Lessons In Designing Great Enterprise Software

Editor’s note: Etan Lightstone is the director of UX design at New Relic.

As people are bringing new expectations of a quality user experience into the workplace, delivering a consumer-grade experience is becoming critical for all business software and is a fundamental way for software makers to compete in the market. While most software makers are increasingly aware of this truth, the challenge today is how to effectively make the transition to building consumer quality software for the enterprise.

A designer without an engineer is an art gallery, and an engineer without a designer is a parking lot 

Balancing designer and engineer perspectives and even designer and product manager perspectives is critical. Instead of having designers slap on a design at the beginning or end of a project, they need to work closely with engineers along the way to properly align design and engineering capabilities.

Lead designers sharing the product definition role with product managers creates a healthy tension and helps to provide checks and balances as multiple people define a project from two different but equally essential lenses:

  1. Designers: the design perspective, focused on usability, workflow, and evoking a good emotional reaction from customers

2.Product managers: the business perspective, focused on increasing profit and market share for the company

We found that the hub-and-spoke model for organizing product design with the rest of the engineering team works well. We deal with problems and do reviews as a single design team: That’s the hub. But we also have spokes: lead designers on every project, completely embedded and focused as members of the project team. Having an embedded designer ensures we always have a representative that’s a single point of contact and in-touch with the day-to-day issues the engineering team deals with in its development life cycle.

Spark an emotional connection

It’s critical to perform user research, utilize the right design patterns, and iteratively test and apply solid interaction design to craft a product experience that best supports customers’ goals and workflow. Make it easy to use so people don’t make mistakes that lead to frustration. We base our product-flow decisions on research that informs our understanding of customers’ goals so we can focus on solving the right problems.

This leads to great functional product designs, but to really develop trust and loyalty, it’s imperative to establish an emotional connection with customers. To delight people with a clever animation in a chart, a helpful indicator while the chart loads, a smart gesture for how a menu snaps back into place. This kind of design helps people make an emotional connection. They may not notice it consciously, but subconsciously it can help build trust and increase the product’s addictiveness.

We see this kind of emotional design in many consumer products, such as the pleasing drag and snap animation on iPhones, the chime that your device sings when you turn it on, or the pulsating animation when I sync my Misfit Shine wearable with the mobile app.

Before beta testing

With a data- and information-heavy UI, you need to test the software with people seeing their own data and information. Properly evaluating our UI/design process requires experimenting and prototyping with real data going through it, before it goes through the beta process of fixing bugs and making slight polishes to the experience.

So we make a point of bringing in customers to test products with their own data and information earlier in the design process. Waiting to do this during beta testing is risky, because you are usually too far along in development to react to big issues in your concept without incurring a huge cost of starting over.

In each development sprint, we usually have some customers on call as a sounding board to review the product and help assess usability. Early in the process, even before the product is fully defined, we try different prototypes with them to identify which ones they react to better, that help them solve the problems that are important to them.

Most software makers try to do this, of course, but it can be extra challenging for enterprise products dealing with data and an infinite number of ways to represent it. The best way to anticipate the questions customers will ask (even if they don’t yet know what questions they want to ask) is to get real-time feedback along the way. There’s a big difference between asking a customer what they want vs. watching them actually use something. You need to do both.

Balance feedback and stay focused

Perhaps the biggest challenge is making the UI simple enough for startups where people wear many hats and powerful enough for enterprise customers who demand more control and flexibility. To maintain that balance, you have to carefully weigh the feedback you get and sometimes eschew seemingly valid suggestions.

For example, when we first built New Relic Servers a few years ago, we began with a very minimalist offering to make sure we didn’t overload people with information. We provided several ways for customers to see what their servers were doing. But one customer wanted something very different: a grid of 30 charts with every metric you could think of at the home-screen level.

For customers like this one, we made it easier to drill down when they need to look deeper. But we didn’t include everything on the first screen, because we didn’t want to overwhelm the majority of our customers with an undifferentiated view of every possible metric. That’s not bringing the right value to the table.

Following simple best practices of optimal design team structure, early alignment and ongoing real-time collaboration with customers, creating user experiences that surprise and delight, and knowing what feedback to incorporate into design will help businesses build software that will become a competitive advantage in an increasingly software-driven world.