Building your brand on Instagram is harder than it was three years ago, and the benchmark data says so plainly. A 2026 study of 35 million posts across 447,613 profiles put average engagement at 0.48%, down 24% year over year (Source: Socialinsider 2026). The accounts still growing are not working harder at the same tactics. They are working on different ones.
Most advice on this topic is recycled from 2019. Filters, hashtag blocks, boomerangs, posting at the golden hour. Instagram has since told everyone, on the record, what it actually measures, and the gap between that and the folklore is where most accounts lose a year.
Key Takeaways
- Instagram named three ranking signals: watch time, likes per reach, sends per reach.
- Sends reach strangers. Likes reach the people already following you.
- Hashtags are capped at five and do not improve reach.
- Reels bring new people in. Carousels hold the ones who arrived.
- Accounts leaning on reposted content lost recommendation eligibility in 2026.
The Three Signals Instagram Named Out Loud
In January 2025, Instagram head Adam Mosseri stated which signals matter most across the platform’s ranking systems: watch time, likes per reach, and sends per reach. Comments were not on the list, which surprised a lot of people who had spent years writing “drop a comment below.”
The useful part came with the split. Likes carry slightly more weight for content shown to people who already follow you. Sends carry slightly more weight for content shown to people who do not. That single distinction reorganises how you plan a month of posts. If the goal is depth with an existing audience, you want posts your followers react to. If the goal is new followers, you want posts somebody forwards to one specific person in a DM.
Mosseri added a note on watch time in February 2025 that gets missed: Instagram looks at both the percentage of a video watched and the raw number of seconds, so a longer video is not automatically punished for a lower completion rate. Details on how each surface weighs these differently are covered in this breakdown of the four ranking systems.
Decide What You Are Known For
Before any of the tactics matter, someone has to be able to describe your account in one sentence. Not your aesthetic. Your subject.
That shows up in three places. The handle, which should be findable and spellable after someone hears it once. The bio, which is indexed by Instagram’s in-app search, so the words people would actually type belong there rather than a tagline nobody searches. And the grid, where consistency does real work, though not the work people think. Matching filters are not a strategy. Being recognisable in a crowded feed before anyone reads the handle is.
Accounts that post everything to everyone have no subject, and an account with no subject gives the recommendation systems nothing to categorise it against.
Reels Bring Strangers, Carousels Keep Them
The 2026 benchmark data breaks engagement down by format: carousels at 0.55%, Reels at 0.52%, single images at 0.37% (Source: Socialinsider 2026). Those numbers get misread as a ranking of what to post. They are better read as a description of two different jobs.
Reels are where unconnected reach happens, because that surface is built to show content to people who do not follow the account. Carousels perform on engagement because the people seeing them mostly already follow you, and a carousel earns dwell time by making someone swipe. Static images sit lowest on both counts and should be the exception rather than the default.
A workable split: Reels to be found, carousels to be understood, and neither posted at a volume you cannot sustain past week three.
Write the Caption for the Send
If sends are what reaches new audiences, the question behind every post changes. Not “will people like this,” but “who would somebody forward this to, and why.”
Posts that get sent tend to do one of a few things. They name a situation precisely enough that a reader thinks of one particular person. They compress something useful into a form worth keeping, like a checklist or a comparison. Or they say a thing the reader has been trying to articulate and hands them the wording.
Engagement bait does the opposite. Asking for a comment on a post nobody would voluntarily share produces a metric that was not on Mosseri’s list in the first place.
Stories Are for Depth, Not Discovery
Stories mostly reach people who already follow you, so treating them as a growth channel misreads what they are for. They are where a following turns into something closer to a relationship.
Replies are the signal worth chasing there, and they have been climbing: a 2026 analysis of 24.3 million posts across 375,000 accounts found Story replies up 88% year over year (Source: Metricool 2026). Polls, question stickers and a genuine answer to what comes back are worth more than another polished graphic. More on what moves that surface specifically in this piece on what actually affects Story reach.
What Quietly Stopped Working
Four tactics are still repeated constantly and no longer hold.
Hashtag volume. Instagram capped hashtags at five per post in December 2025, and Mosseri has said publicly that hashtags do not improve reach. They function as topic labels. Five relevant ones in the caption is the whole of it, and reusing an identical block across every post reads as automated.
Reposting other people’s content. Accounts whose output is mostly reposts lost recommendation eligibility in April 2026. An aggregator account is now a dead end by design.
Uniform filters as a growth tactic. Visual consistency helps recognition. It has never been a distribution signal, and treating it as one absorbs hours that would be better spent on the hook.
Buying followers, likes or views. Beyond the legal exposure, purchased engagement fails on its own terms under a per-reach model: inflating the denominator without moving the numerator makes an account look worse to the ranking system, not better. The same arithmetic applies to what a view count actually measures.
A Cadence You Can Actually Hold
Posting schedules fail for boring reasons. Someone commits to daily Reels in January and stops in February, and the account reads as abandoned rather than selective.
Pick a frequency you can sustain through a busy quarter, then hold it. Three considered posts a week beats seven thin ones, and it beats fourteen followed by silence. Batch the production, keep the publishing spread out, and check Insights monthly on the three metrics Mosseri actually named rather than on follower count, which he has repeatedly said is not a direct ranking input.
One more thing worth noting: in a year-end memo published at the close of 2025, Mosseri said Instagram would prioritise raw, real human content over polished AI-generated material through 2026. Whether that holds is an open question. It is at least a stated direction, and it points away from the over-produced grid that dominated the previous five years.
Conclusion
Building your brand on Instagram in 2026 comes down to a narrower set of decisions than the advice industry suggests. Know what your account is about. Make Reels people send and carousels people finish. Use Stories to talk to the followers you already have. Ignore the hashtag arithmetic entirely.
Growth from here is slower than it was, and the benchmark decline says that is true for everyone, not just you. The accounts that compound are the ones still posting in month nine with a subject a stranger could name.
Read Next
- What Instagram Views Actually Measure, and Why Buying Them Fails
- How the Instagram Algorithm Works in 2026: Signals, Surfaces, and What Actually Changed
- Instagram Story Tips: What Actually Affects Reach, and What Doesn’t
