Facebook Link Clicks vs Landing Page Views: The Gap Test

A low landing page view count is often blamed on a missing pixel. Meta's own pages say something different: a landing page view can be counted through Meta's in-app browser with no pixel at all, and some counts are modeled. Below are Meta's definitions as we read them on 26 October 2026, a gap test you can run on your own account, and the causes in an order that rules out the cheap fixes before you suspect bots. In short, landing page views run lower for five reasons, checked in this order: the wrong comparison, a page that did not load, a visit Meta could not see, people tapping back before the load, and invalid traffic. There is no benchmark ratio here, because Meta publishes none and the figures that circulate online are unsourced.

For how invalid traffic reaches Meta ads in the first place, see our guide to Meta Ads click fraud.

Facebook Link Clicks vs Landing Page Views: What Meta Counts

Link clicks are the widest of the three metrics, outbound clicks count only clicks that leave Meta, and landing page views count only page loads Meta could observe or estimate. Here is each one in Meta's words.

• Link clicks: "The number of clicks on links within the ad that led to advertiser-specified destinations, on or off Meta technologies" (Meta Business Help: Link clicks). Meta's comparison page lists among them "Clicks on ad formats that take someone into a full-screen experience, such as lead forms, Instant Experience and collection" (Meta Business Help: link clicks vs outbound clicks).
• Outbound clicks: "The number of clicks on links that lead to destinations outside Meta technologies." Meta adds: "While some traffic might drop off between an outbound click and a webpage view, outbound clicks give you a closer approximation of the traffic intended for your website or app" (Meta Business Help: Outbound clicks).
• Landing page views: "The number of times a click on an ad resulted in the successful loading of a destination web page, mobile app or shop on Meta technologies. In some cases, this metric may be estimated." Meta says counting "can occur from having the Meta Pixel installed on your website or through the in-app browser on Meta technologies", and where data is partial or missing, "statistical modeling may be used to account for some events" (Meta Business Help: Landing page views).

Accessed 26 October 2026, English (US) versions of each Meta Business Help page.

Two dated changes affect any comparison you run today. Meta's Instant Experience page says: "Starting in June 2025, Instant Experiences will no longer count as a landing page view" (Meta Business Help: Instant Experience). And Meta began removing several metrics labeled as unique on 30 October 2024, with "Unique landing page views" on the list, while "unique link clicks" stays available (Meta Business Help: metrics being removed). So compare totals with totals. Unique link clicks against landing page views mixes two counting rules.

Here is how the three metrics nest for one ad that sends people to a website:

Our verdict: for ads that send people to a website, compare landing page views with outbound clicks, not link clicks. Meta's troubleshooting page points the same way: "If you're using an ad format that opens into a full screen experience, outbound clicks may provide more insight to the performance of your ads" (Meta Business Help: troubleshoot link clicks and outbound clicks). The same page explains when the two match: for a single image ad pointing at your website, Meta reports clicks on the call-to-action button or image "as both a link click and an outbound click", so the two only drift apart when full-screen formats or several destinations are involved. Use link clicks as the denominator only when no lead form, Instant Experience or collection is involved, and write down which one you used.

How to Run the Gap Test (and Why There Is No Good Ratio)

Put the three metrics side by side, divide landing page views (LPV) by outbound clicks, break the result down by placement, then impression device, then time of day, and compare each row with your own account's median. That is the whole test; the rest of this section is how to set it up so the numbers mean something.

Add the columns. Meta documents this path: click Campaigns, Ad sets or Ads, click the Columns dropdown menu, select Customize columns, choose the metrics, tick "Save as a column preset" and click Apply (Meta Business Help: customize columns). Select Link clicks, Outbound clicks and Landing page views. We name the preset "Gap test" so it is one click away next week.

Add the breakdowns. Meta's Ads Reporting help lists these under Delivery: Placement, Platform, Time of day (ad account time zone), Time of day (viewer's time zone), Impression device, Platform and device, and Placement and device; under Time it lists Day, Week and Month (Meta Business Help: breakdowns). In Ads Manager, Meta's path is: select the campaign, ad set or ad, click the Breakdown icon, then click By time, By delivery or By action and select a breakdown (Meta Business Help: navigate to breakdowns).

One edge case: Meta's Ads Reporting page notes that for offsite conversions the Delivery breakdowns available are only Impression device, Platform, Platform and device, and Placement (the Ads Manager page gives a longer list that adds Age, Gender and Country, still without Time of day). Whether that limit applies to landing page views is not stated; if Time of day is not offered next to your landing page view column, use the Day breakdown and look for bad days instead of bad hours.

