adlaneSign in Practical guide
Link clicks vs all clicks in Meta Ads: pick the measure that fits
An interaction with an ad is not always a visit to your website. Meta reports several click measures, and each counts a different action. Before calculating a rate or ranking creatives, state whether you care about any interaction, a link action or traffic that actually leaves the platform, then use that one measure consistently.
By Adlane · Updated
Link clicks vs all clicks: an interaction is not a visit
The broadest click measure counts many kinds of interaction with an ad: expanding the text, tapping the media, opening the profile, reacting, as well as following a link. A narrower measure counts clicks on links in the ad. A narrower measure still counts clicks that take the person off Meta’s platforms to your destination.
Those are three different populations, and they are nested. Every outbound click is a link click and every link click is a click, but not the other way round. An ad with a striking image can collect many interactions and few visits. Whether that is good depends entirely on what the ad was for.
Which Meta click metric counts what
Use this table to choose the measure before you build a report. The field names in your export may differ by tool and version, so check the exact field returned by the report instead of relying on a column simply labelled “clicks”.
| Click measure | What it counts | Use it for | Avoid |
|---|---|---|---|
| All clicks | Any counted interaction with the ad, including media and profile taps | Broad engagement questions | Treating it as website traffic |
| Link clicks | Clicks on links in the ad, including on-platform destinations | Interest in the call to action | Assuming each one left the platform |
| Outbound clicks | Clicks that send people off Meta to your destination | Website-traffic questions and cost per visit | Expecting it to equal analytics sessions |
| Rates built from any of the above | The chosen clicks divided by impressions | Comparing like with like | Ranking ads on CTRs built from different click types |
Keep one click definition across every ad and period you compare
Changing the click type or the denominator can make a metric appear to improve while nothing in behaviour changed. A report that quietly switches from outbound clicks to all clicks between two periods will show a jump in click-through rate that is pure definition. The same applies across ads: a ranking that mixes CTR columns from different click types is not a ranking of anything.
Show the underlying count beside any rate or cost per click. A cost per outbound click on 12 clicks is not comparable with one on 1,200. Rename report columns with the full measurement meaning, such as “outbound click-through rate”, so that readers cannot mistake one for another.
Illustrative example: high engagement, few outbound clicks
Illustrative example: a fictional ad records 2,400 all clicks, 500 link clicks and 310 outbound clicks on 100,000 impressions. Its all-click rate is 2.4 per cent; its outbound rate is 0.31 per cent. A second ad records 900 all clicks, 480 link clicks and 450 outbound clicks on the same impressions: an all-click rate of 0.9 per cent and an outbound rate of 0.45 per cent.
Ranked by all clicks, the first ad wins comfortably. Ranked by outbound clicks, the second ad sends more people to the website. The first ad’s engagement may matter for an awareness objective while telling you little about traffic. Neither ranking is wrong; they answer different questions. The numbers are invented to show the structure and are not benchmarks.
Measure website arrivals separately
Even the narrowest click measure counts something that happened on Meta, not something that happened on your site. People close the tab before the page loads, in-app browsers block scripts, consent choices suppress tags, and analytics counts sessions rather than clicks. An ad-click count will not equal analytics sessions, and the gap is normal rather than a fault.
Use website measurement for questions about arrivals and what happened after them, and keep it labelled as a separate source. Use the Meta click measure for the question it answers, which is how many people took that action on the ad. Put the two side by side with their definitions rather than forcing them to agree.
Compare click metrics in Adlane
In Adlane, connect Meta and select the ad account. Ask a bounded question that names the click type, such as “For these two ads over the last complete month, show impressions, all clicks, link clicks and outbound clicks, with each rate labelled by its click type, using the account’s attribution setting for any results.” Review the evidence with the counts visible before reading any rate.
If you decide to shift budget towards an ad on the basis of a website-traffic question, that change is prepared as a proposal you review and approve; it is not applied from the conversation. New campaigns start paused. Field availability depends on your Meta access and what the provider returns. Adlane is not affiliated with Meta.