Instagram Engagement Rate: What's Good in 2026? (Industry Benchmarks)

"What's a good engagement rate?" is the most common question agencies get from clients. The answer depends on follower count, industry, and content type - a blanket "2% is good" doesn't cut it.

Here are the actual benchmarks for 2026, based on data across millions of accounts.

How to Calculate Engagement Rate

The standard formula:

Engagement Rate = (Likes + Comments + Saves + Shares) ÷ Followers × 100

Some tools only count likes and comments. For a complete picture, include saves and shares - Instagram's algorithm weighs these heavily.

Try our free Instagram Engagement Rate Calculator to check yours instantly.

The Three Competing Instagram Engagement Rate Formulas

The formula above is the common one, but it is not the only one, and the others are not wrong. They answer different questions, and the same post produces a different percentage under each. Before comparing your number to any published benchmark, find out which of these three the benchmark used.

Engagement rate by followers (ER by follower)

(Interactions ÷ Followers) × 100

Asks: how much of my audience did this post move? It is the only formula you can run on an account you do not own, because follower count is public and reach is not. That is why nearly every published benchmark set uses it, and why competitor analysis is stuck with it. Its weakness is the denominator: followers are a list of people who once tapped a button, and many never see the post. If the list is stale or inflated by bought followers, the rate falls with no change to the content.

Engagement rate by reach (ER by reach)

(Interactions ÷ Accounts Reached) × 100

Asks: of the people who actually saw this, how many did something? This is the honest content-quality measure, because it removes the audience-size problem. Reach counts unique accounts, so someone who watched a Reel four times counts once. Its weakness is that distribution failure flatters it: a post shown to 200 people, 40 of them your most loyal followers, can post a spectacular rate while reaching almost nobody. Report it next to the reach number, never alone.

Engagement rate by views or impressions (ER by view)

(Interactions ÷ Views) × 100

Asks: per delivery, how often did an interaction happen? Views count every play or display including repeats, so the denominator is larger than reach and this rate always comes out below the reach-based one. It is not automatically the lowest of the three: a post that reaches fewer people than you have followers still scores higher on views than on followers. The ordering depends on your own numbers, so work it out rather than assume it. Instagram's own surfaces now lead with views rather than impressions as the delivery-side metric, and the two are not interchangeable in older documentation. If a tool hands you an "impressions" figure, confirm what it counts before it goes in a client deck.

The two forks hiding inside every formula

First, a rate can be calculated per post and then averaged, or per account by summing all interactions and dividing once. These give different answers, for reasons set out under mean-of-ratios below.

Second, what counts as an interaction. Socialinsider's Instagram benchmark page defines its rate as likes plus comments divided by followers. Rival IQ's benchmark report defines engagement more broadly, as likes, comments, favourites, shares and reactions over total follower count. Neither is wrong, but they are not the same measurement, and a figure from one cannot be laid beside a figure from the other.

Worked Example: One Post, Four Different Engagement Rates

A single post on an account with 12,400 followers: 486 likes, 37 comments, 92 saves, 21 shares, 9,310 accounts reached, 14,780 views. Total interactions across all four types: 486 + 37 + 92 + 21 = 636.

One post, nothing changed but the arithmetic, and the spread runs from 4.22% to 6.83%. If you have watched two tools report different engagement rates for the same account and assumed one was broken, this is almost always why. An agency can move a client's reported rate by more than half again without touching the content. Pick a formula, write it at the bottom of the report, and never change it mid-engagement.

Engagement Rate by Reach vs Engagement Rate by Followers: Which to Report

Use both, for different jobs.

Report by reach as the content-quality number. It answers what the client is actually asking, which is whether the posts are any good, and it is unaffected by follower list hygiene.

Use by-follower for competitive and benchmark work, because you have no choice. You cannot see a competitor's reach, so any claim about a competitor's engagement rate is a follower-denominator claim, and the only fair comparison is your own follower-denominator number.

The failure mode to avoid is mixing them in one sentence. "Our engagement is 6.8% versus the competitor's 1.9%" is not a finding if your 6.8% is reach-based and their 1.9% is follower-based. It is the same account winning an argument against itself.

