All guides

Why Meta Ads Manager conversions don't match your CRM

Compare the same action, dates, source filters and attribution settings before treating a CRM and Meta reporting gap as broken tracking.

Shahzada Ali Hassan

7 min read0 views

Twelve cobalt marbles on grey paper, with a card frame around five of them.
View full size

Get clinic leads counting again

Recover usable Lead and Purchase events on your existing domain.

$297/mo per subdomain

Start 7-day trial

Payment method required. One trial per domain. Cancel during the 7-day trial.

Get a free consultation

The short answer

Meta Ads Manager and your CRM can disagree while each counts what it was configured to count. Your CRM may record enquiries from every channel, while Meta reports conversions credited to ads under its attribution settings. Compare the same business action, dates and counting rules first. Then investigate whether events are missing from the intended source. A difference alone doesn't prove broken tracking or a data sharing restriction.

Are you counting the same conversion?

Compare completed website enquiries with website Lead events. A CRM contact, a qualified lead and a booked appointment are different actions. Keep each total separate before deciding that Meta under-reports your client's results.

Two reports, two counting rules

Same dates, same time zoneAgree the definitions before comparing totals.

CRM

Every enquiry, from every channel

  • Phone call
  • Email enquiry
  • Website form
  • Website form after an ad click
  • Referral
  • Website form after an ad view

Ads Manager

Only what the attribution settings credit to an ad

  • Website Lead, click-through
  • Website Lead, view-through
  • Phone, email, referral and organic forms are not here.

Both reports can be right. A gap alone does not prove broken tracking.

The CRM lists every enquiry from every channel. Ads Manager lists only the events its attribution settings credit to an ad. Illustrated example.

New website enquiries

What to compare
The configured website Lead event
What to exclude or explain
Phone enquiries, imported contacts and other sources

Appointments created

What to compare
A completed booking event
What to exclude or explain
Calendar opens, time selections and cancelled bookings if your CRM removes them

Unique people

What to compare
An agreed unique-person CRM count
What to exclude or explain
Repeat enquiries that an event-based report may count separately

Qualified opportunities

What to compare
The defined qualification stage
What to exclude or explain
Initial form submissions that never reached that stage
Define the business action before comparing totals.

If the client cares about booked consultations, use the form and booking checks to confirm what fires on completion. Calling every enquiry a booking hides the stage where the numbers diverge.

Most gaps close once the definitions and settings match. If you have matched them and a Meta notice names your Lead or Schedule event as restricted, skip to when Crisper fits.

How do you make the reports comparable?

Write down each report's filters before exporting it. Matching the visible date labels isn't enough if the reports use different time zones, timestamps or definitions of a conversion.

Illustration of a CRM ledger listing phone, email, website and ad enquiries beside an Ads Manager ledger listing only ad-credited rows, under one shared date strip.
Illustration. Fix the dates first, then the definitions, then compare. No numbers are shown because none would be yours.

Business action

CRM export
Form submitted, booking created or qualified opportunity
Meta report
Exact event or custom conversion

Included sources

CRM export
Website only, specified campaign tags or all enquiries
Meta report
Selected ad account, campaigns and data source

Dates and time zone

CRM export
Date field used, range and CRM time zone
Meta report
Range, ad-account time zone and report settings

Counting rule

CRM export
Unique people, submissions or booking records
Meta report
Attribution model, window and All or First conversion

Excluded records

CRM export
Tests, spam, cancellations and duplicates
Meta report
Available exclusions and filters actually applied

Export time

CRM export
Record when the export was made
Meta report
Record when the export was made
A blank reconciliation worksheet. Fill it using the actual account settings.

Keep a private list of the records you're investigating, with their completion times and an internal ID. You don't need to upload a clinic's enquiry text or treatment information to Meta to explain a reporting difference.

Which attribution settings can explain the gap?

Meta can credit an ad interaction that your CRM's source field doesn't record. Its attribution documentation distinguishes standard, incremental and custom models, which use different counting mechanisms. Inspect the model in the actual ad set.

Under standard attribution, Meta describes click-through, view-through and engage-through credit. A person can see an ad and later enquire without clicking its tagged URL. Some accounts still use earlier click definitions during rollout, so record what your account shows instead of assuming one universal window.

