Promote winners
Promote winners moves a message that won on free placements to a paid Ad with a Hypothesis, a stop rule and the minimum budget.
Promote winners moves a message that won on free placements to a paid Ad. It carries a Hypothesis, a stop rule and at least the minimum budget, so money buys confirmation instead of guesses.
Ratings
| Dimension | Rating | Reads |
|---|---|---|
| Scalability | 5 of 5 | very high: higher grows further with spend or automation |
| Cost | 3 of 5 | medium: lower costs less per Customer |
| Latency | 1 of 5 | very low: lower answers sooner |
Budget
| Property | Value |
|---|---|
| Tier | $1k |
| Minimum effective budget | $600 a month |
| Monthly spend | Yields |
|---|---|
| $600 | One well-powered confirmation of one winner on one paid placement. |
| $10,000 | Winners on several placements, each against a holdout. |
| $100,000 | Every placement at scale, cut when incrementality falls below threshold. |
How it runs
A $1k budget buys one well-powered confirmation of one winner. From $10k each placement runs against a holdout, and collected revenue returns to the ad networks’ bidding with refunds as negatives. Launches and raises wait their delay window; pauses never wait.
Processes
Runs as Experimentation, accountable to the Marketing Team.
Also sold as a Service: Demand generation.
Fits
| Fits | Value |
|---|---|
| Family | demand |
| Shelf | Compound: builds an engine that keeps growing (Grow) |
| Moves | Acquisition payback |
| Motions | b2b, b2c |
| Venues | Google Search, Bing, Facebook, Instagram, TikTok, YouTube, Reddit, LinkedIn |
| States | Live, Repeatable, Scale |
Loop
| Step | Value |
|---|---|
| Hypothesis | A message that won on free placements pays back faster as a paid Ad than a new message. |
| Measure | Acquisition payback over 30d, at least 50 observations |
| Learn | The learn step reads the Readout, records Evidence and a Decision, and drafts the next Variant. |
Example
A subject line that won a 2,000-send email test for a Startup selling grant-writing software becomes a search ad headline. At $600 it beats the control headline’s cost per trial by 31 percent.