Run a Small, Capped Paid-Ads Trial Aimed at One Audience

You are a paid-media planner for an Australian small business about to spend its first real money on ads. Your discipline: one audience, one offer, one platform, a hard cap, and a decision at the end. Trials that can't fail teach nothing.

Details:
- Business: [BUSINESS — e.g. 'Clearwater Pool Care, Gold Coast']
- Total trial cap and duration: [CAP — e.g. '$500 over 4 weeks, not a cent more']
- The one audience: [AUDIENCE — e.g. 'pool owners in northern Gold Coast suburbs']
- The offer the ad makes: [OFFER — e.g. 'first clean $49' / 'free green-pool rescue quote']
- Where clicks land and what counts as a result: [DESTINATION + RESULT — e.g. 'booking page; result = booked job']
- What a result is worth: [VALUE — e.g. 'average new customer worth $600/year']
- Prior ads experience: [EXPERIENCE — e.g. 'boosted a few posts once']

Before planning, make the platform call: Google (people already searching for the fix) versus Meta (interrupting the right locals) — choose ONE for this trial using [AUDIENCE] and [OFFER], and defend it in 3 sentences. Name what would have made you choose the other.

Then:
1. Structure the trial: campaign objective in plain English, audience targeting settings described concretely (location radius, the interest/keyword logic), and why ONE campaign with 2 ad variants is enough at this cap — no more.
2. Write both ad variants in full (headlines, primary text/description within platform limits), differing on a single variable (hook or proof), both pointing at [OFFER] with no invented claims.
3. Split [CAP] across the weeks with reasoning (e.g. small week-1 learn budget, pause point, remainder to the better variant) — show the arithmetic; include the instruction for setting the hard cap in the platform so overspend is impossible.
4. Set the maths of success BEFORE launch: using [VALUE], calculate the break-even cost per result, then define three verdicts — clear win / murky / clear loss — as cost-per-result ranges, and what each verdict means for month two.
5. List the 5 checks during the trial (destination page loads on mobile, results tracked at [DESTINATION], frequency not blowing out, comments answered, receipts kept) and the 3 classic first-timer mistakes to avoid given [EXPERIENCE].

Format: 'Platform call' → 'Trial structure' → 'The two ads' → 'Budget split' (table) → 'Success maths' → 'During the trial'. Under 950 words, Australian spelling, numbers explicit.

Rules: never exceed [CAP]; every dollar allocated must appear in the table. Ad copy claims must trace to provided details — Australian Consumer Law treats misleading ads seriously (one-line flag, not a lecture); offers must be genuinely available. If [VALUE] or [RESULT] is missing, stop and ask for it first — a trial without success maths is just spending. Platform interfaces change: describe settings by intent, and tell the owner to confirm current names in the ads manager rather than quoting exact menu paths.

Copy the block above straight into Any AI tool — anything in [BRACKETS] is yours to fill in.