What Is Share Rate in Short-Form Video?

What it actually measures

In its strictest definition, share rate is the count of unique sends on a clip divided by the count of unique accounts the clip reached, expressed as a percentage. Instagram calls it sends and exposes it on the Professional Dashboard under reach. TikTok calls it shares and exposes it on the Creator Center under unique-viewers reached. YouTube Shorts calls it shares and exposes it on Studio Analytics under impressions and reach. The denominator that maps to the ranker's internal weighting is reach, not views, because reach is the surface on which the distribution decision was made.

What the metric isolates is the in-network distribution moment. A like records that the viewer enjoyed the clip; a save records that the viewer wants the clip later; a share records that the viewer wants someone else to see the clip now. The friction is the highest of any positive action on the surface. Mosseri's public guidance, carried in Influencer Marketing Hub's coverage of his sends-per-reach playbook, named the design intent directly: "When you're creating content, think about creating something that people would want to send to a friend. Don't force it, but sends are one of the biggest signals we use in ranking.. The TikTok Newsroom transparency center on the For You feed (newsroom.tiktok.com) names interactions including shares as factors that "are weighted based on indication of user interest," per TikTok's Newsroom.

How to calculate it

The formula is one line. Share rate equals unique sends divided by unique reach, multiplied by 100.

Walk it through with a fictional brand for grounding. Vespera Skin is a 22K-follower Instagram skincare DTC brand running roughly $4M ARR, run by a single founder with one part-time editor. On a Tuesday in April 2026 Vespera ships a Reel breaking down the difference between azelaic acid and tranexamic acid for post-inflammatory hyperpigmentation. The Reel reaches 38,400 unique accounts in the first 72 hours, earns 1,847 likes, 92 saves, and 312 sends. The share rate is 312 divided by 38,400, multiplied by 100, which equals 0.81 percent of reach. The same Reel computed against views (98,200) would post a share-rate-by-views of 0.32 percent, roughly two-and-a-half times lower. Two numbers, two denominators, two diagnoses. The reach denominator is the one the ranker reads against.

If Vespera's last 30 Reels post a median share-rate-by-reach of 0.34 percent, the azelaic-acid Reel is operating at roughly 2.4 times the account's own panel median, and is the kind of clip the Mosseri framework predicts will get a second test wave of distribution beyond the initial seed audience. The number is operator-useful only against the account's own 30-clip baseline and the category's median. The global median is the wrong comparison.

What good looks like by platform

Set the operating range for share rate from the account's own panel, because no industry report publishes a verified sends-per-reach median by audience size. The comparison that holds up in an audit is the account's own 30-clip baseline for similar formats, cross-checked against visible clips from category peers in the same audience tier. For context on how platforms treat the signal, TikTok's Newsroom transparency center names interactions including shares as factors weighted by user-interest indication, and Mosseri's verified 2025 ranking explainer names sends as one of the three top Reels signals.

The second is Buffer's TikTok algorithm guide, which lists the engagement signals — likes, shares, saves, comments and, 'perhaps most importantly,' video watch time — and notes only that 'a rewatch or share may signal stronger interest than a like or follow.' The guide publishes no ranked hierarchy, so the share-first operating read stays the operator's own framing: clips drafted for the share signal should resolve into something transferable (a reframe, a counter-intuitive number, a verdict the recipient will form an opinion about), not something receptive. A clip the viewer can summarize in one sentence to a friend is a clip the ranker reads as ceiling-raising.

A third pattern operators report is that share quality tracks clip length: mid-length clips on Reels and TikTok tend to carry the share signal because they give the audience enough material to summarize, while very short clips can earn high like rates but lower share rates. The shape differs on Shorts. Buffer's TikTok algorithm guide notes that 'a rewatch or share may signal stronger interest than a like or follow' — a relative-interest note, not a ranking.

Uncited working heuristics some operators use (not published benchmarks; validate against your own panel): 0.20 percent sends-per-reach on Reels, 0.40 percent shares-per-reach on TikTok, 0.10 percent shares-per-reach on Shorts.

What I look for when I audit this metric

I run a four-pass diagnosis on share rate before recommending any content change.

The first pass is the dual-denominator audit. I pull sends-per-reach and sends-per-view on the last 30 clips side by side. If the reach-denominator number is below the surface's floor but the view-denominator number is above it, the hook is the bottleneck and the back half is fine. If the reach-denominator passes but the view-denominator does not, the back half is the bottleneck and the hook is doing more work than the rest of the clip.

The second pass is the summarize-in-one-sentence test. I watch the clip with a colleague and ask them, after the cut, to summarize the clip's payload in a single sentence the recipient of a hypothetical DM would understand without context. If the summary holds together as one sentence, the clip is structurally shareable. If the summary requires "you have to watch it," the clip is structurally not.

The third pass is the recipient-graph audit. I pull the share-destination breakdown from the Professional Dashboard on Instagram or the Creator Center on TikTok. If the destinations are mostly story-share (the viewer's own audience), the clip is being broadcast rather than recommended, and the in-network signal is weaker than the raw share count suggests. If the destinations are mostly DM-to-individual, the clip is generating the recipient-graph distribution the ranker reads at full weight.

The fourth pass is the off-platform check. I check for cross-posting volume on adjacent surfaces: was the clip exported to other apps inside the share dialog? A clip that earns high off-platform export volume is generating reach that does not feed back to the ranker but does feed brand awareness. The combined number is the operator-relevant signal; either one alone undercounts.

Common mistakes

The first mistake I see is computing share rate over views instead of reach. The platform documentation on all three surfaces names reach as the ranker-facing denominator. Computing over views inflates the number on clips with heavy multi-loop delivery and produces a vanity metric that does not predict the next distribution wave.

The second mistake I see is treating share rate as interchangeable with engagement rate. Engagement rate aggregates likes, comments, saves, and shares into one weighted number. Share rate isolates the in-network distribution signal at the highest friction level. The two metrics diverge most on tutorial and reference content (high save rate, mid share rate) and on debate-bait content (mid save rate, high share rate). Treating them as one number hides the production decision that matters.

The third mistake I see is chasing the share-rate count rather than the share-rate quality. A useful framing is to build clips around hooks someone would want to copy and paste into a group chat. That framing distinguishes the kind of share that moves the clip into a new feed graph (the group-chat send) from the kind that sits in the viewer's own story (the broadcast). The first signal is what the ranker reads as a distribution multiplier; the second is closer to brand awareness with a weaker ranker read.

Where a planning-first tool fits

The brand-profile analysis I built in a planning-first tool tracks share-rate by destination, by category, and by clip length across an account's last 30 clips and adjacent competitor accounts (one input among several, not a substitute for the summarize-in-one-sentence test above). The operator's call on which payload to draft into the clip sits upstream of any dashboard.

Related Terms

Frequently asked questions