First, confirm click-through is genuinely low
Meta reports more than one click-through figure. All-clicks click-through counts any click on the ad, including taps to expand text, view a profile or open an image; link click-through counts clicks that go to the destination. For a campaign meant to send people somewhere, link click-through is the figure that matters, and it is naturally lower. Make sure you and whoever raised the alarm are reading the same column.
Then compare like with like. Break the ad set down by placement, because the same ad can earn very different rates in different feeds, Stories and Reels, and an average can hide one weak placement among strong ones. Compare against your own account's history for campaigns with the same objective and audience temperature, not against a published industry average. Finally, look at the trend: a rate that was low from day one is this page's subject, while a rate that slid as FrequencyAverage times a person saw an ad. rose belongs to the frequency and creative fatigue guide. The CTR calculator is handy for recomputing rates from raw clicks and impressions when reports disagree.
Read click-through beside conversion rate before rewriting creative, as the CTR versus conversion rate comparison sets out, since a low rate that brings visitors who convert may call for more reach rather than a new ad.
Diagnosis table
Diagnosis
| Symptom | Check | Likely cause | Fix |
|---|---|---|---|
| Video ads get impressions but few views past the opening, and few clicks | Video retention in the opening seconds; thumb-stop on static images | Weak hook or a static image with no reason to stop | Rebuild the opening around a problem, result or person on screen |
| One placement drags the average down | Click-through broken down by placement | Format or crop not suited to the placements served | Supply placement-specific crops or restrict the unsuited placement |
| Rate low across placements; audience very broad or small and heavily reached | Audience size, frequency and the message's specificity | Audience too broad for the message, or saturated | Sharpen the message for the audience, or widen and refresh it |
| People watch but do not click | Offer, headline and call-to-action button | Offer or call to action unclear | State the offer and the next step plainly in the first frame and the button |
| Team reports a low rate that matches account history for this objective | Column used; objective; benchmark source | Reading all-clicks click-through or the wrong benchmark | Compare link click-through with the same objective's history |
Fix, verify and prevent each cause
A hook that does not stop the scroll
In a feed, people decide in the first moments whether to keep watching. A video that opens on a logo or a slow establishing shot, or a static image with nothing surprising in it, is scrolled past before the message arrives.
Fix: rebuild the opening around the problem, the result or a person speaking to camera, and test more than one opening on the same body.
Verify: retention in the opening seconds and link click-through both rise against the original.
Prevent: brief hooks as their own deliverable, using the patterns catalogued in the hooks and UGC guide.
Format or crop that does not suit the placement
Advantage+ placements serve one ad across many surfaces. A landscape video in a vertical placement, or an image whose text falls under the interface in Stories, looks broken and earns few clicks.
Fix: add placement-specific crops through placement customisation, respecting safe zones, or restrict a placement the creative genuinely cannot serve.
Verify: the weak placement's rate rises in the placement breakdown.
Prevent: produce each concept in the main aspect ratios from the start, following the formats and placements guide; Meta's placement documentation explains how delivery spreads across surfaces.
Audience too broad for the message, or saturated
A message written for one segment falls flat with everyone else, and a broad audience means most impressions reach everyone else. At the other extreme, a small audience reached many times stops responding, which pushes CPMCost per 1,000 impressions. up while clicks fall.
Fix: either make the message speak to the broad audience, or narrow delivery to the segment the message suits; for a saturated pool, widen it or rotate in new creative.
Verify: click-through rises without a matching collapse in reach.
Prevent: check audience size and frequency before launch, using the saturation signs set out in the frequency and creative fatigue guide.
Offer or call to action unclear
People can enjoy an ad and still have no reason to click: the price, the benefit or the next step is missing, or the button says something vague.
Fix: state what the person gets and what to do next in the first frame, the headline and the button, and test the offer as a variable on its own.
Verify: link click-through rises while watch time stays similar.
Prevent: include the offer in the creative brief and test it separately from the hook, as the creative testing guide describes when it covers what to vary. The guide to writing Meta ad copy covers what each text field should carry and how to keep the offer visible where placements truncate the primary text.
The wrong number, or the wrong benchmark
A video-views or awareness campaign is not optimised to produce clicks, so its CTRShare of impressions that result in a click. will be lower than a sales campaign's, and an all-clicks figure is not comparable with a link-click figure.
Fix: read link click-through for campaigns meant to send traffic, and compare with the same objective's history.
Verify: the comparison now uses one definition across reports.
Prevent: name the column and objective in every report, and set expectations by objective using the campaign objectives guide.
Prevention: creative QA before spend
Most launch-day click-through problems are visible before a penny is spent. A short quality check on every new ad catches them. Preview the ad in each placement it will serve, on a phone, and check that nothing important is cropped or hidden. Watch the first moments with the sound off and ask whether a stranger would stop. Read the headline and button and ask whether someone could say what they would get by clicking. Check the audience the ad set will reach and whether the message makes sense to all of it. Confirm the objective and the column everyone will judge it on.
After launch, give the ad enough delivery to judge, then read link click-through by placement against the account's history, not by the hour. Keep a simple log of hooks, formats and offers with their rates, so that the next brief starts from evidence. When click-through recovers but sales still do not follow, the problem has moved after the click, which the clicks but no conversions page covers.
A meal-kit brand with a flat launch
A meal-kit brand launches a new campaign with a polished landscape video that opens on the logo. After a week the link click-through rate is 0.42%, against 1.1% for its previous campaigns with the same objective. The placement breakdown shows feeds at 0.9% and Reels and Stories at 0.2%, where the landscape video is letterboxed and the price card sits under the interface. Retention data shows most viewers leave within the first two seconds. The team cuts a vertical version that opens on hands plating a finished dish with "Dinner in 15 minutes" on screen and the price in the first frame. In the following week link click-through reaches 1.0% overall, with Reels and Stories at 0.8%.
To see how a weak link click-through rate inflates the price of each visit, work through the CPC calculator set up for paid social with your own CPM.
Common mistakes
Comparing all-clicks click-through with link click-through. Judging against an industry average rather than your own history. Rewriting hook, offer and audience at once, so nothing is learned. Restricting all placements when one crop is the problem. Calling a launch-day problem fatigue.