What the Benchmarks Actually Say, and Why There Is No Table Here

This article used to carry three tables: average engagement by follower tier, by industry, and by content type. They have been removed, because we could not show you where a single number in them came from.

That matters more than it sounds. An engagement benchmark means nothing without four things attached: which formula produced it, what the sample was, when it was measured, and whether it is a mean or a median. A grid of tidy percentages carrying none of that is decoration, and if you paste it into a client report you are the one who has to defend it.

Here is what you can actually stand behind.

Use a published study, and cite it properly. Rival IQ's Social Media Industry Benchmark Report is the most widely referenced. Its 2025 edition analysed more than 4 million posts, selecting 150 companies at random from each of 14 industries out of a database of over 200,000, and reports the median rather than the mean. It defines engagement rate as total interactions divided by follower count, which is ER by follower, so its figures are not comparable to a reach-based number from another tool. Two of its headline findings: Instagram engagement fell 16% year on year, and carousels outperformed Reels.

Check the industry list before you quote it. It covers Alcohol, Fashion, Financial Services, Food and Beverage, Health and Beauty, Higher Ed, Home Decor, Influencers, Media, Nonprofits, Retail, Sports Teams, Tech and Software, and Travel. If your vertical is not on that list, no published median covers you, and any article handing you one for real estate, home services or B2B SaaS has invented it.

Benchmark against yourself first. Your own last 90 days, on one formula, beats any industry average, because it controls for your audience, your posting mix and your niche simultaneously. The method is in the final section of this article.

Treat follower-tier rules of thumb as folklore. That smaller accounts engage better is directionally true and mechanically obvious: the denominator is smaller, so a fixed core of engaged followers is a larger share of it. The specific percentage bands attached to that idea in most articles, including ours until now, trace back to nothing checkable.

How Reels and Stories Break Instagram Engagement Rate Benchmarks

Both formats violate the assumption the follower formula depends on: that the people who see a post are drawn from the people who follow the account.

Reels are shown to non-followers, so the denominator is the wrong audience

A Reel is recommended into feeds with no relationship to the account, so when one travels, reach can exceed follower count several times over and the numerator collects interactions from an audience the denominator does not contain. The effect cuts both ways at once. By followers, the rate inflates: interactions from strangers are divided by your own follower list, so the ceiling is unbounded and rates above 100% are arithmetically possible. By views, the same Reel deflates: repeat plays swell the denominator while the numerator counts only the one like a person can leave. The same success reads as a spike on one formula and a dip on the other, so segment Reels out and report them on their own line.

Stories have a different interaction set entirely

Stories carry no public like count, so the numerator has to be built from what Instagram exposes: replies, shares, sticker and poll taps, link taps, and profile visits from the frame. Forward taps and exits are navigation, not engagement. Two numbers are worth reporting and are routinely confused:

• Story engagement rate: interactions ÷ accounts reached on that frame.
• Completion rate: viewers on the final frame ÷ viewers on the first frame.

A Story set can hold people to the end while producing almost no interactions, or the reverse. One of the two tells the client half of what happened.

Carousels have a smaller version of the problem: swipe depth is where they earn their saves, and saves are the part of the numerator most likely to be dropped by a tool counting only likes and comments. If your carousels look flat against benchmark, check whether saves are counted before you change the content.

What Affects Engagement Rate

Positive factors

• Posting consistently (4-7 times per week)
• Reels and Carousels over single images
• Replying to every comment within the first hour
• Using 3-5 relevant hashtags (not 30)
• Posting when your audience is online (check Insights)
• Strong hooks in the first 1-3 seconds of Reels
• CTAs that ask for saves and shares ("Save this for later")

Negative factors

• Buying followers - inflates follower count, kills engagement rate
• Posting inconsistently - algorithm deprioritizes inactive accounts
• Ignoring DMs and comments - reduces algorithmic favor
• Using irrelevant hashtags - attracts the wrong audience
• Posting only promotional content - followers tune out

Numerator and Denominator: What Actually Moves an Engagement Rate

The list above is about the numerator. A large share of engagement rate movement comes from the denominator instead, and it goes unnoticed because the denominator has no story attached to it.

