Instagram engagement rate calculator

The standard formula. Use when computing ER from public data without insights access.

Formula. (Likes + Comments) / Followers × 100

Example. Account with 10,000 followers. Last 12 posts averaged 250 likes and 15 comments per post. Average engagement: 265. ER = 265 / 10,000 × 100 = 2.65%.

When to use. Computing ER for accounts you don't own — competitor analysis, influencer-vetting for brand partnerships, public benchmark calculations.

Limitations. Doesn't include saves and shares (Instagram weights both heavily). Excludes accounts with hidden likes from being measurable through this method.

Closer to what Instagram's ranking algorithm weights internally.

Formula. (Likes + Comments + Saves + Shares) / Followers × 100

Example. 10,000 followers. Per-post average: 250 likes, 15 comments, 80 saves, 22 shares. Total engagement: 367. ER = 367 / 10,000 × 100 = 3.67%.

When to use. Internal account-quality measurement, content-experiment evaluation, anything that requires Insights access. Saves and shares require post-author analytics so this formula isn't computable from public data.

Why it matters. Saves and shares correlate with algorithmic distribution boost more strongly than likes do. An account with high extended ER but mediocre basic ER is producing high-quality content the algorithm rewards.

Strips out the follower baseline to isolate content quality.

Formula. (Likes + Comments + Saves + Shares) / Reach × 100

Example. A post got 250 likes, 15 comments, 80 saves, 22 shares — total 367 — and reached 1,200 unique accounts. ER by reach = 367 / 1,200 × 100 = 30.6%.

When to use. Content quality evaluation independent of follower count. Comparing performance across accounts of different sizes. Diagnosing whether low ER comes from bad content or low reach.

Why it matters. Two accounts can have identical basic ER but very different ER-by-reach. The one with higher ER-by-reach is producing better content; the one with lower is either reaching too many wrong-fit users or has a less-engaging post baseline.

Use case. Operator A/B tests where the question is whether a creative variant is genuinely better — ER-by-reach removes the noise of changing follower count between variants.

Take the computed ER and compare it with a like-for-like account baseline.

Step 1: define the comparison set. Use the same account, content type, formula, and reporting window.

Step 2: segment the context. Separate reels, carousels, single images, paid distribution, and major audience-source changes.

Step 3: compare with Insights. Calculate the same formula over equivalent periods and inspect reach, saves, shares, profile visits, and qualified outcomes alongside ER.

Step 4: identify the change. Investigate audience composition, content, distribution, format, timing, or recommendation-eligibility changes, then test one variable at a time.

Full benchmark tables and improvement framework: engagement rate benchmarks.

Five mistakes that produce misleading numbers.

Single-post ER vs account average. A single post that went viral gives a misleading ER if used to characterize the account. Use 12-30 post averages for account ER.

Including outlier posts. A post with 100x normal engagement (algorithmic-luck virality) skews averages upward. Some operators trim the top 10% and bottom 10% before averaging.

Ignoring follower-cohort drift. An account at 50K followers today computes ER over the last 30 posts based on today's 50K. Posts from when the account was at 30K had a different baseline. For drift-aware ER, use per-post follower count at posting time, not current count.

Including reels in feed-post ER. Reels typically have much higher reach than feed posts because they distribute to non-followers. Mixing reel reach into feed ER inflates the number. Compute reel ER and feed ER separately.

Not removing ghost followers. Accounts with 30%+ ghost followers compute ER against an inflated denominator. Removing ghosts before computing produces a more accurate number; without removal, ER is mathematically suppressed regardless of content quality.