Social Analytics

Reels Get 4x Interactions, Carousels 9x Saves of Single Images

Engagement Rate Versus Reach: What Data Science Teaches Online Marketers
Engagement Rate Versus Reach: What Data Science Teaches Online MarketersPexels / pixabay

Metricool's study of 24.3M Instagram posts shows Reels earn 4x the interactions and carousels 9x the saves of single images — and explains the math behind falling engagement rates.

By Nathan Brooks

3 min read

At a glance

  • Metricool analyzed 24.3 million Instagram posts from 375,118 accounts, comparing Jan–Feb 2025 with Jan–Feb 2026.
  • Reels got more than 4x the interactions of single-image posts; carousels got 9x the saves.
  • A post growing from 5,000 to 50,000 reach can see its engagement rate halve while interactions grow fivefold.
  • Engagement rate by reach suits individual posts; engagement rate by followers suits account trends.
  • The study was published June 16, 2026.

Metricool analyzed more than 24.3 million Instagram posts from 375,118 accounts, comparing January–February 2025 with the same months in 2026, and found Reels collected over four times as many interactions as single-image posts. Carousels pulled in nine times as many saves. The study, published June 16, 2026, lands on a problem every practitioner hits: reach climbs, engagement rate falls, and the client wants to know why.

The mechanism is simple arithmetic. Reach counts unique accounts that saw a post; engagement rate divides interactions by that number. When a post escapes your follower base, the denominator grows much faster than the numerator. A post with 300 interactions from 5,000 people runs a 6% rate. If it reaches 50,000 people and collects 1,500 interactions, the rate drops to 3% — even though interactions grew five times over.

Why the rate drops when reach climbs

Platforms distribute content in stages, and each stage dilutes the rate. First, your core followers see it — the heaviest engagers, so the rate peaks. If they respond, the post moves to similar audiences who don't know you yet, and fewer of them interact. Finally it hits Explore, For You, and suggested feeds, where many viewers scroll past in a second.

A lower rate on a high-reach post usually means you're reaching new people. That's growth, not failure.

Which formula should you use?

Two common formulas give different results for the same post. Engagement rate by reach — interactions divided by reach, times 100 — suits judging a single post. A Reel reaching 20,000 people with 1,000 interactions scores 5%. Engagement rate by followers — interactions divided by follower count — suits account trends. That same Reel on a 10,000-follower account scores 10%.

Use the reach-based rate for individual posts and the follower-based version for account-level tracking and competitor comparison.

What data science adds

Raw reach can look impressive while moving nothing. The analytical moves that matter:

  • Count viewers who stayed past a cutoff — commonly the first three seconds of a video — instead of everyone who scrolled by.
  • Watch for signal blur: a burst of broad, untargeted reach can teach the platform the wrong lessons about who your content is for.
  • Track whether high-reach posts convert into profile visits, link clicks, and site traffic.
  • Favor tight communities; posts shared inside niche groups and group chats often travel further than broad-auditude pushes.

Not all engagement carries equal weight, either. Saves and shares signal intent — including purchase interest — while likes take a second and tell you little. Many shares happen in private messages, so they're easy to undercount.

What to track per platform

Each network rewards different actions. On TikTok, engagement rate by reach beats the follower-based version, and completion rate plus average watch time matter most. On X, replies, reposts, and profile clicks are the clearest landing signals. On YouTube, watch average view duration, retention, and thumbnail click-through rate. Facebook Reels behave like Instagram and TikTok: shares and conversation-starting comments outweigh quick likes.

For daily practice, Metricool recommends tracking a saves-to-reach ratio (200 saves on 10,000 reach is 2%), sorting the last 30–60 days of posts by engagement rate to find shared patterns, and comparing formats only against themselves — carousels against carousels, Reels against Reels.

And when a client asks about the dip, the study suggests saying it plainly: "Total interactions went up, but the percentage dropped because the audience was so much bigger. It means we're reaching new people, which is exactly what we want."

As discovery feeds keep rewarding broad distribution, expect the gap between reach and engagement rate to widen — and the practitioners who read both together to keep the advantage.

Source: metricool.com

Nathan Brooks

Show full bio

Staff writer covering companies, deals and strategy in Social platform news and policy at Reach Brief.

More from Nathan Brooks →

Related articles

« Previous articleNext article »