# Meta CAPI & Pixel Checker

> Free self-audit of a Meta Pixel and Conversions API setup: 14 yes/no/not-sure questions, a weighted score out of 100 and a list of likely gaps, each linked to a fix.

Source: https://sterling.company/meta-capi-pixel-checker · Machine-readable spec: https://sterling.company/meta-capi-pixel-checker/spec.json · Publisher: Sterling Lab

## Short answer

The Meta CAPI & Pixel Checker is a free self-audit of a Meta Pixel and Conversions API setup. You answer 14 yes, no or not-sure questions and get a score out of 100 and a list of likely gaps, biggest first, each linked to a fix. It does not scan your website or read your Meta account, and your individual answers are not sent anywhere.

Nothing is scanned and nothing is read from Meta: the checker only scores the answers you give. The weights and band cut-offs are our own definitions, not a benchmark.

## How the score works

Each check has a weight of 1, 2 or 3. Yes earns the full weight, Not sure earns half, No earns nothing. A check marked not applicable is left out of both sides of the sum.

**score = round((sum of weights answered yes + 0.5 × sum of weights answered not sure) ÷ sum of weights of the applicable checks × 100). A check answered na is left out of both sums.**

The checks with weight 3 are core checks. A no or not sure on any of them keeps the band out of solid, even when the score is 80 or higher.

The fix list is sorted by points lost, biggest first: a No loses the full weight of the check, a Not sure loses half of it. Checks that lose the same number of points keep the order in which they are listed below.

## Bands

- **80 to 100**: Solid on paper. The remaining items are tidy-up work. Next step: Fix the remaining items when convenient, then re-run the check after any change to tags, plugins, consent or the server.
- **50 to 79**: Gaps worth fixing. Some of what Meta merges and optimizes on is partial or guessed. A failed or unconfirmed core check holds you here even with a score of 80 or more. Next step: Start with the first item in the fix list. If a core check is named, fix that one before anything else.
- **0 to 49**: Hard to trust. Fix the top items before you change bids or budgets. Next step: Do not change campaign budgets or optimization events yet. Fix the top three items, then run the check again.

## Worked example

An online shop whose pixel and server both send events and that needs no cookie banner: the cookie banner check (consent) is not applicable. The server generates its own event ID, so "Same event name and event ID" is No. Event Match Quality has not been looked at, customer-detail hashing is unknown and the domain verification is unknown, so those three are Not sure. Everything else is Yes.

1. Applicable weight: 27 − 1 = 26
2. Points lost: 3 + 1 + 1 + 0.5 = 5.5, so earned is 26 − 5.5 = 20.5
3. Score: 20.5 ÷ 26 × 100 = 78.8, shown as 79
4. Band: 79 is in 50 to 79, so gaps worth fixing
5. Gaps in order: Same event name and event ID (3 points lost); Event Match Quality checked per event (1); Customer details hashed before sending (1); Domain verified in Meta Business (0.5)

Steps to compute any result: remove not-applicable checks from the total; lose the full weight for each No and half for each Not sure; earned = applicable weight − points lost; score = round(earned ÷ applicable weight × 100); then apply the band and core rules; sort the gaps by points lost, biggest first.

## The 14 checks (27 points in total)

### Events reach Meta

**Events sent by pixel and by CAPI** (weight 3, core)

- Question: Your key events (a purchase, a lead) reach Meta from the browser pixel and also from a server through the Conversions API.
- How to check: Events Manager → your dataset → Overview: each key event should list both Browser and Server as sources.
- If No or Not sure: Meta may see only what the browser lets through. Browsers block scripts, drop cookies and close tabs before a request finishes. Sending the same event from a server as well fills some of those gaps. An event that comes from one source only has nothing to be merged with.
- Fix: [Meta Conversions API setup](https://sterling.company/guides/meta-conversions-api-setup)

**One source per event** (weight 2)

- Question: Each event has one pixel and one server source. A store app plus a Tag Manager pixel, or two CAPI plugins, are not both sending the same purchase.
- How to check: Open your confirmation page with the Meta Pixel Helper extension: one Purchase (or Lead) fire is expected. Then list every app, plugin and container that sends events to the dataset.
- If No or Not sure: The same conversion may arrive from several sources. Our dedup article describes a store with its native Meta app and another CAPI source both running: several streams for one purchase and no shared event ID. Pick one source of truth per event.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**A test conversion seen in Test Events** (weight 1)

