---
title: "How to improve Meta Event Match Quality (EMQ)"
author: "Performetic Ekibi"
url: "https://www.performetic.com/en/blog/how-to-improve-meta-event-match-quality"
published: "2026-11-15T12:00:00.000Z"
updated: "2026-11-15T12:00:20.038Z"
---

# How to improve Meta Event Match Quality (EMQ)

> Meta Event Match Quality (EMQ) scores, out of 10, how well a server event can be matched to a Meta account. To raise it, send email, phone, name, IP address, user agent, fbp and fbc through the Conversions API in the correct format, send events in real time and share the same event_id with the Pixel.

## How do you improve Meta Event Match Quality (EMQ)?

To improve Meta Event Match Quality, add more customer information, correctly formatted, to your Conversions API events: email, phone, first and last name, IP address, user agent, fbp and fbc. Send events in real time, use the same event_id as your Pixel and fix the Purchase event before anything else.

This guide covers what EMQ measures, why ecommerce stores so often score low and the order in which to fix things. If you have not set up the Conversions API yet, start with our post on [server-side tracking and the Conversions API](/en/blog/server-side-tracking-conversions-api); EMQ is the step after that.

## What is Event Match Quality?

Event Match Quality is a score out of 10 that Meta assigns to each event you send from your server through the Conversions API. It reflects how well the customer information in that event can match it to a Meta account. Meta's developer documentation is direct about why this matters: only matched events can be used for attribution and ad delivery optimization. The score is currently shown for web events only.

In practice, a Purchase event with weak EMQ tells Meta "someone bought, but we are not sure who." The algorithm then looks for similar buyers with a weaker signal. Part of the gap between purchases in Ads Manager and orders in your store dashboard comes from exactly this.

## Why is your EMQ score low?

These are the causes we see most in account audits:

- **Pixel only, or a partial Conversions API setup.** Browser blocking and cookie consent mean some events never arrive.
- **Server events carry only IP and user agent.** Meta can treat events with very broad parameter combinations as invalid.
- **Phone numbers in the wrong format.** Numbers are often stored with a leading zero and no country code.
- **fbc gets lost.** The fbclid parameter from an ad click is dropped somewhere in a redirect.
- **Events are batched and delayed.** Orders uploaded hours later weaken both matching and optimization.

## Which parameters affect EMQ the most?

Meta lists email, IP address, first and last name and phone among its high-quality parameters. The table shows where each one comes from and how it should be sent:

| Parameter | Source | Hashed | Format note |
|---|---|---|---|
| em (email) | Account, checkout form | SHA256 | Trim spaces, lowercase |
| ph (phone) | Checkout form | SHA256 | Remove symbols and leading zero, add country code |
| fn, ln (name) | Checkout form | SHA256 | Lowercase, no punctuation, UTF-8 |
| external_id | Customer ID | Recommended | Same format across channels |
| client_ip_address | Server request | No | The visitor's real IP, not your server's |
| client_user_agent | Browser | No | Send as is |
| fbp | _fbp cookie | No | Refresh the value regularly |
| fbc | _fbc cookie or fbclid | No | fb.1.timestamp.fbclid format |

Two notes for stores selling in Türkiye. Phone first: a number stored as "0532 123 45 67" should become "905321234567" before hashing (example number). Then Turkish characters: in some programming languages a capital "İ" picks up an extra dot mark when lowercased. Test the output with a few real names before you hash.

## How do you raise EMQ step by step?

This order gives the most impact for the least effort. Do not try to do everything in one week; check how the score moves after each step.

1. **Start with Purchase.** If your campaigns optimize for purchases, this is the most valuable event and the one where you hold the most customer data.
2. **Add email and phone.** You already collect both at checkout. Normalize them, hash with SHA256 and send them.
3. **Add first name, last name, city and postcode.** The shipping details you already ask for add extra matching power.
4. **Move fbp and fbc to the server.** Store the cookie values with the order and test that fbclid survives every redirect on your site.
5. **Send the visitor's real IP and user agent.** If your server or CDN IP is sent instead, the field is useless.
6. **Add external_id for logged-in customers.** Use the same ID, in the same format, on the Pixel side too.
7. **Send events in real time.** Meta recommends sending events as they happen, or in batches close to real time.
8. **Check deduplication.** If you send the same event through Pixel and the Conversions API, the event name and event_id must match exactly.
9. **Then move up the funnel.** Add whatever customer data you have to InitiateCheckout and AddToCart.

## Why are ViewContent and AddToCart scores always lower?

Because most visitors are anonymous at that stage. They have not given an email or phone number yet, so you only have IP, user agent, fbp and possibly fbc. Expecting Purchase-level scores on these events is unrealistic. Adding external_id and a stored email for logged-in visitors helps, but put most of your effort into purchase and checkout events.

## Where do you check your EMQ score?

You can see it per event in Events Manager, in the event list of the relevant dataset. The same view shows how often each customer information parameter is being sent, which is the quickest way to spot the field that is dragging the score down. After a change, give the score a few days to settle before drawing conclusions.

## What about privacy and consent?

Even hashed, email and phone are personal data shared with an ad platform. Check with your legal adviser that your privacy notice and cookie consent flow cover this sharing; in Türkiye that means KVKK as well. This is not legal advice, but no developer should ship the setup without asking the question.

If you want tracking and campaign structure handled together, our [performance marketing](/en/services/performance-marketing) page explains how we work. To review your current EMQ scores and missing events with us, request a [free growth audit](/en/contact) from the Performetic team.

## Key takeaways

- EMQ scores out of 10 how well server events match Meta accounts; unmatched events do not help optimization.
- The biggest gain comes from adding correctly formatted email and phone to the Purchase event.
- Format phone numbers with the country code and test how your code lowercases non-English characters.
- Move fbp and fbc to the server, send the visitor's real IP and send events in real time.
- Deduplicate Pixel and Conversions API events with the same event name and event_id.

## FAQ

### What is a good Meta EMQ score?

Meta does not publish an official target; it only says that higher match quality is better. In practice, a Purchase event that carries correctly formatted email and phone tends to land in the upper range. Lower scores on upper-funnel events like ViewContent are normal, because most visitors have not shared personal details at that point.

### Can I improve EMQ with the Pixel alone?

The EMQ score is calculated for server events sent through the Conversions API. Advanced matching on the Pixel helps matching, but it cannot recover events lost to browser blocking and cookie consent. For a lasting improvement, run the Pixel and the Conversions API together and deduplicate them with the same event_id.

### How should phone numbers be formatted for Meta?

Remove spaces, brackets, dashes and other symbols, drop the leading zero and add the country code. A Turkish mobile number, for example, becomes a 12-digit string starting with 90. After normalizing it this way, hash the value with SHA256 and send it in the ph parameter.

### Will a higher EMQ improve my ad results?

When more events match, Meta has more data for attribution and optimization, and reported conversions move closer to reality. That is not a guarantee of better results, since creative, bidding and product fit still decide a lot. But expecting strong optimization from a weak signal is unrealistic.

## Sources

- [Meta for Developers: Conversions API best practices](https://developers.facebook.com/docs/marketing-api/conversions-api/best-practices)
- [Meta for Developers: Customer information parameters](https://developers.facebook.com/docs/marketing-api/conversions-api/parameters/customer-information-parameters)
