Include what the last round actually showed, the question you want to answer next, and the creative change that will test it. The team needs enough context to preserve useful work and produce a meaningful comparison. “Make more versions of the winner” leaves both decisions unclear.
Build the next brief around six inputs:
Previous creative: link the exact ads and version identifiers, including the copy and destination used.
Observed results: give the measurement period, audience, placement, and relevant business metric, with the media owner’s notes on delivery or tracking limitations.
Next hypothesis: state what you believe a creative change could improve and why the evidence makes it worth testing.
Changes and constants: identify the message, opening, proof, or visual element to vary, and what should remain consistent for the intended comparison.
Production assignment: list the required variants, formats, approved claims, source assets, review owner, and delivery date.
Learning decision: name who will set up and assess the test, the success measure, and what the team will decide when the evidence is sufficient.
For example, if a product ad attracts clicks but few qualified inquiries, the next question might be whether the opening makes the intended buyer clear enough. That is a hypothesis, not proof that the creative caused the drop-off. The media owner should first check whether the audience, destination, or measurement makes the result misleading.
Google’s experiment guidance recommends a clear hypothesis, defined success metrics, and one variable at a time when isolating an effect. If you intentionally compare complete concepts with several differences, describe the result as a comparison of those concepts. It cannot tell you which individual change caused the outcome.
Designity is a premium, AI-enabled and human-led Creative-as-a-Service platform. Its dedicated Creative Director can translate the agreed learning question into creative direction and specialist work; the Creative Project Manager coordinates versions, feedback, and delivery. Supply the underlying performance evidence and agree who owns test setup, media spend, tracking, and interpretation. Creative production alone does not establish those responsibilities or guarantee a statistically conclusive result.
Keep the approved brief attached to the delivered variants. The next report should identify which creative ran, what it tested, and what remains uncertain, so the following round starts with evidence rather than a new set of guesses.
