About the Integration
The native analytics integration (aka. Attribution Passback) between Triple Whale and Meta is built to allow advertisers to optimize their Meta ads leveraging their first-party attribution data from Triple Whale.
This integration powers Meta Custom Attribution, Meta's new optimization model that uses passed-back attribution data from measurement partners to fine-tune campaign delivery. Triple Whale is one of Meta's launch measurement partners for Custom Attribution.
Availability
As of May 14, 2026, the integration is in Open Beta.
Meta is rolling out Custom Attribution access to eligible Meta-managed advertisers in waves, with broader general availability expected over the coming months.
Eligibility:
Advertisers with a dedicated Meta account manager, sales rep, or client partner
Triple Whale brands on a paid plan with access to Sonar Optimize
If you do not have a dedicated Meta contact today, talk to your Triple Whale CSM, Meta is rolling out Custom Attribution in waves, and we can flag your account for the next wave.
What you get by enabling Attribution Passback
Sharper Meta optimization as Custom Attribution learns from your trusted attribution data over time.
Early access to new Meta ad system updates (see Meta's blog post for context).
Influence on the integration by enabling early, you help Triple Whale and Meta refine the product as it scales.
Setup Instructions
Setting up the integration is straight-forward and should only take a few minutes.
1. Connect Meta to Triple Whale
Log in to Triple Whale
Navigate to your Triple Whale account at app.triplewhale.com.
Ensure that your Meta Dataset is connected to Triple Whale
Head over to Data > Integrations
Find your Meta integration
Ensure that your primary Meta integration is connected. If not, reconnect Meta by clicking “Connect” and completing the OAuth flow.
2. Go to Sonar Settings
The attribution passback integration is a subset of the Triple Whale Meta Sonar (CAPI) Integration. If you already have Sonar configured for CAPI, it will require a small additional configuration to send the additional events.
NOTE - This integration does NOT require you to configure the main CAPI integration (standard events) in order to send attribution data to Meta.
Go to Data > Sonar Enrichment
Click Integrations tab > in Triple Whale and locate the Meta integration.
Note - You do NOT need to configure the standard Sonar CAPI integration in order to use the Passback Integration
Click Complete Setup
Follow the setup wizard instructions
Note - On step 1 of the configuration, if you receive a warning because you do not have the Triple Pixel Theme App Embed or Web Pixel extension, you can skip this step. This integration does NOT require you to have the pixel installed via these methods. Simply click Save and Continue to move to the next step.
On step 1 of the wizard, select the primary Meta dataset for your Business from the dropdown
On step 2:
If you wish to skip sending Standard Meta events, simply uncheck the boxes next to each event
Note: Avoiding Conflicts with non-Sonar CAPI Integrations
If you are running an existing CAPI integration through another provider, be sure to leave the standard events unchecked to avoid sending duplicate standard events through Sonar.
On step 3 of the wizard, you will find Attribution Metadata section
Configure your preferred settings for optimization:
Model - the attribution model you use for making optimization decisions
Recommended: First Click, Last Click, or Linear All/Linear Paid, these have shown the strongest lift with Meta Custom Attribution.
Triple Attribution is not recommended. Triple Attribution closely aligns with Meta's own attribution model, so brands using it as their source of truth are unlikely to see a lift in ROAS from Custom Attribution.Note: Total Impact is not currently supported.
Window - the attribution window you use for making optimization decisions
Performance goal - the outcome you would like to optimize for
Enable the toggle to enable Attribution Passback
On step 4, configure any purchase exclusions (purchase events you don’t want sent to Meta for campaign optimization):
For instance, if you have PoS or $0 orders, you may wish to exclude them. You can do so by configuring order tag exclusions
If you have recurring subscriptions, you can also exclude these in the order tag exclusions
Once you are finished configuring exclusions, click Save and Enable.
Final setup steps
Verify Connection
Once connected, Sonar will begin sending Attribution Events to Meta
In the Sonar Delivery Overview tab you will see event counts for the “UpdatePurchase” event once you have had some purchases occur that are attributed to Meta
Approve the new Custom Events in your Meta Event Manager
Navigate to your Meta Event Manager Dashboard
Once you have received some events, you will see a new custom events called “UpdatePurchase”
You will be asked to approve these events. Approve these events in your Meta event manager.
