Same message (the Cook Candle clears cooking smells) written five ways. We think a scene or a question lands better than a product description, because the reader has to recognise the problem before the solution means anything.
The opening structure of the primary text only. Same three videos, same link, same audience.
£150 spent on each version, or 28 days, whichever comes first. Not before.
Which structure to lead with across Pott product copy. The Tomato round uses the same structures, so the two reads stack.
Each day's line is the whole test so far, so it settles as the numbers grow
Shows whether every version is getting its share
The count behind the cost per purchase
The earliest signal: does the ad earn the click
| Version | Spent | Link clicks | CTR | Purchases | Cost per purchase | Purchases per click | ROAS |
|---|---|---|---|---|---|---|---|
| Question first | £217 | 86 | 1.30% | 8 | £27.15 | 9.30% | 3.14x |
| Review first | £26.59 | 6 | 0.77% | 1 | £26.59 | 16.67% | 3.03x |
| Moment first | £250 | 90 | 1.59% | 10 | £25.00 | 11.11% | 3.12x |
| Buy once first | £52.40 | 10 | 0.66% | 1 | £52.40 | 10.00% | 2.63x |
| Short copy | £42.61 | 13 | 1.07% | 0 | – | 0.00% | 0.00x |
| Day | Question first spend · clicks · purchases | Review first spend · clicks · purchases | Moment first spend · clicks · purchases | Buy once first spend · clicks · purchases | Short copy spend · clicks · purchases |
|---|---|---|---|---|---|
| 12 Oct | £1.17 · 0 · 0 | £0.90 · 0 · 0 | £4.05 · 3 · 1 | £2.32 · 0 · 0 | £2.23 · 0 · 0 |
| 13 Oct | £5.89 · 3 · 0 | £3.83 · 0 · 0 | £31.74 · 11 · 1 | £1.22 · 0 · 0 | £19.94 · 7 · 0 |
| 14 Oct | £35.24 · 20 · 2 | £1.52 · 1 · 0 | £37.95 · 13 · 1 | £5.17 · 0 · 0 | £4.93 · 1 · 0 |
| 15 Oct | £39.84 · 12 · 0 | £4.69 · 1 · 0 | £33.23 · 6 · 1 | £11.79 · 1 · 1 | £10.24 · 3 · 0 |
| 16 Oct | £37.88 · 16 · 0 | £7.63 · 2 · 1 | £37.26 · 13 · 2 | £6.45 · 1 · 0 | £3.01 · 0 · 0 |
| 17 Oct | £32.94 · 12 · 3 | £3.16 · 0 · 0 | £58.34 · 25 · 0 | £11.29 · 3 · 0 | £1.50 · 0 · 0 |
| 18 Oct | £63.02 · 22 · 3 | £4.85 · 2 · 0 | £46.36 · 19 · 4 | £13.92 · 5 · 0 | £0.76 · 2 · 0 |
| 19 Oct | £1.24 · 1 · 0 | £0.01 · 0 · 0 | £1.11 · 0 · 0 | £0.24 · 0 · 0 | £0 · 0 · 0 |
The 5 texts run as options inside 3 ads. Meta chooses which text each person sees, so the texts never get equal spend, and Reels and Stories show no primary text at all, so text spend adds up to less than the ads' total. Per-text results are a steer, not a verdict: a colder text can look weaker while still doing its job for people who are not ready yet.
Spend, clicks and purchases are as reported by Meta for the ads in each test, from the test's start date to today, which is partial. Where versions share an ad set, Meta decides how much each one spends, so compare cost per purchase and purchases per click rather than totals. A test is only called once its rule is met; stopping early on a good week is how false winners get made.