Journal Testing Meta ads for B2B.

Your best-performing ad is the control. Every new concept runs against it, under the same audience and budget, so a win means the new creative is better and not that it had an easier week. When a challenger beats the control, it becomes the new control.

If a new ad changes the hook, the picture, the format and the offer at once, a win tells you nothing you can reuse. Change one, in this order:

1. The hook: the first line of the primary text and the words on the image.
2. The format: a static image, a carousel or a short video.
3. The visual: the picture, the colour field, the layout.
4. The offer: what you ask the buyer to do next.

Most people meet your ad at the size of a thumbnail, moving. Check every headline at that size: if it cannot be read there, it is too long. Use few words on the image, and put the rest in the primary text, front-loaded so the point shows before the “more” cut.

Cheap clicks are easy to buy and rarely what a B2B team needs. Judge each test on the result nearest to a sale that you can measure: booked calls, demo requests or qualified leads. Give each test enough spend and time to mean something before you call it, and do not crown a winner on a handful of clicks.

Write down each concept, the one thing it changed, the result and the dates. After a month the log shows which kinds of hook work for your buyers, and the next round starts from that, not from a blank page. This is what our monthly creative report is: see what Meta Ad Creative includes.

They look like the rest of the category, so the feed treats them like the category. They lead with the logo instead of the buyer’s problem. They carry a paragraph of text on the image. And they test five things at once, so nobody learns anything.