Finding Our Way: Four Lessons About Enterprise Research

We (Omar Vasnaik and Carolyn Wei) joined Facebook’s Ads and Business Platform team as UX researchers a couple of years ago. Our backgrounds were different: one of us came from social computing, while the other had worked on enterprise research at another tech company. But in our first weeks and months doing enterprise research, we both felt out of our depth at times. We quickly discovered that running ads research was a totally different ballgame than the one we’d been playing. After a few instances of having no idea what people were saying or doing, we even had flashbacks of overseas fieldwork — without leaving Seattle.

Over the last couple of years of doing research with advertisers and developers, we’ve worked on cool problems, made plenty of mistakes, and climbed our way up the learning curve. In short, we’ve come a long way.

Spurred in part by a previous Elegant Tools post about building domain expertise in a complex industry, we’ve been talking to each other lately about what we’d tell our slightly younger selves if we could go back in time to those first weeks and months doing enterprise research. We came up with a “care package” of the insights that would have helped us the most. We hope they’re useful not only to new enterprise researchers, but also to anyone finding their way in an unfamiliar domain.

1. Don’t treat enterprise research just like consumer research.

Research is research, right? When we started working with advertisers, we assumed we could run research studies just as we did with consumer products: recruit, ask questions, give a gratuity, and send participants on their way.

Although it does center on people’s needs and behaviors, enterprise research requires a slightly more nuanced approach. Consumer research often involves the discretionary use of a free or accessible product like a search engine or shopping site. Enterprise clients, who may be spending a lot of money on a product or service that’s critical to their business, tend to bring higher expectations. And in most cases, your research is part of an ongoing working relationship. A few simple practices can help you get the information you need while protecting the health of that relationship:

Work with sales partners to find clients.

Recruiting enterprise customers for research isn’t easy. You can’t intercept a developer or advertiser on the street for a quick research session. We recommend working with your sales teams to identify clients who suit your needs — and who want to talk. Such an introduction can give you a head start on building rapport, as well as background knowledge to help you prepare for the research session.

Make research sessions stand out.

Our clients regularly work with account managers and others at Facebook; if we’re not careful, our interview research sessions can feel like yet another business meeting to them. Try disarming participants with creative research techniques. For example, we once asked participants to pin cutout logos of competing companies and products against certain attributes. We’ve also asked clients to imagine the product as a guest at a party and to then describe all their impressions of that person. Techniques like these can break the ice and make the person you interview think about the problem in a different way, which can yield much more enlightening results.

Close the loop with clients.

Like it or not, you’re part of the customer service experience. Clients may raise questions or tech support problems or ideas during the research. Be sure to follow up and connect them with the right folks. We also recommend giving clients some kind of high-level summary of the research findings so that they know they’ve been heard. Partner with the sales team so they can keep clients posted on how the product feature you were testing is progressing toward launch.

2. Build your confidence.

It’s easy to feel intimidated when you’re about to interview a Fortune 500 client, or someone with decades of experience in the field you’re researching. But letting fear drive your approach will limit your effectiveness. Here are a couple of reliable ways to build confidence:

Start small.

Plan your research in such a way that you talk to smaller companies first. Say you want to research the buying lifecycle in an e-commerce company. Try to find small developers who’ve built an e-commerce app, such as a one-person shop or even a student in a dorm room. An informal call with such developers will be much more comfortable than a corporate conference call with a half-dozen people listening in. More importantly, you’ll build your domain knowledge before you talk to more intimidating clients.

Prepare to be lost.

When you run an interview in a complex or technical domain, no matter how exhaustively you’ve prepared, there’s always a chance that you won’t understand what the participant is talking about. When that happens — and it will — be upfront that you’re new to the space, and that you might be asking naive questions. Don’t be afraid to ask participants to repeat themselves or explain things in a simpler way. You can also repeat what you’ve heard or ask the participant to verify implicit points you thought were being made. You’ll be surprised by how much more information you can glean this way. People tend to open up when they see that you’re taking the time to truly understand what they’re saying.

3. Learn the domain in increments.

Getting started doing research in a new domain certainly requires gaining background knowledge. You might feel like you have to burn the midnight oil to catch up with colleagues who’ve been working in the domain much longer. Some background work is essential, but we think it’s wiser to focus on learning the domain as you go. A couple of techniques can help accelerate that process:

Focus studies on one specific customer problem.

When you don’t know the domain well, keep your first research studies narrow in scope. For example, you might plan a study around how ad traffickers create an ad placement, rather than the workflow for the entire marketing team. This will keep your conversations corralled, so that at the end of the study, you’ll have rock-solid knowledge of a specific area. You can then focus a series of studies on the same general topic so that by the end of a quarter, you’re confident about a broader area.

Prioritize your learning.

Learn the parts of the domain or product that are most crucial to your immediate needs. While researching Atlas, a complicated tool for ad agencies, we prioritized learning about the behaviors and needs of the workers who use the tool, rather than about the specific features of Atlas, because we already had so many resident experts on the product. This approach laid the groundwork for more fruitful conversations with ad agency workers. When we didn’t understand a specific workflow they were showing us in the tool, we followed up later with Atlas experts.

4. Make the most of in-house knowledge.

Starting in a complex domain can feel lonely, but a lot of resources are probably available within your company. It helps to know where to look, as well as how to use what you find. We suggest starting with the following:

Use inbound feedback channels.

Sales and support channels within your company likely have a wealth of information — often a mix of problems and requested features. A good way to start building domain expertise is to comb through these channels for useful information.

Talk to the right people.

A lot of questions we had about advertisers and developers weren’t easily answered through external reading, or didn’t seem worth wasting client time on. But most sales and field teams are rich with industry experience and happy to help educate a colleague. Our advertising sales teams often have extensive agency and marketing experience, and because they are internal colleagues, we can ask them very frank questions.

Get tight with your engineer.

Sit down with your engineer and walk through the workflow or even the code. Your engineer can often provide parts of the story that you’ve been missing; together, you can gain a more complete understanding and develop a mutually beneficial partnership. For example, when we were researching how advertisers use deep linking, we had the engineer explain the relevant code. By understanding the nuts and bolts, we better understood advertisers’ challenges with deep linking. And the data gleaned from advertisers was new to the engineer, even though he’d been working in the space for a while.

To our slightly younger selves: that’s about all the advice we have for you! Good luck. We hope it helps you learn faster, work smarter, and get comfortable in your new domain. If it doesn’t, contact us for some can’t-miss betting tips involving future sporting events and political nominations.


Have you been working in a new, complicated domain? We’d love to hear your stories of how you made your way through the space. How has your approach evolved? What advice would you share with newbies?


Omar Vasnaik researches how businesses promote their mobile apps through advertisements on the Facebook mobile newsfeed. He previously worked on a range of enterprise software at Microsoft and CA.

Carolyn Wei works on advertising measurement tools, including Atlas, that help clients better understand the effectiveness of their marketing efforts. She previously researched social communication tools at Google and Microsoft.