- Question: After the last change to tags, plugins or the server, someone completed a test conversion and saw it in Test Events as a Browser row and a Server row.
- How to check: Events Manager → Test Events: open your test URL, complete a purchase or a lead form, and look for two rows for the same conversion, one labelled Browser and one labelled Server.
- If No or Not sure: Nobody has seen both copies of an event arrive. Events that show green in Events Manager have not been tested end to end. If only one side appears in Test Events, one of the two sources is not firing at all, and deduplication is not your first problem.
- Fix: [Conversions API and the duplicate-event trap](https://sterling.company/guides/facebook-conversion-api-gtm-server-side)

### Deduplication and value

**Same event name and event ID** (weight 3, core)

- Question: For each event the pixel and the server send the same event name and the same event ID, so Meta can merge the two copies into one.
- How to check: Test Events: expand the Browser row and the Server row for the same conversion and compare the Event ID field character for character. A missing or different ID means the copies are not merged.
- If No or Not sure: Meta may count each conversion twice, or drop one copy. Without a shared event ID Meta falls back to guessing from customer data. Our dedup article describes the result: the two copies are not merged, conversions look inflated or one copy is thrown away, and optimization learns from unreliable numbers.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**Event ID passed in the right field** (weight 2)

- Question: The event ID is born once, on one side, and handed to both channels: as eventID in the pixel call and as the top-level event_id field of each server event, not inside user_data or custom_data.
- How to check: In your tag template or code, find where the ID is generated and where each channel reads it. Test Events shows an event ID on the server row only when Meta read the field.
- If No or Not sure: The event ID may be present but in a place Meta does not read. The pixel names it eventID, the Conversions API names it event_id, and every layer between your site, Tag Manager and server has to keep one string. If the browser and the server each generate their own, the two never match.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**Value and currency on Purchase and Lead** (weight 3, core)

- Question: Purchase events carry the real order value and currency (a lead event carries the value you assigned to a lead), on the pixel and on the server copy.
- How to check: Test Events: open a test purchase and compare the value and currency parameters of both rows with the real order.
- If No or Not sure: Conversion value is missing or wrong. Without a real value and currency, value optimization and any revenue figure in Ads Manager rest on empty or invented numbers. Check the value on both copies of the event.
- Fix: [Facebook ads not converting: find the leak](https://sterling.company/guides/facebook-ads-not-converting)

**Server events sent soon after the action** (weight 1)

- Question: Server events are sent soon after the action happens, not in a late batch, and you know how long the delay is.
- How to check: Ask whoever runs the integration when the server event is sent: at once from the order or form, or later from a queue or an export. Compare the event time in Test Events with the time of your test.
- If No or Not sure: A late server event may miss its browser twin. Meta merges two copies of an event only when they arrive within the window it uses for matching duplicates. A server event sent much later may arrive as a second, separate conversion.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

### Matching quality

**Event Match Quality checked per event** (weight 2)

- Question: You know the current Event Match Quality of each key event and have looked at which customer details are missing when it is low.
- How to check: Events Manager → Overview → open the event: Event Match Quality is shown per event, with the parameters that would improve it.
- If No or Not sure: Nobody is watching how well server events are matched to people. Event Match Quality is Meta's own gauge of how well a server event can be tied to a real user. A thin set of customer details leaves a good server event hard to attribute.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**fbp and fbc sent with server events** (weight 2)

- Question: The server events include the browser’s _fbp and _fbc cookie values in user_data.
- How to check: Test Events → expand the Server row → user_data: fbp and fbc should be present as strings, not empty.
- If No or Not sure: Server events may lack the click and browser IDs. Our dedup article explains that fbc is the Facebook click ID and fbp the browser ID. Without fbc the server event cannot confirm the click that began the session, and without fbp Meta loses linkage between visits.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**Customer details hashed before sending** (weight 2)

- Question: Customer details sent through the Conversions API (email, phone, name, city, zip) are hashed with SHA-256, in lower case, before they leave your server.
- How to check: Test Events → expand the Server row → user_data: hashed values look like long opaque strings. Plain text, or a warning icon, means a field is not hashed.
- If No or Not sure: Customer details may be sent unhashed, or not matched. Our dedup article notes that Meta expects hashed customer details and that an event can arrive while its enrichment fails. Mainstream libraries hash for you; custom code often skips it.
- Fix: [Meta CAPI installed, attribution still broken: the dedup key](https://sterling.company/blog/meta-capi-pixel-dedup-2026)

