A still photo beat 99.5% of the reels we track

On 3 August 2026 @realmrpatty posted a photograph of a double smash burger. It has 2,425,971 likes and 5,106 comments. It is not a reel — there is no motion, no sound, and no trending audio anywhere in it. We index 22,118 Instagram posts with like counts; exactly 109 of them beat this one.

What the distribution actually looks like

Percentile claims are meaningless without the shape of the data, so here it is. Across 22,118 indexed Instagram posts carrying a like count: the median is 708 likes, the mean is 62,206, the 99th percentile is 1,310,156, and the single largest is 34,879,796.

The gap between the median and the mean — 708 against 62,206 — is the whole story of this platform in two numbers. A handful of enormous posts drag the average up by a factor of 88 while the typical post gets a few hundred likes. Any advice built on averages is describing a post almost nobody has.

The median Instagram post in our index gets 708 likes. Everything you read about “what works” is written about the other end of the curve.

The mechanism

The stored breakdown describes it plainly: an extreme close-up with a slow, exaggerated cheese drip falling the length of the frame, and a cartoon character underneath with its mouth open, waiting to catch it.

1. One detail pushed past realism

Not a nicer photo of a burger. A drip that no real burger produces, running further than physics allows. The exaggeration is the content — it is what makes a static frame worth stopping on, and it is the part most brands sand off in the name of looking professional.

2. A character supplies the reaction

The cartoon does the job a reel does with a face and a voice: it tells you how to feel about what you are looking at. You are not asked to admire the product, you are shown someone wanting it. That transfers to a still image at almost no cost.

3. Nothing depends on sound or timing

No trending audio means no expiry date. A reel built on a sound is dead when the sound dies. This frame works identically in six months, which is why formats like it get recycled for years.

What this does not prove

One post is one post. This is not evidence that photos beat reels in general, and we are not going to claim it is — our index skews toward video because that is what we set out to collect. What it does prove is narrower and more useful: a still image can still reach the top half percent, so the blanket advice that only video works is wrong often enough to cost you.

There is also a measurement caveat worth being straight about. Instagram reports no view count for photo posts, so every number here is likes and comments. We have not compared reach, because for this post the platform does not publish it.

How to run it

1. Pick the single most appetising or most satisfying detail of your product — the drip, the pour, the peel, the snap.
2. Exaggerate it past what the product actually does. Longer, slower, bigger. Realism is not the goal, appetite is.
3. Put a character in frame reacting to it — illustrated, not photographed. The reaction tells the viewer what to feel.
4. High saturation, high contrast, dark background. The product should look lit, not photographed.
5. Skip the trending audio entirely. If the idea only works with a sound, it is a reel idea, not this one.
6. Post it as a photo. The format is the point — it stands out precisely because the feed around it moves.

▶ See the original: https://www.instagram.com/p/DbmG-cnkT_O/