Meta Event Match Quality Score Explained
👉 7.1/10 Facebook Event Match Quality score can actually be better than 9.1/10 Event Match Quality Score.
👉 Event Match Quality Score seems to be an assassment, an average advertisers get with similar setups—not the real accuracy of your data.
👉 Meta says about Event Match Quality “Note that this metric is in development.” which may explain the discrepancy.
👉 My testing indicates you can get a higher Event Match Quality score by sending crappy fake data, if you send more of it. Do not blindly trust random upwork.com freelancers if they promise high EMQ scores.
Update 22. May 2025
A picture is worth a thousand words. Notice how a higher Event Match Quality score does not automatically guarantee more additional conversions reported. Both ‘Complete registration’ and ‘Submit application’ are the same Acuity Scheduling conversion, AB-tested.
For the last four weeks I've been A/B testing different conversion tracking methods for Acuity Scheduling conversion tracking, to figure out which method is the most accurate and has the highest boost in ad performance and return on investment.
In the screenshot below most of the conversions are actually the same conversion— Acuity scheduling booking. So even though the action name is different like Submit Application or Complete Registration, Lead or Subscribe, each of them is actually the same Acuity conversion. I'm just using built-in standard event names.
This has taught me a totally new perspective on conversion tracking. It's been interesting to compare different methods: client-side tracking, server-side tracking, hybrid methods, third-party apps, etc.
So here is the biggest takeaway from testing.
Higher Meta Event Match quality score does not mean more accurate conversion tracking
What do I mean by this? Many people say and have experiences where they increase the event match quality and the ads take off like wildfire, more conversions, cheaper ads, etc.
Most of the time this is true. However, digging in to how the event match quality is calculated and running these tests, I now have a much deeper understanding of what actually works,what is happening and what does the event match quality score actually mean.
The thing I was most curious about was—
Does event match quality score mean Meta is able to match accounts to conversions?
Apparently that is not the case.
Event match quality seems to be more of a benchmark or an average or an assessment. So if we send multiple data points like user agent and IP address, meta will give us a higher event match quality score. Apparently regardless of how well these parameters help meta match users and track conversions.
You can see that a tracking method with a lower event match quality score may still track more conversions. I would definitely trust the higher number of conversions over the event match quality score.
If we compare the actual numbers of conversions, again here the different names all use the same conversion goal, which is the Acuity booking. We can see that the event match quality does not directly correlate with how many conversions we're able to track!
One of the platforms I used has a bug where it is sending user agent and IP addresses server side to Meta, but it's not using the original real IP address of the visitor who converted, but the intermediary IP address of the server.
This bug was a real breakthrough, because this “fake user data” resulted in a very high event match quality score. But these IPs are not relevant at all. They’re totally random and cannot be in any way used to match real users to the ad clicks and conversions. Comparing the event match quality score with the number of actual conversions, knowing that some of this data is "fake", I now treat event match quality just as an indicator, but not the final end result.
In this reddit post we can read amazing results:
When you improve your EMQ score from 8.6 to 9.3, here's what we consistently see:
Cost per acquisition drops by 18% ($42 → $35)
Customer match rates increase by 24%
ROAS improves by 22% on average
Ad spend efficiency increases by ~$2,100/month at $1,000 daily spend
And all this makes sense assuming that the event match quality score is actually real and the data is high quality. But it's important to keep in mind that the event match quality doesn't actually measure how good your data is. It seems to mainly give a rough assessment of what on average happens with others who are using these data points.
So if you send IP, address, email, phone number, first name, last name, it calculates some quality score And if you use another combination, then you get another quality score. Apparently—this is just speculation based on my tests.
But again, this does not reflect how accurate this conversion tracking method is. For that, you actually need to A/B test using a real campaign and comparing the actual numbers of conversions.
If Meta CAPI by itself has higher Event match quality score, should I stop using the Meta Pixel entirely?
This was another thing where my hypothesis differentiated from the end result.
After digging in, it seemed that the Meta systems by documentation and public information do not merge deduplicated event information. So they seem to say that they use whichever event they get first. Pixel is probably faster even if the Conversions API (CAPI) has more data points and higher quality data.
However, in my testing the hybrid method of using Meta Pixel and Meta conversions API togethe as recommended outperformed the pure server-side tracking by a small margin. This leads me to suspect thatMeta is actually a little smarter under the hood than what they tell us. I think it's likely Meta actually merges and combines deduplicated event data somehow. Otherwise, I don't think these results would make sense because thePixel event arrives instantly. The server side takes longer, but the Pixel event in my testing was missing phone number, email, and a lot of other data points.
Looking to get higher Meta event match quality scores and track more conversions?
We go deep. We test. We deliver. And we’d love to help.
Get in touch here