No Such Thing as a Random Sample of Five People

Written by Sean Murphy. Posted in Customer Development, skmurphy

Q: I have 3D printed a couple prototypes of my product. I am going to get user feedback by letting security guards–my target market–test if for free for a few days. How many prospect should I have test it before I can determine if there is a market for the product.

Since you are 3D printing and can iterate I would start by getting feedback from five folks: see Why You Only Need to Test with 5 Users by Don Norman for some research on why this is a reasonable place to start.

You will more than likely end up talking to more folks before deciding if there is a market or not but working with a small group at one time and testing iteratively will be easier to manage and teach you things more rapidly.

There is No Such Thing as a Random Sample of Five People

Be aware that there is no such thing as a random sample of five people and that there will be unconscious biases in your selection process that you will need to adjust for (e.g. availability, channels or methods used to solicit, reasons for responding, geography location, etc..) , for that you need to get to larger numbers depending upon what your hypothesis is for the level of interest.

There are unconscious assumptions built into Norman’s suggestion for five, they are not “random people.” They have to be willing to test your product, have the ability to understand the need or problem it’s designed for, have the ability to use the product, etc..

Trackback from your site.

Leave a comment

Quick Links

Bootstrappers Breakfast Link Startup Stages Clients In the News Upcoming Events Office Hours Button Newsletter SignUp