Hold Rate
Paid MediaAlso: Video Hold Rate · Audience Retention Rate
Quick definition
Hold Rate measures how many people keep watching a video ad past a given point, shown as a percentage of everyone who started it. It shows whether a video holds attention after the opening seconds, unlike view-through rate, which only tracks full completions, or click-through rate (CTR), which measures clicks.
How it varies across Australia
Hold rate drops fastest in the opening seconds across Australian social feeds regardless of platform. Vertical formats built for TikTok and Instagram Reels tend to hold attention longer than horizontal creative repurposed from older placements. Category and creative quality move the number more than platform choice does.
Explore paid media performance across Australian industries →What it actually means
Picture a room of a hundred people watching a presentation. Thumb-stop rate tells you how many turned their heads when the presenter walked in. Hold rate tells you how many were still in their seats five seconds later, and fifteen seconds after that. Both matter, but they measure different failures.
Hold rate is usually tracked at multiple checkpoints, such as three seconds, ten seconds and full length, then charted as a curve. A steep early drop tells you the opening frame isn't earning attention. A slow bleed throughout tells you the middle of the ad is dragging.
It sits alongside view-through rate, thumb-stop rate and video completion rate as one of the numbers that describe how a video ad actually performs, as opposed to click-through rate (CTR), which only measures the click at the end. Cost per mille (CPM) tells you what attention costs. Hold rate tells you whether you kept it.
Platforms like Meta and TikTok report hold rate curves natively, but most advertisers glance at a single completion figure and move on. That's a mistake. The shape of the curve is more useful than the endpoint, because it shows exactly where in the video people gave up.
The curve is more diagnostic than any single completion number, because it shows exactly where in the video people gave up.
How to calculate it
Hold Rate = Viewers still watching at a set point ÷ Viewers who started, multiplied by 100
Worked example. 10,000 people started your video ad. At the 15 second mark, 3,200 were still watching. Hold Rate = 3,200 ÷ 10,000 x 100 = 32%.
The Australian context
Australian attention spans on mobile skew short across every platform, and data costs still shape behaviour outside major cities, where users are quicker to skip on slower connections. Video ads built for metro audiences on fast Wi-Fi often show a hold rate cliff when the same creative runs regionally.
Testing hold rate by location before scaling spend nationally catches this early. Agencies that only report a national average miss the split between capital city performance and regional performance, which can differ enough to change a media plan. Vertical formats built for Reels and TikTok tend to hold better than repurposed television creative, because television pacing assumes a captive audience that mobile feeds don't give you.
Where people get this wrong
Hold Rate vs Video Completion Rate
| Hold Rate | Video Completion Rate | |
|---|---|---|
| What it measures | Attention retained at any checkpoint during playback | Percentage who watched the entire video |
| Best used for | Diagnosing exactly where creative loses people | Judging whether the length suits the format |
| Reported as | A curve across multiple time points | A single end percentage |
| Sensitive to video length | Less sensitive, checkpoint based | Highly sensitive, shorter videos score higher by default |
Related terms
Common questions
What's a good hold rate for a video ad?
There's no universal benchmark because format length and platform change the shape entirely. What matters more is whether your hold rate curve beats your own previous creative and whether the drop-off points make sense for your story structure.
Does hold rate affect ad costs?
Platforms like Meta and TikTok factor watch behaviour into their relevance and quality signals, which influences how cheaply an ad delivers. Video ads that hold attention longer often earn lower cost per mille (CPM) over time because the algorithm rewards content people don't skip.
How is hold rate different from thumb-stop rate?
Thumb-stop rate measures whether someone stopped scrolling in the first moment the ad appeared. Hold rate measures whether they kept watching after that. One is about the interruption, the other is about the follow-through.
Can hold rate predict conversions?
Not directly. Hold rate tells you about attention, not intent. Pair it with click-through rate (CTR) and downstream conversion data before deciding a high hold rate is actually driving business results.
Debrief
Get the next one
No spam. No fluff. Just the next article, straight to your inbox.
Keep exploring
About New Rebellion
New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.
How we think →