Denominator moves that look like content changes

• Follower growth. A campaign that adds 3,000 followers enlarges the divisor immediately, while those followers take time to start interacting. Engagement rate falls during successful growth, which is the usual reason a client asks why the number went down in a month everything went right.
• A follower cleanup. The divisor shrinks, the rate jumps, the content did not change.
• Which follower count you divided by. Followers at the start of the window, at the end, or at the moment the post went out are three different numbers on a growing account. Pick one rule and apply it to every post.
• A distribution shift. If Instagram serves more of your posts to non-followers, the follower-based rate rises and the reach-based rate typically falls, because the added audience is colder.
• Format mix. Shipping more Reels changes the average denominator across the whole set even if every individual post performs identically to last month.

Numerator moves worth separating

A save is being filed for later; a share is being used as social currency. Collapsing both into one interaction count with likes hides which you produced. The metric also cannot see comment quality, since a tagged friend and a paragraph each count as one, and some tools count the account's own replies as comments on its own post, inflating the rate on anything you moderate heavily.

How to Compare Instagram Engagement Rates Honestly

A comparison is only valid when five things match on both sides. Any mismatch and you are comparing measurements, not performance.

1. Same formula. Follower, reach, or views, picked once and stated.
2. Same interaction set. Likes and comments only, or including saves and shares. That gap alone was nearly a full percentage point in the worked example above.
3. Same window, same length. A quarter against a month always favours the month with the hit post in it.
4. Same content mix. Nine Reels is not comparable to nine static images, on any formula.
5. Same aggregation. Per post and averaged, or summed and divided once. Not the same number.

The mean-of-ratios trap

This is the aggregation error that survives every other check, so do the arithmetic once. Three posts on a 10,000-follower account:

Calculate the reach-based rate per post, then average the three: 12.5%, 3.0% and 5.0% give 6.83%.

Now sum first: 1,150 interactions ÷ 34,400 reach = 3.34%.

Both are correct, and they differ by a factor of two. The averaged version gave post A, which reached 400 people, the same weight as post B, which reached 30,000. The summed version weights each post by how many people it actually touched.

There is an exception worth knowing. With a follower denominator held constant, the two methods agree exactly, because the divisor is the same on every post. Per-post follower rates of 0.5%, 9.0% and 2.0% average to 3.83%, and 1,150 ÷ 30,000 is also 3.83%. That is why the error survives review: teams verify the method on follower data where it cannot fail, then carry it to reach data where it can.

Use the summed version for reach and views. Use either for followers, as long as the follower count is fixed.

Report a median and a range, not a mean

Instagram performance is skewed. Most posts land in a narrow band and one occasionally goes far beyond it, which drags a mean and leaves a median alone. Report the median for the window, the highest and lowest post, and the post count. Three numbers and a sample size is a finding; one number is a claim. A month with eight posts, one of them a Reel that travelled, does not support a statement about whether the strategy is working.

Why Instagram Engagement Rate Benchmarks Disagree With Each Other

If two benchmark reports do not agree, they are not in conflict. They measured different things, and each disclosed it in a methodology section most readers skip. The variables that separate published sets:

• What counts as an interaction. Socialinsider's Instagram benchmarks page states its rate as likes plus comments divided by followers. Rival IQ's benchmark report describes engagement as likes, comments, favourites, shares and reactions over total follower count. Adding saves and shares to a numerator raises every figure in the table.
• Sample construction. Rival IQ's report describes selecting 150 companies per industry from a database of over 200,000, with a minimum of 1,000 Instagram followers, reported as a median. Socialinsider's page describes analysing 35 million posts across 447,613 active pages over a full calendar year. Those two samples are answering different questions about different populations.
• Median versus mean. A median suppresses outlier posts and a mean includes them, so one dataset reported both ways yields two credible, different benchmarks.
• Follower floors. A sample excluding accounts under 1,000 followers removes the tier with the highest rates, pulling the all-industry figure down.
• Timeframe and account type. Distribution behaviour changes across a year, so a quarter is not a year. Brand pages and creator accounts also differ, and some sets mix them.
• Paid reach. If promoted posts are in the sample and the denominator is followers, paid distribution inflates the numerator with no corresponding change below the line.

None of that makes those reports untrustworthy. It makes the number non-portable. Quote the source alongside the figure, or do not quote the figure.

