← All tools
Mediation utility · Auction floors

Floor price & fill simulator.

Fill rate answers one question; the business is usually asking another. See what a floor does to fill, eCPM — and the number that matters, revenue per eligible opportunity.

Set the typical bid, how widely bids vary, how many bidders compete and your current floor. The simulator runs 6,000 seeded auctions per floor level.

Your numbers

Spread 0.5 means bids cluster tightly around the median; 1.2 means a long tail of occasional high bids.
Fill at current floor——
eCPM at current floor——
Revenue / 1,000 opps——
Best floor (model)——

—

Floor vs fill rate, eCPM and revenue per 1,000 opportunities (indexed to no floor)
Fill rateeCPMRevenue per opportunity
How it's calculated

The method, in plain terms.

  1. Bids modelled as a log-normal distribution (median, spread)
  2. Fill = P(bid ≥ floor); eCPM = E[bid | bid ≥ floor]
  3. Revenue per 1,000 opportunities = Fill × eCPM
Worked example

With a $6 median bid, raising the floor from $2 to $10 lifts observed eCPM sharply while fill collapses; revenue per 1,000 eligible opportunities peaks at a floor well below the one that maximises eCPM.

Defaults on this page are illustrative starting points, not benchmarks for your game. Replace them with your own numbers.

Embed this calculator

Free to embed on your blog, wiki or course page with the credit link intact.

FAQ

Common questions.

Does raising the price floor increase ad revenue?

Sometimes. A higher floor always raises observed eCPM and lowers fill. Revenue rises only if the extra price on the impressions that still clear outweighs the value lost on the impressions that no longer fill. In second-price clearing there is usually a revenue-maximising floor; in pure first-price clearing a floor mostly removes low bids.

What is revenue per eligible opportunity?

Ad revenue divided by every opportunity where an ad could have been served, filled or not. Unlike eCPM, it cannot be improved simply by rejecting cheap impressions, which is why it is the right scorecard for floor changes.

Why does the auction type matter?

In second-price clearing, the winner pays the higher of the second bid and the floor, so a floor between the top two bids raises the price. In first-price clearing, the winner pays their own bid, so a floor cannot raise the price of a bid that already clears it — it only turns low bids into no-fills (though in practice it can change how bidders shade their bids).

How realistic is this model?

It is deliberately simple: bids follow a log-normal distribution with the median and spread you set, bidders are independent, and there is no bid shading. Use it to build intuition about the fill–price trade-off, then test real floor changes against revenue per eligible opportunity by geo and placement.

Calculator by The Game Scientist ↗