What does a test cost?

The shape of the number before you open the wizard: what drives the price of a test, what common sizes run, and why the quote you approve is the price you pay.

Updated

Every test shows its exact price in the wizard before you launch, so you never commit blind. But when you are budgeting a study, you want the shape of the number before you get there. Here it is.

How credits work

One credit is one dollar, always. You can buy credits in packs (100, 250, 500, 1000, or 2500 at a time, with bonus credits on the three larger packs), or add exactly what a single launch needs in one click. Unused credits stay in your wallet for the next test. The full mechanics are in Credits and packs.

The four things that move the price

  1. How many people watch. Every participant is a real person paid for their time, so the sample size is the biggest lever. Double the people, roughly double the cost.
  2. Who they are. The general population is the base rate. A shaped audience (a specific age band or gender) draws from a smaller pool, and reaching those people costs more per person. See Choosing an audience.
  3. How long the piece runs. Participants are paid for their time, so a longer piece means a longer session and a higher per-person rate. A ninety-second version and a six-minute one are different budgets.
  4. How long the survey is. Up to 4 questions ride inside the base estimate. Past that, each stretch of extra questions adds session time, and the pay follows.

What a test runs, roughly

For a short video, a general audience, and a short survey, a recruited test works out to about 4 credits per person. That gives you honest budgeting anchors:

  • 24 people is around 100 credits.
  • 48 people is around 200 credits.
  • 200 people is around 800 credits.

These are approximations, and the drivers above move them. The wizard's quote is the exact price: it reflects your video, your audience, and your survey, and nothing is charged until you approve it. A paid test also has a floor of 25 credits, because below that a sample is too small to tell you anything.

The test cost quote in the wizard showing the credit total before launch
Screenshot: testing-credits-1
The wizard quotes the exact credit cost before you launch. The number you approve is the number you spend.
The wizard quotes the exact credit cost before you launch. The number you approve is the number you spend.

What happens at settle

A test reserves credits for the full sample size. When it settles, only completed responses are charged. If someone never finished the video, or if they failed a screener question, that response doesn't count in your result and isn't charged. Unfilled slots also return their credits automatically.

Voided responses are held for a human review on our side. We never auto-reject a participant. If you think a response should not have counted, write in and we will look at it.

The minimum charge across all completed responses is 25 credits because below that a sample is too small to tell you anything.

All of the above is the price of HoldFast recruiting an audience for you. If you bring your own viewers instead, sending the test out with a share link to your list or community, that costs zero credits. You pay for recruiting, not for testing.

Questions

Why is my quote higher than the worked examples here?

The examples assume a short video, a general audience, and a short survey. A longer video, a narrower audience, or a survey past the included question count all raise the per-person cost, and the quote reflects your exact setup. The quote is the real price; the examples are for budgeting.

Can I flag a bad response and get it refunded?

Today there is no flag button. A response that did not finish the video or failed a screener is never charged. Voided responses are reviewed by a person on our side, and if you think a response should not have counted, write in and we will look at it.

Ready to try it?

Host video, collect frame-accurate review, and see who's watching. Free to start.

Start free