Six questions answer all of this, and an honest methodology section answers all six. If a report will not, it is not a benchmark, it is a graphic. Which denominator? Which interactions? Median or mean? How many accounts and how chosen? What date range, and how old is it now? Per post, or per account across the window?

Building Your Own Instagram Engagement Rate Benchmark

Published benchmarks are a sanity check, not a target. The benchmark that matters for one account is that account's own recent history, because it holds industry, audience, cadence and content style constant automatically. No published set can do that.

1. Fix the formula. Reach denominator, interactions including saves and shares, per post.
2. Take a rolling window. The last 30 posts, or the last 90 days, whichever is steadier for the posting rate.
3. Record the median, the range and the post count. The range is what tells you whether a new post is genuinely unusual or inside normal variance.
4. Segment by format. Separate medians for Reels, carousels and static posts. A blended median moves whenever the format mix moves.
5. Re-baseline after any structural change. A follower cleanup, a rebrand, or a shift in posting frequency or format mix invalidates the prior baseline. Start a new one and note the date.
6. Keep the raw counts, not just the rate. Interactions, reach and followers stored separately means you can recalculate any formula later. Storing only the percentage means you can never audit it.

With that baseline in place, a published benchmark table becomes what it should be: a check that you are in the right order of magnitude for the account size and industry, not a number to chase.

How Fake Followers Kill Engagement

If an account has 50K followers but 10K are bots, their engagement rate is calculated against 50K - making it look terrible even if real followers are engaging.

Before analyzing engagement, audit the account for fake followers using our free fake follower checker. Clean up ghost followers to get an accurate engagement baseline.

This is the clearest argument for keeping a reach-based rate beside the follower-based one. Bots never view posts, so they never enter the reach denominator. An account with a polluted follower list can show a poor follower-based rate and a healthy reach-based rate at once, and a wide, persistent gap between the two is itself the diagnostic: the problem is the list, not the content.

One warning about the cleanup. Removing 10,000 dead followers raises the follower-based rate immediately, and that jump is not a result. Note the purge date on the chart so nobody reads it as performance six months later.

Tracking Engagement Over Time

A single snapshot doesn't tell you much. Track engagement weekly to spot trends:

• Rising engagement - your content strategy is working, double down
• Flat engagement - time to experiment with new formats or topics
• Declining engagement - check for fake followers, posting frequency changes, or content fatigue

For agencies managing multiple accounts, tracking this manually is unsustainable. Use a CRM with built-in analytics to monitor all client accounts from one dashboard.

What to Tell Clients

When reporting engagement rates to clients:

1. Compare to their tier benchmark, not to influencers with millions of followers
2. Show the trend, not just the current number - improvement matters more than absolutes
3. Highlight saves and shares - these are stronger signals than likes
4. Connect engagement to business results - "Your engagement increased 40% this month, and we saw 23% more DM inquiries"
5. Print the formula in the report footer, denominator and interaction set included. One line, and it ends every argument about whose number is right.
6. Explain the denominator when growth is the story. If the account added followers fast, say plainly that the rate will dip while the new audience settles, and show the raw interaction count rising beside it.

If creator sourcing and vetting is taking more time than the campaigns themselves, this is the part brands most often hand to a specialist influencer marketing agency.

Interactions, reach and follower count plotted as three lines tell a client more than the ratio does, because a ratio can move for two opposite reasons and three lines cannot.

---

Track engagement rates across all your client accounts automatically. Start your free Inflowave trial

For more Instagram growth strategies, follow us on YouTube.

---

Explore More

• Feature: Analytics Dashboard - Track Instagram metrics that actually matter
• Free Tool: Engagement Rate Calculator - Check any account's real engagement rate
• Free Tools - All free Instagram marketing tools
• Feature: AI CRM - Tie analytics to real revenue with Instagram CRM
• Compare: Inflowave vs ManyChat - Full-stack analytics vs chatbot-only
• Instagram Analytics Guide - Stop tracking vanity metrics
• Best Instagram Hashtags - Most hashtag advice is years out of date
• Feature: Link in Bio - One bio link, every destination tracked
• Feature: Unified Inbox - Every DM, comment and message in one place
• Pricing - Plans built for Instagram businesses