Most operators hold two conversion numbers that do not agree: the one the platform reports and the one the backend confirms. The usual assumption is that one of them is wrong. Usually neither is — they are counting different populations, because click attribution and view attribution see different people. Across six independent measurements of ThumbAd campaigns, between 89% and 96.2% of confirmed conversions arrived on paths a click-based system never sees. Put your two numbers in below and the gap is reconciled line by line, with the arithmetic shown rather than asserted. It runs entirely in your browser: no email, no account, nothing sent anywhere. It can also tell you your current measurement is fine.