  1. In Ads Manager, select the campaigns and reporting dates you want to compare.
  2. Open Columns: Performance, then choose Performance and Compare attribution settings.
  3. Select the available attribution windows relevant to your comparison.
  4. Choose All conversions or First conversion where that control is available, then select Apply.
  5. Export the comparison and save its settings beside the CRM export. Explain any change in the total before changing the tracking installation.

These controls come from Meta's comparison instructions. All conversions and First conversion answer different questions. Neither option automatically means unique people in your CRM.

When does the gap point to broken tracking?

Look for a missing step in event collection before calling the difference attribution loss. A successful CRM submission establishes that the business received an enquiry. It doesn't establish that the intended Meta source received the event.

Totals change when you compare attribution windows

What it supports
Counting settings explain part of the difference
Next action
Report the chosen definition consistently

A completed submission produces no event on the intended source

What it supports
A collection or routing problem needs investigation
Next action
Use the missing-conversions diagnostic

The event arrives twice for one completion

What it supports
Duplicate senders or identifiers need inspection
Next action
Check the browser/server duplicate contract

A notice identifies the event as restricted

What it supports
Restriction scope needs a separate review
Next action
Inspect the notice and affected event

The event arrives, but ad credit is unclear

What it supports
Receipt is established; attribution is unresolved
Next action
Review campaign settings and genuine ad-driven activity
Evidence guides the next step. No percentage gap diagnoses a restriction.

For missing events, start with Facebook Ads Manager not showing conversions. After a repair, follow the verification procedure rather than treating a successful request as the finished result.

If a notice confirms restricted Lead or Schedule events on a clinic domain, Crisper's recovery workflow is relevant. It recovers usable standard events on the existing domain. It doesn't make the CRM and Ads Manager adopt the same counting rules. Read the restriction guide before choosing that path.

What should you tell the client?

Report the business totals and the advertising totals with their definitions. Say how many website enquiries the CRM recorded, what Meta credited under the selected model, and which collection checks passed. Keep purchases, leads and appointments separate.

  1. State the action and date range each number describes.
  2. Name the filters or attribution settings that explain a known difference.
  3. Separate those explained differences from events you couldn't verify.
  4. Assign the remaining investigation to the person who owns that step, then repeat the same exports after the fix.

Do not label the whole difference wasted spend or calculate recovered revenue from it. That would require evidence about what the ads caused, beyond a mismatch between two reports.

When should you use Crisper?

A mismatch between Meta and the CRM is usually a counting difference, and the steps above settle it. Crisper fits when a notice confirms a restriction and the website's Lead or Schedule event is unusable. A fix for that has to make the standard event usable again on the clinic's existing domain. Crisper recovers the standard event server-side, with no new domain, no redirect landing page and no pixel reset, as the recovery setup describes. It does not change how the CRM and Ads Manager count.

Your CRM can send its conversions through Crisper too. Connect your CRM by webhook: any CRM that can send an outbound webhook, including GoHighLevel or Salesforce, can do this. If your CRM cannot send a webhook, that route is not available. Agree with the clinic whether the conversion is an enquiry or a booked appointment, and keep that definition in the worksheet.

Scaled AI runs Meta lead-gen for addiction treatment and behavioral health centers. It has 10+ treatment-center domains recovered on Crisper, with usable Lead events on the same restricted domains. The Scaled AI case study shows the setup and the result in its own words.

Crisper does not win Meta appeals or change how Meta classifies the site, and a received event is not proof of attribution. The clinic stays responsible for what it shares with Meta. Setup takes about 5 minutes when you have DNS access, access to the Meta integration and the ability to add the tracking snippets, and the founder sets up the first client domain with you.

Set up in about 5 minutes. Events in Events Manager within about 30.

~5 min setupwithin ~30 min
Setting up
  • 7-day free trialon every eligible domain, one per domain
  • $297/moper subdomain
  • We set up the first domain with youfounder included
  • 30-day money-back guaranteeand cancel any time in the trial
  • We keep working with youuntil it is fixed

Payment method required at checkout. One trial per domain. The 30 minutes start once tracking is installed and events fire; it is visibility in Events Manager, not same-day ROAS.