One of the first things I learned how to when I started my position at IBM was how to create a screener. Due to design being a newer area in my business unit, it was hard to get access to current product. I had to be creative and go outside to recruit.

Define the Problem and Set Your Target

First, you have to make sure you’ve defined the problem and understand it. Based off the problem, you can figure out who you need to target. Who is the person experiencing the problem and working in it?

Research, research, research

Before you can even start writing a screener, you have to do a lot of research. Why? Because a screener is intended to weed people out to make sure you’ve hit your target. In order to do that, you need to have a baseline understanding of the job description of the person you’re trying to recruit.

My first project was persona validation for a database administrator. I was trying to recruit database administrators so that I could learn more about a database administrator! It was extremely challenging to write the screener.

I read a lot of articles and a few book chapters about database administration. Then I took my first pass!

Formulate the screener

I looked at a lot of examples, but roughly landed on this as my framework:

  • Industry (open ended)
  • Company size (single choice)
  • Employment status (single choice)
  • Job title (single choice)
  • Experience in field (single choice)
  • Job level (single choice)
  • Tasks (open ended)
  • Tools used (open ended)
  • Regular job functions performed (multi-punch)
  • Challenge (open ended)
  • Tools (multi-punch)

I vary the questions by open ended, single choice, and multi-punch. There are answers they HAVE to select, may select, and some I review.

In the questions that are multiple choice, I always include some fake answers or answers that look real to trip people up. I ask the open ended questions to both eliminate those who’ve made it past my trick questions and to make sure I’m targeting what I want.


No matter how hard you try, the screener isn’t going to be perfect! So I get other researchers to peer review them for me. I also have just learned through trial and error of launching screeners with questions that could be formulated better.


After doing all this work, I launch the screener and recruit participants!