okay" at the very end? If we can get to that place, no interface is required other than the one in which I say "okay."


Smart integration

Looking at this from the perspective of an integrated platform, let’s say my stay at the Marriott is the twentieth expensed by a Coupa employee in three months, and we don’t have a contract with them. The system knows all that and it takes this expense report information and flips it over to procurement and sends a message: Hey, maybe it’s time to negotiate a contract with Marriott. The message contains all the historical information that a person in procurement would need to do that.

This is the best interface, because it presents you information only at the point of decision, where human interaction is needed. Behind the scenes, the machine does the thinking for you based on the rules you’ve given it. This is one of the next frontiers in enterprise software. 

The new grail

The enterprise software of the ‘90s was all about feature function - the more, the better. We've all discovered, both in business and in our personal lives that it's not about more features. It's about how easy it is to use the features.

We discovered that because we're getting bombarded with information and choices. The most precious commodity today is time, and the Holy Grail is not taking up any of it at all. So, while it’s important to be building a newer, better, cleaner, fresher, easier-to-use UI, making the UI all but disappear is the ultimate goal.

Instead of being slaves to our screens, we will become exception handlers. That frees us up to make higher order contributions. To be strategic, coming up with ideas and making decisions that a computer can't make. Solving difficult problems, or inventing new products and services, or creating designs that could only come from a human mind and hand.

Obviously this concept has a far broader appeal than the spend management space, but it’s what we’re striving for here in our corner of the world, because that’s the world we want to live in, and the one we want to create for our hundreds of thousands of users all over the world. Although once the user interface disappears, we may have to call them something else. Maybe just ‘people.’