Avoiding overlapping with randomizers and different text versions | Experience Community
Skip to main content

Avoiding overlapping with randomizers and different text versions

  • May 6, 2026
  • 1 reply
  • 33 views
R

Okay, so let me first explain my thesis set-up. I am researching CTA’s (call-to-action) in e-mail texts and want to find out if people can identify phishing from not-phishing by the CTA that is used. To do this I gathered 4 e-mail texts from a fraud database. These e-mail texts have been classified as phishing mails and each have different CTA. I also gathered 4 texts from my own inbox which are non-phishing mails, with a CTA. I have created a typology of four common used CTA’s in phishing and marketing. Then I manipulated the texts so every version has every CTA. This rounds up to 32 versions (8 texts x 4 CTA’s). I am going to test with a survey. I have created one with a randomizer of 4. Now I have a sticky situation. I want to make sure every respondent gets different texts and different CTA's. I want to avoid having to use multiple different surveys because I am messy and have little faith in me keeping track of everything. So how can I make sure every respondent gets 4 diferent texts with each a different CTA in my Qualtrics survey? Is it even possible at all?

 

PS.

I thought about maybe using 4 randomizers, 1 per CTA, but there is a risk that the same text will appear twice and it may cause carry-over effects or confusion. Also thought about embedded data but I am scared that the randomizer will maybe select the one that is then supposed to be skipped and then the respondent will only see 3 instead of 4 texts. 

 

Thank you so much! 

1 reply

Lpena
Qualtrics Employee
Forum|alt.badge.img+4
  • Lpena
  • Qualtrics Employee
  • May 13, 2026

Managing randomization for 4 different texts and 4 unique CTAs per respondent can definitely be done by leaning into the Survey Flow Randomizer and Embedded Data. I'd suggest creating 8 separate blocks—one for each of your phishing and non-phishing texts. Inside those blocks, you can house the different CTA versions and use display logic so only one shows up based on the respondent’s assigned data.

In your Survey Flow, you'll want a Randomizer set to pick exactly 4 of those 8 blocks. To make sure you don't repeat CTAs, you can use Embedded Data at the start to assign a unique ID to each CTA, ensuring they rotate through the chosen blocks without any overlap. This keeps things clean and avoids those carry-over effects you're worried about. 

If this setup starts feeling like a bug or isn't behaving quite right, you might want to head over to the Support team so they can take a closer look at your specific flow. You can get in touch with them through the https://support-portal.qualtrics.com/ by logging in and hitting the "Submit a Ticket" button. Good luck with the survey! 🙂

This reply contains AI usage for clarity and language purposes.
Terms & ConditionsAccessibility statement