The arithmetic, with no numbers borrowed from anyone:

As a pure illustration: if your Feed row has an LPV rate of A and the median across placements is M, the row is flagged when A is less than three quarters of M. Nothing about A or M comes from a benchmark. They come from your account, over your date range.

Why your median and not a published ratio. The rate depends on your page speed, your audience's devices, your placements and your formats, so a figure from someone else's account tells you nothing about yours. Your own median moves with all of those and leaves the outlier rows standing out.

Why viewer's time zone. The account time zone tells you when you paid; the viewer's time zone tells you what hour it was for the person who tapped. A run of clicks without loads at 3 a.m. viewer time deserves a closer look, since most real audiences are asleep then. That is our reasoning, not a Meta statement.

• Gap test worksheet, method as of 26 October 2026 (Meta metric definitions read that day).
• 1. Date range. Last 14 to 28 complete days, one objective, learning phase days left out where you can.
• 2. Columns. Link clicks, Outbound clicks and Landing page views, saved as a "Gap test" preset.
• 3. Rate. Landing page views divided by outbound clicks for website ads. Use link clicks only when no lead form, Instant Experience or collection is involved, and note which you used.
• 4. Cuts. Break down by Placement first, then Impression device, then Time of day (viewer's time zone).
• 5. Baseline. Your account's median LPV rate across placement rows. No external benchmark exists.
• 6. Flag rule (our rule of thumb, not a Meta figure). A row with at least 100 outbound clicks and an LPV rate under three quarters of the median. Recheck next week before acting.
• 7. Diagnose. Walk causes 0 to 4 below, in order, for each flagged row and write a verdict: page, measurement, abandonment, suspected invalid, or unresolved.
• 8. Re-run. After any change, such as a page speed fix or a placement exclusion, repeat over the same length of date range.

The 100-click floor is there because small rows swing. On a row with 20 outbound clicks, five failed loads move the rate by 25 percentage points; on a row with 400, the same five move it by about one. Below the floor, a low rate is noise until the next week confirms it.

Copy this table into a sheet and fill it from your own account:

Why Landing Page Views Are Lower Than Link Clicks: Causes in Triage Order

Landing page views run lower than link clicks for the five reasons listed above, and the order matters because each one is cheaper to rule out than the next. Meta names two of the causes itself: "The web page isn't fully loading or the app failed to open after a link click" and "You may have specified multiple link click destinations while creating an ad that doesn't click through to your landing page" (Meta Business Help: link clicks vs landing page views).

Cause 0: The Comparison Is Wrong

If you divide by link clicks on an ad that opens a lead form, an Instant Experience or a collection, part of the "gap" is clicks that were never meant to reach your site. The same goes for ads with several link destinations, some of which do not lead to your landing page. Since June 2025, an Instant Experience load does not count as a landing page view at all. Fix: switch the denominator to outbound clicks for website ads and rerun the worksheet. On ads with lead forms, Instant Experiences or collections, this step alone can remove much of the gap.

Cause 1: The Page Did Not Load

A slow page, a chain of redirects (tracking hops, http to https, country or language redirects), a heavy script that blocks the first render, or an app link that fails to open all lose the visit before Meta can count it. Check: open the live ad on a phone inside the Facebook and Instagram apps, on mobile data rather than office Wi-Fi, and watch how long the page takes to appear. Then load the final ad URL in a redirect checker and count the hops. Fix: cut hops, send the ad straight to the final URL, and make the first screen render before anything else loads.

Cause 2: Meta Could Not See the Visit (Pixel and In-App Browser)

A missing Meta Pixel does not zero your landing page views, because Meta also counts loads through its in-app browser and may model some events; what a missing, broken or blocked pixel does is leave visits outside the in-app browser uncounted. Meta's landing page view optimization page goes further: starting in July 2025, landing page view is available as an optimization event for advertisers with pixel sharing challenges (Meta Business Help: landing page view optimization).

Our inference, not Meta's wording: a consent banner that holds the pixel until the visitor accepts can leave more of those visits uncounted, for people who decline or ignore it. Meta does not say how its in-app browser counting interacts with consent. Modeling adds a smaller caveat: on low-volume rows, some of what you read may be an estimate. Check: confirm the pixel fires a PageView on the exact landing URL in Events Manager, including after the consent choice. Fix: repair the pixel and keep it on every landing page. If signal loss is also hiding your conversions, the server-side route is in our post on fraud-filtered conversions through the Conversions API. Meta does not document server-side events as a way to count landing page views, so treat that as a conversion fix, not a gap fix.

