What is hold rate in Meta Ads, and which formula should you use?
Learn what Meta Ads hold rate measures, why teams calculate it differently, and how to diagnose the part of a video that loses viewers after the opening.

Quick answer
Hold rate is a practitioner-built video-retention ratio, not one universal Meta metric. Define the numerator and denominator first, then use it to diagnose whether an ad keeps attention after the initial hook.
Quick answer: hold rate measures attention after the hook
Hold rate is a practitioner-built ratio for the share of early video viewers who continue to a deeper viewing milestone. A common version divides 15-second plays or ThruPlays by 3-second video plays and multiplies by 100. Some teams instead use a completion milestone such as 25%, 50%, or 95%, so the label alone does not identify the formula.
Meta does not publish one universal Ads Manager metric called hold rate. Meta publishes the underlying video-play metrics; advertisers, agencies, and analytics tools combine selected milestones into their own ratio. Always write the numerator and denominator beside the result before comparing ads or accepting a benchmark.
- Common continuation formula: 15-second plays or ThruPlays / 3-second video plays × 100.
- Alternative retention formula: selected completion milestone / selected early-play milestone × 100.
- What it answers: how much initial video attention survives to a defined later point.
- What it does not prove: that the viewer clicked, converted, or became profitable.
Choose a formula that matches the video and decision
There is no responsible way to choose a hold-rate formula without considering video length. A 15-second threshold is impossible for an eight-second asset unless the reporting metric substitutes a completion rule, while it captures only an early portion of a 60-second demonstration. A percentage milestone is easier to compare across different lengths, but it answers a different question from an absolute time threshold.
Use one written metric contract for a specific analysis: the exact Meta columns, report scope, dates, placements, video lengths, and calculation. Keep that contract unchanged while comparing creatives. If the contract changes, rename the metric rather than silently carrying the same hold-rate label forward.
- Short ads: use a completion-based milestone that every compared creative can reach.
- Similar-length ads: an absolute threshold can make the comparison easier to interpret.
- Mixed lengths: segment by duration or use the same percentage milestone.
- Agency reporting: show the formula in the dashboard, export, and client definition sheet.
Know what the underlying Meta metrics count
Meta defines a 3-second video play as a play lasting at least three seconds, or nearly the full length when the video is shorter than three seconds. Meta says it counts plays separately for each impression and excludes replayed time. This is an early playback event, not a unique-person count or proof of deliberate interest.
Meta describes ThruPlays as videos played to completion or for at least 15 seconds. That makes ThruPlay useful as one possible deeper milestone, but it is still a platform video-play metric rather than a sale or qualified lead. The hold-rate calculation inherits every scope, placement, duration, and reporting difference in its two component counts.
Read hook rate and hold rate as different transitions
Hook rate asks whether a served video impression reached an early viewing milestone. Hold rate starts after that early milestone and asks whether attention continued deeper into the asset. A creative can therefore have a strong hook and weak hold, or a modest hook and strong continuation among the people who stayed.
Build the chain explicitly: impression → 3-second play → selected deeper milestone → outbound click → landing page view → qualified outcome. Find the first weak transition before deciding which part of the ad to change.
- Weak hook and weak hold: inspect the opening, audience-message fit, pacing, and overall creative concept.
- Strong hook and weak hold: inspect the handoff from curiosity to message, proof, demonstration, and offer.
- Strong hold and weak outbound response: inspect the call to action, offer clarity, and whether the video creates a next step.
- Strong attention and traffic but weak outcomes: investigate the destination, measurement, offer, and downstream conversion path.
Diagnose the mid-video drop without chasing a universal benchmark
A universal good hold-rate threshold is unreliable because the ratio changes with the chosen formula, video length, placement mix, audience, objective, autoplay context, and reporting window. Use an internal controlled baseline before borrowing an external benchmark.
Compare ads under the same objective, date range, account time zone, audience context, placement scope, and metric contract. Then review the actual timeline around the selected milestone. These are diagnostic hypotheses, not automatic causes: confirm the suspected drop with controlled creative tests and downstream outcomes.
- Map when the problem, product, proof, offer, and call to action first appear.
- Check whether the opening promise is resolved, contradicted, or delayed in the next section.
- Inspect placement-level delivery before blaming one edit across every format.
- Test one responsible layer at a time while preserving a control creative.
- Judge the test with both retention and the campaign's qualified business result.
How this differs from nearby Meta Ads questions
This guide owns the early-to-deep continuation path: define hold rate and decide whether the body of the video retains attention after the opening. The hook-rate guide owns the impression-to-early-play transition and the first seconds. The ThruPlay guide defines Meta's platform milestone and the choice between video-view goals rather than a derived ratio.
The video-views-without-purchases guide begins after viewing already exists and diagnoses a downstream conversion gap. The creative-fatigue guide is longitudinal and asks whether response decays as an audience repeatedly sees an aging ad. Related routes: /blog/what-is-meta-ads-hook-rate, /blog/what-is-meta-ads-thruplay, /blog/why-meta-ads-video-views-but-no-purchases, and /blog/why-meta-ads-creative-fatigue-happens.
How an AdSpecIt-style audit helps
An AdSpecIt-style audit can place the hold-rate question beside the account evidence that gives it business meaning: impressions, spend, CTR, outbound clicks, landing page views, placements, creative identity, conversion volume, CPA, and ROAS under one consistent scope. That helps distinguish a retention problem from weak delivery, a placement shift, poor post-view response, or an otherwise healthy creative being judged by the wrong metric.
The connected audit should not pretend that a derived hold rate exists when its component video milestones were not ingested. Capture the exact 3-second, ThruPlay, 15-second, or completion columns from Ads Manager when the calculation itself is required, state the formula, and use the audit to decide whether to revise the opening-to-message transition, isolate a placement, improve the next step, or leave commercially healthy creative alone.
Sources and further reading
Research checked on 2026-10-27. Meta can change metric definitions, calculation methods, performance goals, and reporting interfaces, so verify the current Help Center pages and the exact columns in your own account. Hold rate is a practitioner label; the sources below define the underlying Meta metrics and provide practical context for evaluating early video attention.
- Meta Business Help Center — About Video Ad Metrics on Facebook and Instagram: https://www.facebook.com/business/help/1792720544284355
- Meta Business Help Center — 3-Second Video Plays: https://www.facebook.com/business/help/743427195703387
- Meta Business Help Center — 15-second ThruPlays: https://www.facebook.com/business/help/471190536725647
- Meta Business Help Center — About ThruPlay: https://www.facebook.com/business/help/2051461368219124
- Motion — How to Stop a Scroll in 3 Seconds (+50 Great Thumbstop Examples): https://motionapp.com/thumbstop-guide/how-to-stop-a-scroll-in-3-seconds/
Want AdSpecIt to audit your own account?
Connect your Meta Ads account, get a free score, and see the issues most likely to be hurting campaign performance before you spend more.
Get your free audit