Definition
Floor budget of a market test: definition and calculation
The floor budget of a market test is the amount below which a market test produces no usable conclusion, because it never reaches the number of conversions required to tell a signal from chance. It is calculated backwards: you start from that number, move up to the clicks through the expected conversion rate, then to the budget through the cost per click. It does not depend on what the business is willing to spend.
The backward calculation
- The number of conversions required. It is set before any amount. Google Ads recommends evaluating a campaign over 30 days with at least 30 conversions; Meta places the end of the learning phase near 50 events over seven days. If the test compares two segments, the number applies to each.
- The number of clicks. Divide the conversions by the expected conversion rate, taken from your analytics over twelve months.
- The budget. Multiply the clicks by the cost per click found in the target market.
- The duration. Check that the weekly pace reaches the threshold of the platform you chose.
Why half a budget doesn't give half an answer
An advertising test cannot be scaled down like an order. It produces a single number, an observed conversion rate, which only becomes readable above a certain volume. Below that volume, the gap between two cities comes from chance as much as from the market.
A budget below the floor therefore doesn't buy a partial answer. It buys no answer at all, and it spends the money anyway. Everyone then reads the resulting number according to what they hoped for.
When the floor is out of reach
This is the normal situation for a small business. Three exits exist:
- Narrow the question. The number of conversions required depends on how many things you try to learn at once: one offer, one segment and one city cost a single floor.
- Switch levers. A less contested engine can deliver the same signal for a cheaper click. An event more frequent than a sale reaches the volume faster, without proving that people buy.
- Postpone. The postponement is written down with the missing amount and a review date.
Spending anyway is not a fourth exit.
Floor budget = (conversions required ÷ expected conversion rate) × cost per click
The number of conversions is set first, for each segment compared. The conversion rate comes from your analytics; the cost per click comes from the target market, not from your current campaigns.
Suppose a fictional business wants to compare two cities. It requires 30 conversions per city, so 60 in total. With an assumed conversion rate of 2.5%, it needs 2,400 clicks. At an assumed cost per click of $4, the floor comes to $9,600.
It only has $5,000. Rather than spreading that amount over both cities, it tests one: 1,200 clicks, or $4,800. It gets a narrow answer, but a readable one.
We set the number of conversions before we talk about money. The amount comes on the last line of the calculation: it follows from the question asked, it is not decided in a meeting. When an owner offers to cut the budget in half, we suggest cutting the question in two instead.
We also check the two assumptions the calculation rests on. A conversion rate taken from memory or incomplete tracking makes the floor wrong, on the optimistic side. So measurement gets fixed before the test, never during it.
Not to be confused with
- Useful conversion threshold
- The useful conversion threshold is a monthly number of conversions below which an automated bidding strategy cannot hold its target. The floor budget turns the number of conversions a test must produce into dollars.
- Marketing budget
- The marketing budget is set for the year, from the company's means and goals. The floor budget is calculated for a single test, from the proof it has to produce.
- Stop criterion
- The stop criterion sets the result and the date that decide whether investment in a market continues. The floor budget sets the amount below which the test cannot conclude anything.
Related concepts
- Useful conversion threshold
- Learning phase
- Stop criterion
- Conversion rate
- CPC
- Market entry cost
- First market signal
Further reading
- How much advertising budget do you need to test a new market?
- New markets: managing the learning phase of your ad campaigns
- New market testing: when should you continue, pause or stop?
- Google Ads location targeting: setting up a market test
- Microsoft Ads for B2B: how to test a new market
Related services
Frequently asked questions
My budget is below the floor. Should I launch anyway?
No. A budget below the floor doesn't buy a partial answer: it buys no answer and gets spent anyway. Three exits exist: test a single offer with a single segment, switch levers for a cheaper click, or postpone the test with a written review date.
Where do the conversion rate and cost per click in the calculation come from?
The conversion rate comes from your own website, measured over twelve months in your analytics rather than from memory. The cost per click is found in the target market, for example with the Google Ads Keyword Planner. Both are assumptions: the calculation is only as good as your tracking.