Cause 3: People Tapped Back Before the Page Loaded

Meta describes this use of the metric directly: "You can compare landing page views to link clicks to understand how many times people may have clicked on your ad but left before your website loaded" (Meta Business Help: Landing page views, linked above). Accidental taps while scrolling and people who change their mind both land here. In the worksheet this usually looks like a modest gap spread across most placements, not one row far below the rest. Fix: the same speed work as cause 1, plus ad creative that sets an honest expectation of the page. If your ad set optimizes for link clicks, test the landing page view performance goal; Meta says it "may prioritize higher-quality clicks" (Meta Business Help: landing page view best practices).

Cause 4: Invalid Traffic

Bots, click farms and incentivized taps can produce clicks with no load behind them. Meta's documented filtering is narrower than that. Facebook is "accredited for Sophisticated Invalid Traffic (SIVT) detection and filtration" for Feed, Instream video, Right Hand Column and Reels ad impressions and video plays, and Meta adds that "it's unlikely that all invalid activity can be identified and removed from the report results" (Meta Business Help: invalid traffic methodology). Audience Network employs "General Invalid Traffic Detection" for its impressions (Meta Business Help: Audience Network methodology). Verified negative: as of 26 October 2026, we found no Meta Help page saying link clicks are filtered for invalid activity, or that filtered clicks are not billed. The same methodology page does say that when events from a single user source exceed a defined threshold within a given time, later events are marked as non-billable and not reported, but it describes impressions and video plays, not link clicks. For what that means for credits, see whether Meta refunds invalid clicks.

Check: the tell is concentration. One placement, one device type or a run of odd viewer-time hours sits well below your median after causes 0 to 3 are ruled out. Read more on what counts as invalid traffic and on why Audience Network carries more invalid traffic. Fix: if the outlier is Audience Network, exclude Audience Network at account level rather than relying on ad set checkboxes; for other placements, weigh the row's results before you cut it.

A landing page view is not proof of a human. An automated browser that runs your page's scripts loads the page like anyone else, so if it arrives through the in-app browser or fires your pixel, Meta can count it. That is our reading of Meta's definition. The gap only shows clicks that never landed; it says nothing about the ones that did.

The Second Check: Compare Landing Page Views With GA4 Sessions

Meta's reports only show Meta's side, so the second check is your own analytics, such as Google Analytics 4 (GA4): if landing page views and your analytics sessions disagree in the same placement and on the same days, you have found where to dig. Tag every ad URL with UTM parameters, then compare Meta's landing page views with sessions from that source, day by day and campaign by campaign.

Expect the numbers to differ. Meta and your analytics use different counting rules, different attribution, and consent affects them differently, so equality is not the goal. Look for direction and concentration instead:

• Sessions far below landing page views on every row: usually measurement on your side, such as consent settings or a tag that loads late.
• Sessions close to landing page views, but engagement near zero in one campaign or placement: the page loaded, and something that does not behave like a person loaded it. This is where the "not proof of a human" point turns practical.
• Both counts fine, conversions flat: outside this test. The gap is not the problem there.

The analytics side of this, including the filters and segments that separate bot sessions from real ones, is in our recipe to detect and filter bot traffic in GA4.

What the Gap Can't Show, and What ClickFortify Adds

Comparing landing page views with outbound clicks is a good first filter for Meta ads and a poor final answer, because it only sees clicks that failed to land and misses every bot that loads the page. It cannot tell a slow page from a bot with certainty. Some counts are modeled. And if your traffic is low, two weeks may not be enough for a row to clear the 100-click floor.

ClickFortify protects Google Ads and Meta Ads campaigns, and on Meta it works on the other side of the gap, the visits that do reach your landing page. It scores the sessions that land on your site against dozens of behavioral, device and network signals and flags invalid traffic. On ad sets where it detects fraud, it can switch off Audience Network automatically (opt-in per ad account) and push the publishers you block to Meta's block list. When the CAPI proxy is enabled, conversions sent through ClickFortify are checked against the visitor's fraud score, and flagged events are dropped or tagged before Meta receives them. It does not block IP addresses on Meta, because Meta does not let advertisers exclude IP addresses.

Two limits, stated plainly. ClickFortify cannot see a click that never loaded your page: that is the gap itself, and only Meta's reports show it. And it does not see native Instant Form submissions, because those never load your site; validate those leads in your CRM.

There is a 7-day free trial, and no credit card is required. See Meta Ads protection for how the pieces fit together.

The Short Version

• For website ads, divide landing page views by outbound clicks, not link clicks, and compare totals with totals.
• Read the rate by placement, impression device and viewer-time hour, against your own account's median. No published benchmark exists.
• Rule out the comparison, the page, measurement and early back taps before you suspect bots.
• Act on concentration: one placement, device or hour band far below the median, two weeks running.
• A landing page view is not proof of a human, so pair the gap test with a check on the visits that did land.