Introduction
In today’s privacy-first marketing landscape, performance marketers face growing challenges in capturing accurate data due to the decline of third-party cookies and stricter global privacy regulations. Many advertisers are turning to server-side tracking solutions like the Meta Conversion API (CAPI) to overcome these hurdles. However, simply implementing server-side tracking isn’t enough to maximize ad performance.
The real game-changer? Event Match Quality (EMQ).
What Is Event Match Quality (EMQ)?
When a user completes an action—such as clicking “Purchase” on your eCommerce site—this event data (e.g., email, phone number, or other identifiers) is sent via the Meta Pixel or Conversion API to Meta and compared with their user database. Event Match Quality (EMQ) is Meta’s scoring system to assess the quality and completeness of this user data. Measured on a scale from 1 to 10, a high EMQ score indicates that your data includes enough identifying details (e.g., email, phone number) to connect events with real users confidently.
A higher EMQ score leads to better ad targeting, accurate attribution, and campaign optimization, improving performance metrics like return on ad spend (ROAS) and cost per acquisition (CPA). Conversely, a low EMQ can cripple your ability to retarget effectively or optimize campaigns to reach the right audiences.
Figure 1. How Event Match Quality Works
How Does the Scoring System Works?
Scale: EMQ scores range from 1 (low) to 10 (high). Meta recommends aiming for a score of 6 or higher to achieve strong ad performance.
Identifiers Used: Data points like email, phone number, name, city, state, ZIP code, and advanced parameters (e.g., gender, date of birth) are considered.
Match Type: Meta primarily uses deterministic matching, where hashed user data matches exactly with data in its database. When exact matches aren’t possible, probabilistic matching—using partial data points—comes into play.
Why Is EMQ Important?
If you’re sending conversion data, you might assume you’re covered—but the quality of that data is critical. Here’s how EMQ directly impacts your advertising performance:
Improved Targeting and Optimization: Meta’s machine learning models rely on high-quality data to build lookalike audiences, identify high-intent users, and optimize ad delivery. A higher EMQ means more complete, accurate user signals feed into Meta’s machine learning models, lowering your cost per acquisition (CPA) and boosting return on ad spend (ROAS).
Better Attribution: A high EMQ allows Meta to accurately attribute conversions to the right ad campaigns and creative assets. This clarity helps you make informed decisions about your budget and strategy.
Reduced Data Loss: Low EMQ can result in unmatched events, which Meta cannot connect to real users. This means lost opportunities for optimization and remarketing. High EMQ minimizes unmatched events, giving you a more complete picture of your performance.
Competitive Advantage: While many advertisers now use CAPI or other server-side technologies, few take steps to actively improve their EMQ. Meta generally recommends an EMQ score of 6 or higher.
How Does EMQ Work in Meta’s CAPI?
The Meta Conversion API (CAPI) enables server-to-server data transfer, bypassing limitations of client-side tracking like cookie restrictions or ad blockers. To better understand how CAPI works, you can read our article about it here. Here’s how CAPI strengthens your EMQ score:
Secure, Accurate Data Transfer: With CAPI, user data is hashed (using SHA-256) and securely transmitted, reducing the risk of data corruption or blocking by client-side restrictions (like ad blockers).
Cross-Channel Attribution: By consolidating data from web, app, and offline sources, CAPI provides a holistic view of user behavior. This unified data allows Meta to match events across multiple touchpoints more effectively. When you have consistent identifiers (like an email address used both online and offline), Meta can match events more accurately.
Event Deduplication: CAPI eliminates the duplication between pixel and server-side events, preventing inflated conversion counts and preserving data accuracy. If Meta receives the same event from both your pixel (browser) and server, it must identify them as the same occurrence, maintaining data precision.
Enhanced Custom Audiences: With high-quality match data, you can create powerful retargeting lists, such as users who abandoned their carts or made a recent purchase. These audiences consistently outperform broader targeting strategies.
What are the Core Elements Affecting EMQ?
To maximize your EMQ score, focus on these key elements:
Data Consistency & Formatting: Normalize identifiers like emails and phone numbers before hashing for better matching. For example:
Convert all emails to lowercase and remove spaces.
Include country codes in phone numbers to ensure compatibility.
Hashing these parameters using Meta’s recommended method (SHA-256) also ensures data security and consistency.
Volume and Quality of Identifiers: The more data points you provide, the better your match rates. For example:
Combine emails with phone numbers, names, or ZIP codes. Multiple identifiers increase the odds of successful matches.
Leverage omnichannel data (e.g., an email from a newsletter signup plus a phone number from an in-store loyalty program).
Timestamp Accuracy: Ensure event timestamps align with the actual time of the user’s action. Misaligned timestamps can reduce match success rates.
Platform-Specific Strategies: Optimize identifiers based on platform-specific nuances. For example:
Use advertising IDs for mobile app events.
Ensure CRM data matches the format Meta expects for offline conversions.
Privacy Compliance: Always obtain user consent to collect and share data. Regulations like GDPR and CCPA require clear opt-ins, and non-compliance can lead to legal repercussions and potential data filtering by Meta if the user hasn’t opted in.
How are Events Matched to a Customer?
Now, let’s explore the mechanics of how Meta takes the event data you send and matches it to a real customer on its platform. This section is crucial, as understanding these mechanics can help you fine-tune your data approach.
