Questions, answered.
- To passYour build passes at 23 of 25, with at least 3 of 5 in every area. Your interview passes when your answers average 4.5 of 5 (22.5 of 25 on your result), at least 4 in 5 of them score 4 or 5, no more than one misses the question outright, and every answer is your own and matches your build.
- AttemptsUp to 5 attempts per cert. Attempts that end in our error don’t count.
- CreditsYour build runs in your own Clay workspace and uses your credits.
Before you start
Yes. Use AI tools and templates as you like. In the interview, tell us what you built, what you used and why.
Yes. At the end of the brief you can bring a real project from work or swap in a job listing. We turn it into a short scenario like the practice one, and your build is scored the same way.
The build
No. We grade the finished build, not your credit history.
We check that excluded rows can’t reach a paid column. A row that turns bad after a paid step is normal: count it in your cost per qualified account.
No. Run IDs only matter if you build a Clay workflow. For a table, your table link and your test rows are enough.
Paid Ads needs a platform that tells you how many of the people you uploaded it matched. LinkedIn and Google Customer Match both show this on the audience once it finishes processing.
If your platform only shows an estimated audience size, pick another platform for this project.
The interview
Your interview opens once your build passes: 23 of 25, with at least 3 in every area and its required checks met. To get certified, your interview has to pass too.
If your build is close, revise it with your feedback and submit it again. To pass, a build and the interview you take on it both need to meet the full passing standard.
That interview didn’t count because of a problem on our side. Your redo is free. Start it again from your project page when you’re ready.
Scores and attempts
Your build passes at 23 of 25, with at least 3 of 5 in every area. Your interview passes when your answers average 4.5 of 5 (22.5 of 25 on your result), at least 4 in 5 of them score 4 or 5, no more than one misses the question outright, and every answer is your own and matches your build.
Your brief is a short scenario, with no checklist. The build is scored on how well it solves that scenario, shown in your table and test rows, and the interview on how you explain and defend the choices you made.
Started on an older brief with must-haves? Show each one in your table or test rows.
Up to 5 attempts per cert. Attempts that end in our error don’t count.
Your best build and the interview you took on that same build. An interview counts for the build it was about, so a new build needs its own interview.
Once a build and its interview both pass, a later, lower score never takes that away.
On an older brief with must-haves, they’re read on the build that counts, so show them in that build.
Re-sharing a link we couldn’t open. We ask you to share it again, and the build is graded once it opens.
An interview redo when the interview broke on our side or a question you needed never came up.
Yes. No one signs off a pass, so the score you see is your result.
Two exceptions. If something broke on our side (we couldn’t read your build, or the interview glitched or skipped a question), you get a free redo. If you think the grader missed something your build shows, ask for a second read.
Your certification
Once your build and its interview both pass. A person on the Clay team confirms every result before it’s final. Your certification shows on your project page once yours is confirmed, with a public page anyone can check.
What changed
- All three practice briefs say what Vantis is: a fictional security pentesting consultancy.
- This FAQ no longer gives a date for when certifications release.
- Up to 5 attempts per cert. Attempts that end in our error don’t count.
- You can type your interview instead of speaking it.
- Common questions you ask from your project page get an answer right away.
- Your interview opens once your build passes, at 23 of 25.