Fazley Rabbi"The ad was great for the first few days. Now the same ad costs far more for each order, and I don't know why." Very often this is ad fatigue. The same people have seen the same image or video again and again and have stopped responding, or the people who were most interested have already bought. The answer is not finding one perfect ad. It is a steady supply of new ones, tested in a simple and fair way.
Signs an ad is tired
- The cost per message, lead or order climbs while nothing else has changed.
- The same people are seeing it many times; the frequency figure in your reports keeps rising.
- Comments say "seen this already", or reactions drop.
- Clicks fall even though the ad is still reaching people.
Before blaming the creative, rule out the other usual causes: stock running out, a price change, delivery problems, slow replies, or a competitor's sale.
What to test, from biggest effect to smallest
- The angle: the reason to buy. Price, quality, a problem solved, a gift idea, fast delivery, or trust (cash on delivery, easy exchange).
- The format: a short video, a single photo, a carousel of several products, or a simple customer-style phone video.
- The hook: the first second of the video or the first line of text. "Is your child's skin getting dry this winter?" works differently from a product name.
- The details: colours, caption length, headline wording, the button.
Most of what you learn comes from new angles and hooks, not from changing a button colour.
A weekly testing routine
- Keep a creative sheet. A Google Sheet listing every ad you make: its angle, format, hook, start date and results, measured as cost per order or lead from your own records.
- Make two to four new creatives each week. Change one main thing each time, usually the angle or the hook, so you can learn from the result.
- Test them side by side with the same audience and budget so the comparison is fair. Ads Manager also has built-in testing tools if you prefer a more formal test.
- Give each test enough time and spend to show a pattern before deciding. A day or two with a handful of results is mostly noise.
- Keep the winners and pause the losers, and write one line in the sheet about why you think each result happened.
- Refresh before the old winner fades, not after. When a winner starts to slow down, the next set should already be ready.
Making creatives without a big team
- Shoot on a phone in good daylight. Simple, real videos can do as well as polished ones, so do not wait for a studio.
- Record several different openings in one session and pair each with the same main video.
- Turn customer questions and messages (with permission, names hidden) into ideas for angles.
- AI tools such as ChatGPT or Claude can help draft hooks and captions in Bangla and English. Read every draft as a customer would before using it; they can sound generic.
- Make both square and vertical versions so the ad fits different placements.
Mistakes that waste tests
- Changing many things at once, so you never know which one mattered.
- Judging after a few hours.
- Killing an ad because of one bad day, or keeping a dying ad out of attachment.
- Testing only designs when the offer itself is the problem. If no creative works, check the offer and the reply process with this diagnosis checklist.
In-house versus outside help
Your team can shoot, write and keep the sheet; the product knowledge is yours. A specialist helps with the test structure, reading results correctly and deciding when to put more budget behind a winner. That is part of my social media and ads management work.
The smallest useful first step
Find the ad that has been running longest and write down three angles you have never tried. This week, make one simple phone video for the strongest angle and run it next to the old ad. If you would like help setting up the routine, get in touch.