CAPI Match Rate
Definition
CAPI Match Rate: CAPI match rate — surfaced by Meta as Event Match Quality (EMQ) — is how well Meta can tie your server-side Conversions API events back to a real person using the identifiers you send. The more strong match keys you include, the more conversions Meta can attribute and optimize toward.
What CAPI match rate means
When you send a conversion to Meta through the Conversions API, Meta tries to match that event to a specific person in its system. The match rate — which Meta reports as Event Match Quality (EMQ) in Events Manager — reflects how successfully it can do that, based on the customer-information parameters you include with each event. A higher match rate means Meta can attribute more of your purchases to the right people, which directly improves how well its delivery algorithm optimizes your campaigns.
This matters because every conversion Meta can't match is a conversion that effectively didn't happen as far as optimization and reporting are concerned. Low match quality starves the algorithm of the very signal it needs, which shows up as higher cost-per-acquisition even when your real-world sales are fine.
The match keys that drive it
Meta matches on the customer-information parameters in the event's user_data object. Per Meta's Customer Information Parameters documentation, the high-value identifiers (sent SHA-256 hashed, except the cookie/IP fields) include:
- em — email (the single strongest key for most ecom)
- ph — phone number (E.164 format before hashing)
- fn / ln — first and last name
- external_id — your own stable customer ID
- fbp — the Meta browser cookie (not hashed)
- fbc — the click identifier derived from
fbclid(not hashed) - client_ip_address and client_user_agent
The rule of thumb: send more keys, hashed correctly, on every event. Email plus phone plus external_id, combined with the browser fbp/fbc from the pixel, gives Meta several independent ways to find the person.
What a healthy match rate looks like
Meta scores EMQ on a 0–10 scale in Events Manager. As an illustrative reading of that scale: above roughly 7.5 is healthy, above ~8.5 is strong, and below ~6.0 is a warning sign that Meta cannot match most of your purchases. These are interpretive guardrails, not a published Meta benchmark — the number that matters is the trend on your own events after you add match keys.
Illustrative example
A store sends Purchase events server-side with only email. EMQ reads around 6. Adding phone (E.164, hashed), external_id, and passing through fbp and fbc from the browser lifts EMQ into the 8s, and the share of purchases Meta can attribute rises accordingly. Because Meta now "sees" more conversions, the delivery algorithm optimizes more accurately and cost-per-purchase tends to settle lower. The exact numbers are an illustrative scenario — the durable lesson is that match keys, not budget, were the lever.
How to improve CAPI match rate
- Send email, phone, and external_id on every event — not just Purchase, but earlier events too.
- Hash correctly. Normalize first (lowercase/trim email, E.164 phone) then SHA-256. A formatting error silently zeroes a key's value.
- Pass through fbp and fbc from the browser into the server event so cookie and click identity travel with the server-side conversion.
- Deduplicate pixel and CAPI events with a shared
event_idso the same purchase isn't double-counted. - Capture identity early (e.g. at email capture) so even pre-checkout events carry a strong key.
We break the seven highest-impact fields down in depth in the 7 CAPI fields that change match rate, and compare browser-pixel vs server-side matching in server-side CAPI vs browser pixel.
Common mistakes
- Hashing un-normalized data. "John@Email.com " and "john@email.com" hash to different values; only the normalized one matches.
- Sending keys on Purchase only. Earlier funnel events benefit from match keys too, and they help the algorithm learn.
- Trusting your server's view of match rate. Your server only knows what it sent; the EMQ number in Events Manager — scored after Meta's matching pipeline — is the one that drives optimization.
How Admaxxer improves match rate
Admaxxer ships server-side conversion tracking included, not as a separate add-on — capturing first-party identity, hashing it correctly, passing through the browser fbp/fbc, and deduplicating against the pixel automatically. It then surfaces which match keys are present on each event and how match quality is trending, so improving CAPI match rate becomes a dashboard you can act on rather than a guessing game inside Events Manager.
Related glossary terms
Continue exploring the DTC ad-analytics vocabulary — every term in this glossary cross-links to the next.
-
Ad Frequency
Ad frequency is the average number of times each unique person in your target audience saw your ad during a…
-
Attribution Window
An attribution window is the time period after an ad click or view during which a resulting conversion is…
-
Average Order Value (AOV)
Average Order Value (AOV) is total revenue divided by the number of orders in a period. It is one of the…
-
Blended ROAS
Blended ROAS is total revenue across all channels divided by total paid ad spend. By using one revenue number…
-
Cohort LTV
Cohort LTV (lifetime value) measures the cumulative revenue per customer within a specific acquisition cohort…
-
First-Click Attribution
First-click attribution assigns 100% of a conversion's credit to the first marketing touchpoint a user had…
-
Incrementality
Incrementality is the revenue a marketing channel actually caused — the conversions that would not have…
-
Linear Attribution
Linear attribution splits a conversion's credit evenly across every marketing touchpoint in the user's…
-
Marketing Efficiency Ratio (MER)
MER (Marketing Efficiency Ratio) is total revenue divided by total marketing spend across all paid channels.…
-
Meta Ad-Set Learning Phase
The learning phase is the period during which Meta's delivery system is still gathering signal on a new or…
-
Payback Period
Payback period is the number of days it takes for a customer's cumulative gross profit to equal the cost of…
-
Performance Max
Performance Max (PMax) is Google Ads' goal-based campaign type that serves across all Google inventory —…
-
Pixel-to-Conversion Discrepancy
The pixel-to-conversion discrepancy is the gap between orders reported by your storefront (Shopify,…
-
ROAS (Return on Ad Spend)
ROAS (Return on Ad Spend) is revenue generated divided by the ad spend that generated it. It is the most…
-
iOS 14.5 Attribution
iOS 14.5 (released April 2021) introduced App Tracking Transparency, requiring Apple users to explicitly opt…
Frequently Asked Questions
What is a good CAPI match rate?
As an illustrative reading of Meta's 0–10 Event Match Quality scale: above about 7.5 is healthy, above ~8.5 is strong, and below ~6.0 is a red flag that Meta can't match most of your purchases. These are interpretive guardrails, not a published Meta benchmark — watch the trend on your own events.
How do I improve CAPI match rate?
Send hashed email, phone, and external_id on every event; pass through fbp and fbc from the pixel; normalize data before hashing (lowercase email, E.164 phone, then SHA-256); and deduplicate pixel and CAPI events with a shared event_id. More correctly-formatted match keys is the single biggest lever.
What is the difference between match rate and Event Match Quality?
They refer to the same thing. 'CAPI match rate' is the informal term; 'Event Match Quality' (EMQ) is the label Meta uses in Events Manager and the 0–10 score it reports. EMQ is the authoritative number because it's measured after Meta's own matching pipeline runs.
Why does Meta show a different match rate than my server?
Your server only knows what parameters it sent. Meta's match rate is scored after its hashing and matching pipeline runs against its own user graph. The number in Events Manager is the one that drives optimization, so always trust that over any server-side estimate.
Which match key matters most?
For most ecommerce, email (em) is the strongest single identifier, followed by phone (ph) and external_id. Combining several keys plus the browser fbp/fbc gives Meta multiple independent ways to find the person, which is more robust than relying on any one key.
Does a low match rate raise my CPA?
Effectively yes. Every conversion Meta can't match is a conversion the algorithm can't learn from or attribute, so it optimizes on partial signal. That typically shows up as higher cost-per-acquisition even when your real sales are healthy — which is why match rate is worth fixing before raising budgets.