Static Ad Design: What Actually Stops The Scroll
A static ad gets well under a second to earn a second look. Here is what the ones that get it have in common, and the mistakes that cost the rest.
Video gets the attention in every conversation about paid social, and static ads quietly keep carrying accounts. They are faster to make, cheaper to vary, and the quickest way to find out which message your market actually responds to.
They also get less time than anything else on the screen. A thumb moving through a feed gives a still image well under a second to earn a second look. Here is what the ones that earn it have in common.
One Message, One Focal Point
The most common reason a static ad fails is that it is trying to be a landing page. Three benefits, a discount, a testimonial, four product shots and a logo, all at the same volume, so nothing wins and the eye slides off.
Meta dropped its old rule limiting text to a fifth of the image years ago, but the reason it existed has not gone away. An image that is mostly words reads as an advert before it reads as anything else, and gets scrolled past.
The Anatomy Of One That Works
The visual hook
The product, a face, or a striking situation, large enough to read at thumbnail size. This is what stops the thumb, and it has to work before a single word is read.
The headline
Six to eight words that state the benefit or the problem, not the brand name. Written for someone who has never heard of you, because most of the people seeing it have not.
One line of proof
A number, a rating, a recognisable stockist, a guarantee. One is enough, and it should sit close to the headline so they are read together.
The offer, if there is one
Price, discount or free trial, set so it is unmissable but secondary to the hook. An offer that is bigger than the product is selling the discount, not the product.
A small, clear brand mark
Big enough to recognise, small enough to stay out of the way. A logo the size of the headline is the clearest sign an ad was approved by committee.
Design At The Size It Is Seen
- Review on a phone, not a monitor. An ad that looks balanced at full size on a 27-inch screen is often unreadable at the width of a feed. Check every draft on the device it will run on.
- Contrast beats colour. A headline that disappears against a busy background loses more than any palette choice gains. If it does not read in greyscale, it will not read in the feed.
- Respect the safe zones. In a 9:16 frame the platform's own buttons, captions and profile details sit over the top and bottom of the image. Keep text and logos out of those areas, and check each platform's current guidance, because the zones change.
- Make it look native, not cheap. The ads that perform often look like something a person would post. That is a design choice, made carefully, and it is not the same as an unfinished design.
We design static and motion ad creatives as testable sets, one strong concept in every size and variant a paid account needs.
See The ServiceThe Variations Worth Testing
Static ads are where testing is cheapest, so test the things that actually change who responds. Change one at a time, or the result will not tell you anything you can reuse.
| Format | What it does |
|---|---|
| Product on colour | Clean, fast to read, and the baseline every other test is judged against |
| Problem and solution | Names the frustration first, then shows the product fixing it |
| Before and after | Makes the result the hook, where the claim can be shown honestly |
| Comparison | Your product against the usual alternative, on the one point you win |
| Review card | A real customer line set as the design, with the rating beside it |
| Offer-led | The price or discount as the headline, for audiences who already know you |
Within a format, the headline angle is the next lever: lead with the benefit, the problem, the price or the proof. One of these will beat the others for your product, and it is rarely the one the brand team expects.
Where AI Helps, And Where It Does Not
AI has changed the economics of variation more than anything else. The same product in six settings, a new background for each season, a scene that would have needed a shoot: these are now an afternoon rather than a booking.
What it has not changed is the idea, the headline and the brand accuracy. Those still need a person who knows what the ad is for. Our rule is that the product is always made from photos of the real product, and the world around it can be generated. The honest comparison with a real shoot goes further into where that line sits.
Five Mistakes That Cost The Most
- Too many words. If the ad needs a paragraph, the message is not decided yet.
- A product too small to recognise. At feed size, a product that fills a tenth of the frame is a smudge.
- Stock-photo sameness. An image a viewer has seen in five other ads this week is an image they have learned to skip.
- One version. A single design tells you whether that design worked, and nothing about why.
- Judging on day one. Give each variant enough spend to have been seen by a comparable audience before calling it.
Statics and video work best together: statics to find the message cheaply, video to carry the winners further. We cover the video side in building a testing system rather than a hero film.
Common Questions
- Do static ads still work on Instagram and Facebook?
- Yes. They are faster and cheaper to vary than video, which makes them the quickest way to find the message that works before you invest in motion.
- How much text should a static ad have?
- As little as carries the message, usually a headline of six to eight words and one line of proof. An image that is mostly text reads as an ad and gets skipped.
- How many static ad variations should we test?
- Start with three to six variants of one concept, changing one thing at a time, such as the format or the headline angle, so the result tells you something you can reuse.