BidTune
USAGE-BASEDMonitoring, experiments, the holdback comparison, and the experiment ledger. Every non-empty render counts. Each month the bill is capped at the extra revenue measured against the 10% holdback.
Fourteen days free after activation. Then $0.03 per 1,000 measured impressions.
Never pay more than BidTune earned you. No proven lift, no charge.
Monitoring, experiments, the holdback comparison, and the experiment ledger. Every non-empty render counts. Each month the bill is capped at the extra revenue measured against the 10% holdback.
Your trial starts when your site is activated and collecting data. Setup review comes first. No credit card required to get started.
First findings land in about 2 days. Each test needs at least 72 hours and 2,000 sessions in each arm. A site-wide test needs about 7,408 sessions over 14 days (about 530 a day). The smaller arm is 30% of the traffic outside the 10% holdback.
Estimate your monthly costThe rate stays $0.03 per 1,000 measured impressions. All non-empty renders count, including refresh and other demand. A better Prebid bid can raise prices across the whole auction, so the lift is not limited to Prebid wins.
You never pay more than the extra revenue BidTune measurably earned you versus the 10% holdback. No proven lift means no charge. The amount due is the smaller of the usage fee and that measured lift.
The holdback keeps your original setup. We compare revenue per session against it, on complete days, and use a conservative estimate, not the headline number. Conservative means the low end of that comparison: the 2.5th percentile of 1,000 resamples of those days, not the average. If that low end is not above zero, the charge is $0. With fewer than seven comparable days, or no rendered revenue on the holdback, there is not yet a measurement, so the cap does not apply and the usage fee is charged.
These examples use the default share, 100% of the measured lift. A lower share only lowers the cap.
10,000,000 impressions ÷ 1,000 × $0.03 = $300 usage fee.
Measured lift versus the holdback: $800.
Amount due = $300. You pay the usage fee.
50,000,000 impressions ÷ 1,000 × $0.03 = $1,500 usage fee.
Measured lift versus the holdback: $400.
Amount due = $400. You pay only the lift.
20,000,000 impressions ÷ 1,000 × $0.03 = $600 usage fee.
Measured lift versus the holdback: not above zero.
Amount due = $0. You pay $0.
The holdback is what the lift is measured against. Turning it off, or setting it to 0% or 100%, leaves nothing to compare, so the cap does not apply and the amount due is the usage fee. Settings asks you to confirm before the holdback is turned off.
The Billing page lists three figures for the open month: usage fee, measured lift, and amount due. The invoice, sent after the UTC month closes, is the amount due. Until then, daily rows are the list price.
Estimate the usage fee from measured impressions. The rate is $0.03 per 1,000 measured impressions. Your other partners bill on their own terms.
The calculator shows the usage fee for a normal paid month after your trial. The amount due is capped at measured lift. It does not model the introductory free period or forecast revenue.
Amount due is capped at measured lift. If the lift is smaller than this fee, you pay the lift. No proven lift, no charge.
Each non-empty ad render observed by the integration adds an impression, including refresh and other demand. Each slot and each refresh can add another. A better Prebid bid can raise prices across the whole auction, so the lift is not limited to Prebid wins. Pageviews and sessions are different units.
Observed renders across the site, including other demand where captured
Optimized and holdback traffic
Each non-empty refresh
Pageviews or sessions themselves
Bid requests and bid responses
Empty ad-slot render events
Usage includes observed demand beyond Prebid, while the optimization readout evaluates Prebid revenue. Legacy days without render events use recorded Prebid wins.
14 days free after activation. No credit card to get started.