Talking to strangers in the street: Recruiting by intercepting people


Intercepting is an exercise in self-awareness. Who you choose and how you approach them exposes who you are and what you think. What your fears are. The inner voice is loud. As a practice, we worry about bias in user research. Let me tell you, there’s nothing like doing intercepts for recruiting that exposes bias in the researcher.

Why would you do recruiting by intercepting, anyway? Because our participants were hard to find.

Hard-to-find participants walk among us

Typically, we focus recruiting on behaviors. Do these people watch movies? Clip coupons? Ride bicycles? Shop online? Take medicine?

The people we wanted to talk to do not take part in a desired behavior. They don’t vote.

We did intercepts because we couldn’t figure out a way to find the people we wanted through any conventional recruiting method. How do you recruit on a negative behavior? Or rather, how do you find people who aren’t doing something, especially something they are likely to think they should be doing — so they might lie about it?

They’re not going to answer an ad on Craigslist. It is unlikely that we could find them through panels or databases, as registration status and voting history are probably not data points collected.

When we talked to voter advocacy groups, of course they only knew about people who were registered to vote. We wanted people who were off the voting grid.

It’s not uncommon for retail companies to have mystery shoppers who chat up other people in their stores. The head of Intuit, Scott Cook, instituted follow-me-home studies long ago: find someone in a store who is buying your product, have a conversation, invite them into your study, and then see if they’ll let you follow them home and watch as they open, install, configure, and use your stuff for the first time. It’s what Cyd Harrell and Nate Bold call in-time recruiting. The user is in the moment, not just in the market. There’s no question about the motivation.

We wanted to learn about information challenges for people eligible but not voting. We were less about being at the right time and more about being in the right place. We needed to go where these people might be.


The art of the intercept

Much of the success of intercepting people to do interviews and testing on the street has to do with how you design your approach. There’s a lot of intention here.

What you look like and who you are

First, there’s what you look like. We did not want to look like the younglings who work for Greenpeace or World Wildlife Fund (you’ve seen them in their day-glo vests), and we were certainly not in the class of Latter Day Saints missionaries. We were researchers and we wanted people to get that intuitively. Also, as we were doing this work in cities, we did not want to be mistaken for beggars, homeless people, or hustlers.

A couple of things about us made that easy: We’re two middle-class, middle-aged white women wearing sensible shoes, carrying clipboards, with kindly expressions. One person asked me if I was a professor at the university we were in proximity of. Clothes are important. Props (like clipboards with signs on them and ID badges) are important. Smiling is important.

But also, be comfortable. You may be on your feet for hours. Out in weather. So plan for being warm enough, cool enough, sun-screened enough — keeping in mind that you have to carry everything you need for the day without looking like a bag lady.

Your first words to a prospect: What you are and aren’t

In my previous intercept experience, I started by saying, quickly, something like, “Can I ask you a couple of quick questions about X?” But after I watched a pro — a colleague who works for the League of Women Voters registering people to vote — I tried different things with different people. But mostly what worked was some variation of:

Hi. My name is Dana. I’m a researcher and I’d love to ask you a couple of questions about voting and elections.

It seems like it would take a long time to say all that. But there’s something about introducing yourself that gets people to stop and pay attention. However, we ran into plenty of skepticism. Especially on city streets in California, where there are a lot of petition drives. So the follow-up was often something like,

I’m not selling anything. I don’t work for any of the campaigns or parties. I don’t have a petition. I just have a couple of questions. Ok?

That’s usually all it takes.

Pro tips:

  • Know what you want to end up with out of your study and why. This makes it easier to flex with changing conditions and a widely varied available population.
  • Be ready with any appropriate, quick, and friendly question. You have to be “on” and energized without being freaky.
  • Use your own style.
  • Cast your question in a culturally acceptable way. (We had young helpers in a heavily Chinese neighborhood who spoke to prospects in Mandarin and addressed their elders as “auntie” or “uncle.”)
  • Be inquisitive rather than confrontational.
  • Work the researcher angle. At least, this worked for me. People want to help you, if they can.


Hit rates and time

Not everyone wants to help you, though, and some folks just won’t talk to strangers. And of course, hardly anyone qualifies for your study when you’re out in a random pool. For our most recent study, we conducted 55 20-minute interviews over 3 weeks in which we did 1 or 2 days of intercepts in each of 4 cities. There were days when we stood (or paced) for 3 or 4 hours and managed to do only a couple of interviews each.

Just for fun, we set goals for how many interviews we wanted to do in a day. Honestly, this made me a bit more assertive with approaching people, which was good. But we also had plenty of time to do the study, so if a spot on the street didn’t work well, we might move. Or we’d bail on the day and start fresh the next day. Most days, we made our quotas without stress.

Pro tips:

  • Give yourself plenty of time.
  • Consider setting goals for the day.
  • Don’t expect everyone you approach to say yes.


