Ad verification, in plain English
Findoutwhatyouradmoneybought.
You buy 10,000 ads. Your report accounts for 8,460. The rest is written off as a “discrepancy” — every month, at roughly the same size.
The problem
Where the money goes missing
Two systems give you two numbers. Both sound certain. The difference gets written off as a "discrepancy" and everyone moves on. It is not a rounding error — it is a chunk of your budget, and it is the same size every month.
How it works
Every ad. One reason. No mystery.
We start at the number on your invoice and follow every single ad. Each one shows up exactly once, at the point it stopped, with one plain reason why. Add the rows back up and you land on your invoice again — which is how you know nothing quietly went missing on the way.
Why trust it
Why you can trust the number
A number is only worth as much as the method behind it. Ours is deliberately boring, and deliberately easy to check.
The numbers always add up
Start at the number on your invoice. Every ad appears once, with one reason. Add them back up and you land on your invoice again. We check that on every run, and code that breaks it does not ship.
One script for everyone
The same file for every partner, every time. Nothing custom to build, nothing to maintain, and no chance of one partner quietly running a version that reports different numbers.
We keep the working out
We store how long an ad was actually on screen, not just a yes or no. When the industry moves the goalposts, we re-answer for data we already have instead of asking you to start again.
A reason, not just a number
"Nobody scrolled far enough" and "your script never loaded" are different problems for different people. We tell you which one, and what it cost.
Questions
Questions people ask us
We take every ad you paid for and tell you what happened to it. Did it reach the website? Did it load? Could anyone see it? Was it a real person? You get one clear reason per ad, and the cost attached to each reason.
Because they count different things. Your buying platform counts ads it paid for. Your measurement tool counts ads its code managed to run in someone’s browser. Everything that goes wrong in between falls down the gap, and most tools just label the difference "discrepancy" and move on. We name each cause and price it.
A viewability tool grades the ads it managed to measure. It cannot tell you about the ads it never saw — and those are usually where the money goes. We start from your invoice, not from what happened to be measurable.
They send theirs instead and we match it back through the logs. We label those matches differently from exact ones. We will never quietly mix a guess into a number you are about to show a client.
No. The script is 2.1KB — smaller than a small photo — and loads in the background. Our server replies in about two milliseconds and writes to disk first, so even if our database goes down, nothing is lost.
One campaign, one script, one week. It is the same script for every partner, so there is nothing to build per partner and nothing to maintain once it is live.
A single chart that starts at the number of ads you were billed for and ends at the number that genuinely worked, with every drop-off in between labelled and costed. If the two ends do not add up, that is our bug, not your problem — and we check it on every single run.
Get started
Try it on one campaign
Put our script on one campaign for a week. You will get a single chart that starts at what you paid for and ends at what actually worked, with every drop-off named and costed in between.
Or write to hello@pixtru.com