**Cookie banner and pixel agree** (weight 1)

- Question: Your cookie banner and the pixel agree: the pixel starts after consent is given, and you know what the server sends for visitors who never use the banner.
- How to check: Open your site in a clean browser, accept the banner and check that the pixel fires after Accept. Then ask who owns the server side what it sends when there was no banner choice.
- Not applicable when: No banner needed
- If No or Not sure: The pixel and the server may follow different consent rules. Our dedup article describes a consent bridge that holds back the pixel while the server event fires unconditionally, so Meta sees only one half of each conversion. The consent article covers the same pattern for Google.
- Fix: [Consent Mode v2 and missing GA4 conversions](https://sterling.company/blog/consent-mode-v2-ga4-attribution-gap)

### Settings and upkeep

**Domain verified in Meta Business** (weight 1)

- Question: The domain your events come from is verified in your Meta business account.
- How to check: Meta Business settings → Brand safety → Domains: your website domain should be listed and verified.
- If No or Not sure: The domain may not be verified. Domain verification is a Meta setting that ties your site to your business account. If it is missing or unconfirmed, open the setting and verify it before you rely on the event setup below.
- Fix: [Meta Ads service: CAPI deduplication and account setup](https://sterling.company/services/meta-ads)

**Events prioritized in Aggregated Event Measurement** (weight 2)

- Question: You have opened the Aggregated Event Measurement settings for your domain and the events you optimize for are in the priority list, with the most important one first.
- How to check: Events Manager → Aggregated Event Measurement (or Web Event Configuration): compare the list and its order with the conversions your campaigns optimize for.
- If No or Not sure: The event priority list may not match your campaigns. Aggregated Event Measurement is Meta’s setting for which events from a domain are available and in what order. If nobody has compared the list with the conversion your campaigns optimize for, the campaign may be aimed at an event that is not configured.
- Fix: [Meta Ads service: CAPI deduplication and account setup](https://sterling.company/services/meta-ads)

**Monthly check against real orders** (weight 2)

- Question: In the last month someone compared the purchases or leads Ads Manager reports with real orders or leads in your shop, CRM or inbox.
- How to check: Compare last month’s Meta-reported conversions with your own records. A steady gap is normal; a sudden drop, or a number that stops moving, means something broke.
- If No or Not sure: Meta numbers are not checked against real orders. Our dedup article starts from accounts where Ads Manager reports fewer purchases than the shop or the CRM. A regular comparison with real orders is what shows that a setup has drifted.
- Fix: [Server-side tracking service](https://sterling.company/services/server-side-tracking)

## FAQ

**Does the Meta CAPI and Pixel checker connect to my Meta account?**

No. It asks questions and scores your answers. Nothing is loaded from your site, your pixel or your Meta account, so the result is only as good as the answers. Each question has a short note on where to look in Events Manager.

**Why does “Not sure” count as half?**

An unknown is not a confirmed failure, but it is not a pass either: nobody is watching that check. Half the weight keeps it visible in the score and on your list.

**Is the score a benchmark against other advertisers?**

No. The weights and the 80 and 50 thresholds are our own definitions, chosen to rank what matters most for deduplication and attribution. They are not drawn from industry data, and the score does not compare you with anyone.

**What happens to my answers?**

They stay in your browser tab, are not sent anywhere, and disappear when you leave or start over. Our site analytics receives only an aggregate: that the check was started and finished, with the final score, the band and the number of gaps, and the score again if you click the audit link. The individual answers are never sent.

**Why can a high score still stay in the middle band?**

The three checks with weight 3 are the events reaching Meta from both the pixel and the server, the shared event ID, and the value and currency. A No or a Not sure on any of them keeps the result out of the top band, even when the score is 80 or higher, because the rest of the setup cannot make up for events that are not merged or carry no value. This rule is our own definition, like the weights and thresholds.

**Where do the questions come from?**

From our published article on the Pixel and Conversions API event ID, our Meta Ads and server-side tracking service pages, our Conversions API guides, and Meta’s own terms: Conversions API, event ID, Event Match Quality, Aggregated Event Measurement and Test Events. Where our site gives no fact, the check is worded as an action to carry out, not as a statistic.

**What does the free audit add on top of this?**

We look inside your Meta dataset and the rest of your tracking and test the same points directly, instead of relying on answers, then send a written report with the fixes in order.

## Next step

If you want the same points tested inside your Meta dataset and the rest of your tracking, request the [free conversion tracking audit](https://sterling.company/services/conversion-tracking-audit).
