Ordering study participants - best practices
What is a drop-rate, how many testers you should order and what to require from the panel provider
When you are using your own panel (order own participants) there are a few things to keep in mind.
What to require from a panel provider?
The panel of your choice should, of course, provide you desired target, but as just important is that participants will do the eye-tracking test. Many of them don't want to enable webcam access or don't follow the instructions properly, so they won't complete the task.
First of all, make sure to target participants by those who are willing to use the webcam. They also need to complete the study on a desktop device only. Next, the test should be done on the Chrome web browser or the latest MS Edge web browser (based on chromium). Attending the test from Firefox may cause technical issues and other browsers won't work at all.
These are crucial requirements that must be met. You are welcome to use our recommended instructions while ordering participants (please find them at the bottom of this article).
How many testers won't participate in the study? What is an "offset"?
There will always be some participants that will return the study or that something won't work for them. This is why it's so important to make them follow the instructions to reduce such cases. It shouldn't be more than 50% of the ordered number of testers.
An offset is the number of results that are low or very low in terms of quality. This is usually between 15 and 20% of all obtained results.
For studies with a greater number of participants the risk of encountering such cases is reduced (in percentage) - i.e. finding 5 problematic participants won't affect study for 100 participants (5%), but it's a lot in the case of study for 10 participants (50%!).
The number of participants that don't want or can't do the task also depends also on the study type - it's higher for live website testing studies.
There are two ways to approach ordering participants on your panel:
- If you'd like to have 100 usable results, order between 100 and 120 participants (we don't count here participants that can't or don't want to participate) and then order additional ones if needed (if there isn't enough good quality results).
- Order between 150 and 200 participants at once to be sure you meet the required minimum, without adding more later.
What to do if you think you have too many low quality results?
If you find too many participants with low or very low quality results, so you can't obtain any usable results for your study, please contact us. We'll be happy to talk with you, analyse the possible causes and, if needed, increase your participants limit (so you won't have to pay extra).
Consider using RealEye participants
If you don't want to struggle with ordering participants, you may want to consider hiring RealEye participants. This costs $5 per participant (no matter the study time or your plan), but we will make sure, that you receive only good quality results.
- we provide you with as many results, as you ordered (we'll take care of the drop-rate and participants with technical issues),
- we approve only usable results (good quality),
- you'll receive results within 1 business day.
- we can't provide very specific targets (ethnicity, age, etc. is ok, but please contact us before ordering participants, we'll check if we can provide anything more specific),
- you can't use external tool integration with RealEye participants - so you won't be able to redirect participants to an external survey after the eye-tracking part,
- you can't test live websites.
Instructions which we're using to hire participants:
In this study we'll ask you to participate in eye-tracking session using your webcam. Please:
- use a desktop device like laptop/PC/Mac(book),
- use Chrome web browser,
- make sure you have a good quality, working webcam plugged in,
- make a webcam very steady and below your eyes level,
- make sure you're in good lighting conditions - the light source is in front of you and illuminates your face evenly,
- don't rush through the calibration process,
- if you wear glasses, make sure there are no reflections on them,
- try to keep your head still during the whole test,
- if you're using a laptop, make sure it's plugged in.
- (for website testing) please, share your screen or browser window (NOT a tab) when asked.
We value your privacy. We don't store any image or sound. We use your camera only to calculate eyes position.
See also: How to get accurate results