Methodology
How Campaigntrial produces its numbers
Campaigntrial models how a synthetic audience may respond to your offer. It does not predict real conversions.
Generating the population
Each run creates synthetic profiles with demographic, behavioural and device attributes. Profession affects income only; age and income make small adjustments to technical familiarity and price sensitivity.
Inputs
Scores use the offer, price, page signals, audience and campaign details you enter. Campaigntrial does not inspect a live page.
Scoring a response
Each profile is scored for relevance, need, comprehension, trust, price acceptance, ad alignment and alternatives. Seeded variation prevents identical reactions.
Comparisons
The same inputs and seed reproduce the same population. Change one input to compare its effect.
Reading the metrics
| Metric | Reported as | Meaning |
|---|---|---|
| Simulated interest | Share of audience | Reached the interest threshold of 55 |
| Purchase intent | Share of audience | Reached the purchase intent threshold of 55 |
| Comprehension | Average score | How well the audience understood the offer |
| Trust | Average score | How credible the offer was to them |
| Price acceptance | Average score | Whether the price was defensible to them |
| Ad alignment | Average score | How closely the page matched the pre-click promise |
| Bounce likelihood | Average score | How likely they were to leave without engaging |
These scores are not conversion rates or forecasts.
Limits
These are modelled scores, not conversion forecasts. Campaigntrial does not assess visual design, brand equity, live competitors, seasonality or checkout behaviour. Validate decisions with interviews, real traffic and controlled tests.