Nonprofit User Interviews… on a Shoestring Budget

If there’s one constant in the nonprofit world, it’s that we seldom have as much money or time as we’d like for technology projects. But you don’t need a lot of either to gather some very useful information about your potential users — information that will make a big difference in the quality of what you build. This research can also save in the long run — if your stakeholders have trouble agreeing, research can cut through differences of opinion with actual facts.

User interviews are one of my favorite research techniques when money is tight. It’s amazing how much you learn from talking to four or five actual people. These interviews often insert a very needed “outside perspective” into a project, with perhaps a total of eight to ten hours of work. That’s almost always worth it, in my book.

Getting Started

With a little bit of prep, you’ll be off and running.

1. Write up your overall research goals. It’s useful to define what you generally want to know prior to deciding what exactly you want to ask… as those are often separate things.

2. Create a list of questions to ask your interviewees. In general, it’s better to stay away from people asking directly for their opinions about your organization or project. Instead, understand how they currently do things, what’s important, what sites they like and why, what information they look for, etc. People are typically happy to share their opinions (often, at great length), but those opinions may or may reflect what they actually do in real life.

3. Recruit people to interview. Choose people outside your organization, but who ideally have some kind of connection to your mission of the goals of your project. It can often be useful to talk to some people who are familiar with your organization and some people who aren’t. Remote interviews work very well; with the boom in video conferencing, there’s not even much benefit to in-person interviews. 

4. Schedule interviews. I generally do 30- or 45-minute video conference interviews (phone also works). Don’t underestimate what a pain it is to sync schedules to find time times; you might want to draft a junior person to help do the actual calendaring.

Conducting Interviews

People are typically happy to talk, making conversations relatively easy. 

1. Ask your questions approximately in order. You want to ask generally the same questions to each person, so you can compare your data across, but there’s no need to be a robot. Go ahead and paraphrase your questions or follow up on interesting things they say.

2. Take notes. I typically copy the list of questions over for each person, and then use that as a template to add notes during the interview. If you don’t take notes, it’s very easy to take notice when people say what you want to hear, and pay less attention otherwise. 

3. Wind up interviews by seeing if they had more to say. I often finish by asking “So based on what we talked about, was there anything else I should have asked?” Most people will say no, but I’ve had people— or whole sets of interviews— where it came out at the end that they had additional really important things to say.

Drawing Conclusions

You probably now have about ten pages of notes. That’s not that many; you don’t need to rely on fancy research techniques to analyze them. 

1. Analyze commonalities. Read back through your notes. What was important, or unimportant to a number of people? What conclusions can you draw? 

2. Present back to stakeholders. You don’t need a formal report — likely just a document or a short set of slides in the simple format of  “Three of the five people we interviewed felt that…”

3. Apply your learnings! What does the information mean for your project? In your presentation, suggest some things that you should make sure to do or change based on this new knowledge.

In This Really Research?

A lot of people ask me whether talking to four or five people really gives you enough information to make decisions. Yes and no. A few thoughts on that:

• Talking to four or five people is in fact a reasonable research sample if the interviewees have similar things to say.  If three to five of your five interviews all had the same perspective, that’s pretty solid data that you can use to make decisions.

• However, perspectives held by one or two people in that group aren’t as useful, and you shouldn’t put much stock in them. It’s difficult to know if those people had a strange outlier perspective or if more people would share it if you had spoken to more people.

• If pretty much everyone says different things, and you can’t find commonalities, that’s a sure sign that you haven’t spoken to enough people. Or, perhaps, that the group of people you spoke to was too different from each other. Could you speak to only people who will use the project in one particular way, and gain a deeper understanding of those people?  Or add more interviews?

• As a rule of thumb, you can know that you’ve spoken to enough people when interviews start to become boring—when you’re hearing the same things over and over. That’s not likely to start happening at only four or five, but you might start seeing the signs of that… and that’s a good thing. 

In general, pretty much any user research is better than none. Don’t be afraid to take on a straightforward set of research like this. Your users will thank you in the end!