Deterministic Matching (Exact Match): Deterministic matching is the primary method Meta uses to connect event data to users. This involves exact matches between the hashed identifiers you send and the hashed data already stored in Meta’s user database.
Hashed Identifiers: Once your server (or client) hashes user identifiers—like email or phone number—Meta checks if the hash matches a hashed version of the same information in its user database.
Data Normalization: Before hashing, it’s best practice to trim spaces, convert all letters to lowercase, and remove any special characters or name tags. A mismatch in formatting can result in a missed match.
Phone Normalization: Phone numbers should include the country code and be stripped of formatting symbols, like dashes or parentheses.
Probabilistic Matching (Partial or Fuzzy Match): When deterministic matching isn’t possible (e.g. if the email or phone number isn’t provided or is inaccurate), Meta can use probabilistic matching to connect data. This involves comparing multiple partial data points to make an estimated match:
Secondary Identifiers: Data like first name, last name, ZIP code, city, and date of birth help Meta make connections even if primary identifiers (email/phone) fail to match. Eg., if an email or phone is unavailable, but the first name, ZIP code, and date of birth match a user profile in Meta’s database. This could result in a successful partial match.
This approach is less common than deterministic matching but can still boost match rates if you consistently provide secondary identifiers.
Data Completeness is critical: if only a partial address is provided, the chances of a correct match drop drastically.
Cross-Device & Cross-Platform Insights
Cross-Device Login Matching: Meta knows if a single user logs into Facebook on both a desktop browser and a mobile app. When an event is sent with a mobile device ID, Meta can map that ID back to a Facebook user.
Mobile Device IDs: For mobile app events, including a user’s advertising ID (IDFA for iOS or AAID for Android) significantly improves match rates. Meta associates these IDs with user profiles when available.
Unified IDs: If your data strategy includes a single internal ID (e.g., a loyalty program ID or CRM user ID) that you also send to Meta, it can link multiple events from different channels to the same user, raising the match probability.
Lookback Windows: Timing is critical for Meta’s matching logic. Meta uses a “lookback window” to match events to user activity. This is the time frame in which Meta can link an event timestamp to a user’s historical behavior:
Event Timestamps: If a user visited your site 15 days ago and you send a purchase event with a timestamp that aligns with the user’s historical activity, Meta can match it.
Keeping this in mind, it’s essential to send events promptly with the correct timestamp or risk the event falling outside Meta’s matching window.
User Login vs. Guest Checkout:
Logged-In Users: Events tied to logged-in users often match better. For instance, if your site or app requires a user to log in via Facebook or with the same email they use on Facebook, the event data becomes inherently more matchable.
Guest Checkouts: Events tied to guests can pose a challenge if the user enters minimal or erroneous data. Implementing form validation can help reduce typos or incomplete fields.
By aligning your data collection practices with these matching mechanisms, you can substantially increase your EMQ score. The key is to understand how Meta cross-references the data points you send and to supply those data points in the most complete, normalized, and hashed manner possible.
Common Challenges and Solutions
Even with a solid strategy, pitfalls can arise. Here’s how to address them:
User Privacy & Consent Issues
As regulations like GDPR and CCPA evolve, obtaining explicit user consent becomes paramount. Without consent, user data cannot be legally or ethically used for matching.
In some cases, failing to gather consent results in data being blocked or removed from your feed before it even reaches Meta, lowering EMQ.
Incomplete Data
Problem: Missing or partial identifiers reduce match rates. A single piece of data—like an email—may not suffice for accurate matching, especially if the user uses multiple addresses.
Solution: Capture multiple data points through different mediums and channels like forms, CRMs, and loyalty programs.
Formatting Errors
Problem: Typos or inconsistent formatting (e.g., missing country codes) can cause big drops in match rates. For instance, missing the country code in phone numbers or having variations like
gmail.con(typo) effectively kills matching opportunities.Solution: Use validation tools to clean and normalize data both when collecting and before submitting to Meta.
Technical Misconfigurations
Problem: Duplication or incorrect parameters in your CAPI setup.
Solution: Regularly test events using Meta’s debugging tools. Using Meta’s Test Events tool can help you confirm if your server is sending the right parameters in real-time.
Lack of Data Unification
Problem: Relying on a single data source, such as your website pixel or email lists, limits insights.
Solution: Combining multiple data sources (in-store, mobile app, offline CRM) can drastically improve match rates, but it requires thoughtful setup and data unification.
Conclusion
Optimizing your Event Match Quality (EMQ) is essential for maximizing ad performance on Meta’s platform. By improving your data collection, leveraging Meta’s CAPI, and maintaining compliance with privacy regulations, you can achieve higher EMQ scores and unlock better targeting, attribution, and optimization.
Where to go from here?
Perform an EMQ Audit: Use Meta’s Events Manager to see your current EMQ score, identify areas for improvement, and track changes over time.
Refine Data Collection: Ensure your forms, CRM, and checkout processes capture detailed, clean user identifiers.
Implement or Optimize CAPI: Confirm your server-side events are set up correctly, hashed, and not duplicated.
Monitor Compliance: Stay updated on regulations like GDPR/CCPA and include user consent checks to maintain ethical and legal data practices.
Set Targets: Aim for an EMQ of 6 or higher as a starting goal. Continually refine your matching process to push into the 7+ range for even better ad performance.
Abhimanyu Atri
Marketing Associate