Location, location, location

One of the biggest challenges was choosing where to try to intercept people. We wanted people who didn’t vote or voted infrequently. Fortunately, they don’t hide under bridges. They’re everywhere. But we were pretty sure that there were places where prospective voters were more likely and less likely to be. Legend, experience, and data told us that we were looking for people of racial and ethnic minority groups. We also decided that we wanted people who were between age 18 and 60. It seemed unlikely that the outcomes of our project were going to change the habits of older people.

We looked for cities that had a high proportion of racial and ethnic mix, with ethnic neighborhoods. The next trick was to find the right place within the neighborhood to do our epic hanging around.

You want a place where there is plenty of foot traffic, but not where people are on their way to somewhere. Train stations — maybe. Subway stops, not so much. Our greatest success was at a community center one evening. There were a lot of people coming and going for events, but there were also people just hanging out, who had time to spend with us and were glad to do it. We had called a week or two ahead to ask if it would be permissible for us to be there, if we needed to rent a room, and whether there were any rules we needed to know about. When we arrived, the director was delighted to see us, made some introductions, and gave us a tour. It was a fantastic evening. (Ask us about “drag bingo” sometime.)

There were also a couple of places where we had handlers or intermediaries. In one neighborhood, a community activist took us around to some of the spaces where she regularly worked with people or groups. In one day, we met our guide at a diner, and then we moved on to a community college, next we went to a multi-use complex that had a farmers’ market going next to a vocational school and a senior center, then a library (which was closed), and on to a neighborhood strip mall. That was a peak day for the number of interviews we did, but moving locations took time and energy.

We had general success at or near libraries. The large city library where we met our three guys at the top of this post had a lot of foot traffic from the neighboring university as well as other people from the community. However, there were a lot of homeless people loitering. Some were qualified for our study and some were not. Unfortunately, they tended to be less grounded in reality and less likely to vote no matter what the outcome of our study was.

Pro tips:

  • Cafes and restaurants may not be welcoming, or they might, especially if you offer to buy their wares for you and all your participants (and maybe offer a bit more). Or you know the proprietor.
  • Rooms in libraries are often far away from the foot traffic where you might do the intercepting.
  • What you are doing may be seen as soliciting, which is illegal in many situations, and uncool in others. For example, we thought we might try out a laundromat in a city that had a homeless problem. Not an option for us, either.
  • Colleges may want you to go through their IRB, unless you or your contact knows the provost or president who can approve of your being there.
  • Community centers can be safe, fun, target-rich environments with a little advance planning.


Pair up for safety

This gets us to safety and awareness of your surroundings. How were 2 white, middle class, middle-aged, women standing on street corners in ethnically diverse neighborhoods going to blend in? This was less a concern about our research and more a concern about possibly becoming targets for bad people. We did all of our street intercepts during daylight hours. We conducted separate interviews, but we were usually within viewing range of each other, and when we weren’t, we either checked in first or SMS’d new locations. (I would never try this kind of field work without a cell phone handy.) We did interviews in the evening when we could be indoors at a place where the clientele was generally known to our contacts.

Pro tips:

  • Work in pairs.
  • When working outside, work during daylight hours, unless you know the neighborhood to be safe.
  • If you decide to separate, agree on a meeting time and place to check in.
  • Have a plan for emergencies.


About eye contact

There are those people out there who would be perfect for your study, but when you see them, they actively avoid you. They turn away, cross the street, or just look down. It is fascinating to be deliberately ignored. When it seemed like that was happening a lot, though, I took at as a signal that I needed to check myself. What was my posture like? How was I holding myself? Was the sign for my study showing and readable? Was I showing any signal that I should not be approached? Often I couldn’t think of anything to change — but the awareness brought me back to my intention, and I’d snag somebody soon after, no longer unapproachable.



Talking to strangers

Recruiting well for user research is one of the greatest challenges of day-to-day work in experience design. Focusing on behavior makes it easier. But what how do you find people who don’t do a thing? How do you find people who you want to motivate toward a behavior but they’re not going there? How do you find people who are hard to find through the usual channels? We decided to go where our people were: streets, libraries, community centers, churches, malls. Sure enough, there they were. All we had to do was talk to enough strangers.

2 thoughts on “Talking to strangers in the street: Recruiting by intercepting people”

  1. Wonderful post. Thanks for sharing with concrete examples grounded in actual things you’ve tried.

    The "eye contact" paragraph was a bit confusing: do I need to check myself to look more like a researcher looking for people to talk to, or less? I’ve found standing more upright and smiling more or looking more researchy didn’t necessarily create a more welcoming or approachable environment.

    1. Ah, yes, I see what you mean. At least in the U.S., you should try to make eye contact, as the researcher. But if someone avoids your eye contact, they’re trying to escape. Now you have to decide whether to pursue, and if so, how.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.