Most ecommerce brands don’t have a creative problem. They have a learning problem. New ads go live, some do well, some don’t, and nobody can say why. The next batch is a guess, so results stay random. A creative testing process fixes that. Each ad becomes a small experiment, and each experiment makes the next round of creative better.
Here is the process we use, simplified so you can run it with or without an agency.
1. Start with one clear question per test
Every test should answer a single question. “Does a problem-first hook beat a product-first hook for this audience?” is a good question. “Will these five new ads work?” isn’t, because when one wins you won’t know which difference made it win.
Write the question down before you brief the creative. It keeps the team honest and makes the results much easier to read.
2. Test the big levers before the small ones
Not every variable is worth testing. Button colours and small copy tweaks rarely change results on their own. Test the elements that change how someone feels about the offer first:
- The angle: the reason someone should care, such as saving time, looking good, or feeling in control.
- The hook: the first second of a video or the headline on a static ad.
- The format: a founder talking to camera, a product demo, user-generated content, or a designed static.
- The offer: a bundle, a free gift, or a guarantee versus a straight price.
Once you’ve found an angle and format that work, test smaller variations inside them: new hooks, different first frames, alternative calls to action.
3. Keep everything else the same
If you change the audience, the budget, and the creative at the same time, you can’t learn anything from the result. Run creative tests with a stable setup: same campaign objective, similar audiences, and enough budget per ad for each one to get a fair read.
4. Decide what “winning” means before launch
Pick the metric that answers your question and agree on it up front. For a hook test, thumb-stop rate or hook rate (3-second views divided by impressions) tells you whether people stopped scrolling. For an offer test, look further down the funnel: cost per purchase or return on ad spend.
Agreeing in advance stops anyone from choosing the metric that flatters their favourite ad after the fact.
5. Log every test, especially the losers
Keep a simple test log with the date, the question, the variants, the result, and one sentence on what you learned. After a few months it becomes your most valuable creative asset. It shows what your customers respond to, and it stops you from re-testing ideas that already failed.
6. Turn learnings into the next brief
This is where most testing processes break. Results get reported, but they never reach the people making the next ads. Close the loop: every new creative brief should reference what the last tests taught you. That’s why we keep creative production and media buying in one team. The people reading the numbers are the same people making the next round of ads.
A testing rhythm that works
- Review last cycle’s results and update the test log.
- Choose one or two questions for the next cycle.
- Brief and produce the variants.
- Launch with a stable setup and agreed success metric.
- Read the results, document the learning, and repeat.
Want to see how this fits into a full engagement? Our process page walks through each stage, or you can book a free call and we’ll look at your current ads together.
