Growing an Instagram Account Without Making Reels

We did not set out to run an experiment. We set out to keep posting. The account we look after had a stretch where nobody wanted to be on camera, the editing time for short video kept eating whole afternoons, and the numbers were not rewarding that time anyway. So we asked a narrower question than usual: if we take Reels off the table for a while, does anything else on Instagram still move an account forward, or do we just accept a slow fade? What follows is a record of what we tried, what we saw, and where our explanation gets shaky.

What we actually did

We made carousels the default format. Not the only format, but the one we planned around. A carousel is the multi-image post where a viewer swipes sideways through slides, and it was the format we could produce consistently without a camera, a script, or a hook we had to perform.

Our working structure for a carousel was roughly this: a first slide that states a specific problem or claim, three to six slides that work through it one idea at a time, and a final slide that either asks something or points somewhere. We kept the slides plain. Text on a background, occasional simple graphics, no stock photography of people pointing at whiteboards.

We also changed how we wrote captions. Instead of treating the caption as a summary of the slides, we treated it as the place where the argument continued, because the slides can only carry so much. And we started replying to comments as comments came in rather than in one batch at the end of the day, on the theory that a conversation that looks alive is more useful than one that looks archived.

We did not stop posting Reels entirely. We posted fewer of them, and the ones we posted were lower effort. We are not claiming a clean comparison, because it was not one.

What happened

Saves went up. That was the clearest change we noticed. Carousels get saved in a way that our Reels did not, which makes sense on reflection: a Reel is consumed and gone, while a carousel with several slides is something a person can return to. We started seeing the same posts saved repeatedly over days rather than in one burst.

Shares went up too, and they looked different from Reel shares. People were sending carousels into direct messages, which we take as a stronger signal than a passive share, though we cannot prove that.

Reach was slower. A Reel could find a cold audience in a way a carousel generally did not for us. Carousel reach looked more like a widening circle than a jump: it moved through people who already followed us, then through people adjacent to them, and it kept trickling rather than spiking.

Profile visits and follows seemed to come from a different place. With Reels, follows often arrived right after a single video. With carousels, follows seemed to arrive after somebody had seen two or three of our posts, which is a slower path and harder to attribute.

We should be clear about what we are not saying. We did not run a controlled test. We changed several things at once, we have no way to isolate the format from the caption changes or the reply timing, and we are not going to put a number on any of this because we do not have one we trust.

What seems to explain it

Our opinion, and we want to label it as opinion, is that carousels and Reels are asking the viewer for different things. A Reel asks for a few seconds of attention and gives a quick payoff. A carousel asks for a deliberate swipe and gives a structured payoff. Those attract different people and produce different behaviour.

The save behaviour fits that. If a post contains several distinct pieces of information, saving it is rational, because you will want it later. If a post is a single moment, there is less reason to save it.

The direct-message shares fit it too. Sending someone a carousel is a way of saying "this is the thing I was trying to explain to you." Sending a Reel is often just "look at this." One is a reference, the other is an amusement, and references travel differently.

The slower follow pattern also fits. Somebody who has seen three of your carousels has a clearer idea of what your account is for than somebody who saw one video. That person may be more likely to follow and more likely to stay.

What is still unclear

We do not know how much of this is the format and how much is the topic. Carousels suit explanatory content, and our account is explanatory. An account built on performance, humour, or personality may get nothing from this approach, and we would not be surprised if it got less than nothing, because a carousel is a poor container for a joke.

We do not know how the distribution works. We can see that carousels reached people, and we can see that Reels reached people faster, but we cannot see why, and we are not going to invent a mechanism. Anything we said about the algorithm here would be a story, not an observation.

We do not know whether the effect holds. A stretch of weeks is not a trend. It is possible that we were simply posting better than usual, or that the topics we chose happened to land, and the format had little to do with it.

We also do not know what we gave up. Every carousel we made was a Reel we did not make, and the reach we did not get from those Reels is invisible to us. That is the part of this we are least comfortable with.

If we wanted to settle any of it, the honest answer is that we would need to run the thing properly: hold the topic constant, vary only the format, and give it long enough that a good week does not fool us. We have not done that.

What we are doing next

We are keeping carousels as the spine of the account, because the format suits what we are trying to say and because we can sustain it. Sustainability matters more than we used to think. A format you can produce every week beats a format you can produce for three weeks and then abandon.

We are going to keep posting some Reels, but we are going to stop treating them as the only door to a new audience. If a Reel is the right container for something, we will make it. If it is not, we will not force it, and we will accept that the reach will be slower.

We are going to watch saves and direct-message shares more closely than likes, because those are the two signals that changed most and the two that seem to predict a follow rather than just a view. We are going to keep replying to comments as they arrive, since that costs nothing and we have no reason to stop.

And we are going to keep the question open. The useful thing about running an account over time is that you get to find out whether you were right. We suspect carousels will keep working for accounts that have something to explain, and we suspect they will do very little for accounts that do not. We would rather say that plainly and be corrected later than pretend we know now.

If you are considering the same trade, the version of it we would suggest is this: pick the format you can actually sustain, give it enough weeks to show a pattern, and watch saves and shares instead of views. Paid promotion is always available if you want reach faster, though it buys attention rather than interest, and the two are not the same thing. What we found is that a slower format asked more of the people who saw it, and the ones who stayed seemed to have chosen us on purpose.