Your Conversions API is connected. Events are flowing. Events Manager shows green ticks. Yet your Meta ads still cannot find buyers reliably, costs keep creeping up, and your team insists “tracking is fixed”. Here is the uncomfortable truth: an event arriving is not the same as an event Meta can use. If Meta cannot match that event to a real person, it cannot learn from it — and that is exactly what Meta Conversions API event match quality measures.
Think of it this way. The Pixel tells Meta “someone bought”. The Conversions API tells Meta “this specific person bought, and here is how we know”. Without the second half, Meta sees conversions but cannot connect them to the people who saw your ads, so optimisation stays blind. You can spend more, test more creative and rebuild audiences, and still get the same flat results — because the learning signal itself is weak.
The direct answer: what Meta Conversions API event match quality really means
Meta Conversions API event match quality is Meta’s score, out of 10, for how well the customer information sent with your server events lets Meta match each event to a Meta account. It is calculated from the quality of the information you send and the share of events that actually match, using recent data. Higher match quality means Meta can attribute and optimise better; low match quality means many of your conversions are effectively invisible to optimisation.
Why “CAPI is firing” is not enough
Most tracking audits stop at delivery: is the Purchase event arriving, and is deduplication working? Those checks matter, but they miss the identity layer. A server event can arrive perfectly, with the right name, value and timestamp, and still score poorly if it carries almost no customer information — no hashed email, no phone number, no click ID, no browser ID. Meta receives the event, but cannot confidently tie it to anyone.
This happens most often after a platform migration or a “quick” CAPI setup: a Shopify app, a server-side GTM container or a direct integration sends events, but nobody configured the customer information parameters, or the checkout does not pass them at the moment the event fires. The dashboard looks healthy. The match quality score tells the real story.
Note the scope, too. Event match quality applies to website events sent through the Conversions API. Pixel-only events do not get a score, and app or offline events follow different rules. If you only look at Pixel events, you will never see this diagnostic at all — which is why many teams miss it for months.
The 6-point Meta Conversions API event match quality diagnostic
1. Check the score per event, not just overall. In Events Manager, open your dataset and look at event match quality for Purchase, Lead, CompleteRegistration and AddToCart separately. A strong Purchase score can hide a terrible Lead score, and if Lead is your optimisation event, that is the one that matters.
2. Audit which customer parameters you actually send. Meta ranks email and click ID (fbc) highest, followed by phone number, country, date of birth, external ID, browser ID (fbp) and Facebook login ID, then name, city and postcode. If your events send only IP address and user agent, your ceiling is low. List, event by event, which parameters arrive — and which are blank in practice, not just in theory.
3. Fix normalisation and hashing before adding more fields. Emails must be trimmed and lowercased, phone numbers must include the country code with symbols and leading zeros removed, and contact fields must be SHA-256 hashed. IP address, user agent, fbc and fbp must not be hashed. A single formatting error — an unhashed email, a phone number without a country code — can break matching for that whole parameter. Meta’s own Conversions API documentation sets out these rules, and they are worth checking field by field rather than trusting a plugin’s defaults.
4. Capture identifiers at the right moment. The click ID and browser ID must be captured when the visitor arrives, not reconstructed later. If your checkout or lead form loses the fbc value between landing page and conversion, Meta loses its strongest matching signal. The same applies to email: capture it on the form or checkout that triggers the event, with proper consent, rather than trying to join it afterwards.
5. Check coverage, freshness and deduplication together. A high match quality score on a tiny share of events will not fix optimisation. Check event coverage (the share of Pixel events also sent via CAPI), how quickly server events arrive, and whether browser and server copies share the same event name and event ID within 48 hours. A CAPI event match quality problem is often really a coverage or deduplication problem in disguise — fix all three, or the score will mislead you.
6. Respect consent and privacy rules in every market. Sending more customer information only helps if you have the lawful basis and consent to send it. Rules differ between the UK and EU, the United States, Canada and Australia, so your consent setup, data processing terms and parameter choices must match each market you serve. Better matching must never come from data you should not have sent.
What good looks like: a before-and-after checklist
| Check | Weak setup | Strong setup |
|---|---|---|
| Parameters sent with Purchase | IP and user agent only | Hashed email, phone, fbc, fbp, external ID where available |
| Event match quality score | Below 4, or no score visible | Steadily improving; aim for 6 or higher where possible |
| Coverage vs Pixel | Under half of Pixel events also sent via CAPI | High coverage, with deduplication confirmed |
| Freshness | Server events arriving hours late | Events arriving promptly after the action |
| Consent | All parameters sent by default | Parameters gated by consent, per market |
Do not chase a perfect 10. Some businesses, such as guest-checkout stores or lead forms that ask only for a phone number, will never send every parameter, and that is fine. The goal is a steady, honest improvement in the parameters you can legitimately send, measured over weeks, not a one-day spike. Compare your score with your own baseline, and with your cost per result, rather than with another advertiser’s score.
Cross-checking with other analytics helps here too. If Meta reports conversions that your own analytics cannot reconcile, the problem may be deduplication or attribution, not match quality alone — Google’s Analytics help on conversions is a useful second source when you compare counts across platforms.
Frequently asked questions
What is a good Meta event match quality score?
The Meta event match quality score runs out of 10 and, in Meta’s guidance, suggests aiming for 6.0 or higher where possible. Treat that as a direction, not a pass mark: a store with guest checkout may score lower for legitimate reasons, while a logged-in membership site should score higher.
Why is my CAPI event match quality low when events are firing?
Because delivery and matching are different things. Events can fire perfectly while carrying too little customer information, badly formatted information, or missing click and browser IDs. Audit the parameters actually sent, fix normalisation and hashing, and capture fbc and fbp at arrival.
Does event match quality affect ad performance directly?
It affects how well Meta can attribute conversions and optimise delivery. Better matching gives Meta a clearer learning signal, which supports more stable optimisation. It will not rescue bad creative, a broken landing page or the wrong audience — it removes a tracking ceiling so those other fixes can work.
Should I send every customer parameter I have?
No. Send the parameters you legitimately hold, with consent where required, correctly formatted and hashed. Sending data without a lawful basis creates privacy risk and will not sustainably help performance. When in doubt, send less, correctly, rather than more, carelessly.
How long does it take for event match quality to improve?
The score reflects recent data, so correctly configured events usually move it within days, but stable improvement takes a few weeks of consistent volume. Judge the trend over at least two weeks, alongside coverage and cost per result, not a single day’s score.
How AgencyAxiom does this for you
AgencyAxiom audits and fixes Meta tracking end to end: Pixel and Conversions API setup, deduplication, parameter and hashing checks, and event match quality diagnostics, with GA4 and server-side tracking checked alongside so no platform is trusted in isolation. It is the same tracking discipline behind our Meta Pixel double-counting fix and GA4 conversion discrepancy diagnostic. If your CAPI events are firing but Meta still cannot optimise, book a free consultation and ask for a free tracking audit — we will show you your real match quality, coverage and deduplication picture, and the exact fixes in priority order. You can also explore our tracking and advertising services to see the full system.