You open the report and the two columns do not agree. Link clicks say one number. Landing page views say a much smaller one. Ask anywhere and the first answer you get is that your site is too slow. Sometimes that is right. But Meta's own documentation lists two reasons these metrics diverge, and the second one takes about two minutes to rule out — so it should be the one you check first.
The gap is not automatically a page-speed problem. The two columns count different events, and Meta's own documentation gives two reasons they diverge: people who leave before the page finishes loading, and clicks that never went to your site at all. Check the destination first — it is free and takes two minutes, while rebuilding page speed is a project.
The two metrics are not measuring the same event
Meta defines a landing page view as the number of times a person clicked a link and your site successfully loaded. It defines a link click as clicks on links in your ad — and it says plainly that if your ad has more than one destination, a click can send someone to a different place entirely. Its example is the headline pointing at one destination and the button pointing at another.
So one metric counts an intention and the other counts an arrival. A gap between them is not automatically a defect. What matters is which of the two documented causes is producing yours.
Cause two, checked first, because it is faster
Open the ad preview and click every clickable thing in it: the headline, the image or video, the call-to-action button, the page name, the description. Watch where each one goes.
- If any of them lead somewhere other than your landing page, part of your link-click count is made of clicks that were never going to land anywhere you can measure.
- Profile and page-name taps are the usual offenders — someone taps the brand name, lands on your Facebook page, and you were billed for the interest.
- This costs nothing to check and it either explains the gap or eliminates itself in two minutes.
Cause one, the famous one
The page did not finish loading before the person left. Test it the way the person experienced it, not the way you experience it: a phone, on cellular data, not your office wifi and not a cached desktop browser. Tap the link in the live ad and count the seconds until the page is usable.
Do these in this order. The destination check is free and takes two minutes. Rebuilding page speed is a project. Ruling out the cheap cause first is worth more than being right about the expensive one.
Why the gap is expensive rather than merely untidy
You are billed on clicks, not on arrivals. In one account I looked at, link clicks ran around 550 while landing page views sat near 60 — roughly nine out of ten clicks never became a visit. Cost per link click was about $0.26, which reads fine. Cost per landing page view was about $2.30, which does not. Same spend, same ads, two very different stories depending on which column you read. (Numbers rounded.)
That is also why swapping creative at this stage tends to do nothing. Creative decides whether someone clicks. This gap happens after the click. Better creative sends more people into the same leak.
The order that actually saves money
- Check every destination in the ad. Two minutes, free.
- Test load time on a phone on cellular. Ten minutes, free.
- Only then start arguing about creative, audience, or budget.
We built our optimizer to watch the step-to-step ratios rather than the headline numbers, because the headline numbers are where this particular problem hides — an account can post a perfectly respectable cost per click for weeks while almost nobody is arriving.