Before testing: Allow ~14 days of attribution data to accumulate before running an A/B test or applying this to a live campaign. Meta's model needs this window to learn from your data.
Congratulations, you are now successfully set up with the Sonar <> Meta analytics integration.
FAQ
Will this integration interfere with my existing Meta CAPI provider?
No. You are safe to run this integration alongside any other CAPI integration. While this integration uses Sonar's CAPI connection to send attribution data to Meta, the Attribution events sent to Meta are unique to this integration (labeled Update Purchase events) and will not interfere with standard conversion events you may be sending through any other existing CAPI integration. There is no risk of de-duplication errors for your primary conversions or events.
This integration is configured on the same UI/screen as Sonar's standard CAPI integration. To avoid conflict with another CAPI solution, do not check the boxes to send the standard conversion events through Sonar (see the callout in the setup instructions above).
Will this integration impact my existing campaigns/Meta performance?
Not immediately. Configuring this integration will have no immediate impact on existing campaigns or Meta performance.
The integration sends data to Meta to power Custom Attribution.
Once Custom Attribution is live on your account, you will be able to choose which campaigns use this data for optimization when configuring your campaigns in Meta.
How do I know the integration is configured correctly?
If configured correctly:
You should see the Attribution Metadata setting toggled on AND the Meta Sonar integration toggled ON
You will begin to see new custom events in your Meta events manager called "Update Purchase" events.
You can check your Meta Sonar dashboard and should see Attribution Passback events in the event monitoring table
Note, you will only be able to confirm the integration is sending the attributed order data after enabling the integration AND after an attributed purchase has occurred
Why do I see so many more Attribution Passback events than orders?
While this integration is enabled, Triple Whale will send one event for every supported attribution model, for each order.
How do I use this data in Meta?
Once Custom Attribution is live on your account in Meta, you will be able to choose which campaigns use this data for optimization when configuring your campaigns in Meta.
Meta is rolling Custom Attribution out in waves to eligible Meta-managed advertisers, with broader general availability expected over the coming months.
How long before I can test this in an A/B test or on a campaign?
Plan for a ~14-day data collection window after enabling Attribution Passback before you run an A/B test or apply Custom Attribution to a campaign. Meta needs enough passed-back attribution data to learn from before it can meaningfully influence delivery. Testing before this window is complete may produce unreliable or inconclusive results.
Why isn't Triple Attribution recommended for this integration?
Custom Attribution works best when you feed Meta a signal that differs from what it already uses. Triple Attribution closely aligns with Meta's own attribution model, so brands using it as a source of truth may not see a lift in ROAS. First Click, Last Click, or Linear give the model a more distinct signal to optimize against.
Testing Custom Attribution: How to Run an A/B Test
Once your Attribution Passback integration is live and sending events, the best way to see the impact of Custom Attribution is to run a simple A/B test, a Custom Attribution campaign against a business-as-usual (BAU) campaign, and compare the results in Triple Whale.
The goal is straightforward: confirm whether optimizing Meta delivery against your Triple Whale attribution data outperforms Meta's standard, in-platform attribution. Because attribution methodology and data timing differ between the two systems, always measure the winner in Triple Whale, not in Meta Ads Manager. Triple Whale is your source of truth for cross-publisher performance and budgeting decisions.
Before you begin, make sure:
Your Attribution Passback integration is set up and enabled (see the setup steps above).
Meta has received roughly 2 weeks of high-quality signal and your data quality shows High (see Confirm your data is ready below).
You can run the test for at least 2 weeks (the test can run in any region).
Confirm your data is ready
Meta needs about two weeks of high-quality Custom Attribution signal before the feature becomes available for campaign setup. You'll get an email when it's ready, but you can check status anytime:
In Meta Events Manager, click Datasets in the left menu.
Select the dataset receiving your purchase events.
Open the Custom attribution source attached to those events.
Review the custom attribution quality page for connection status, active issues, and parameter coverage.
Confirm the connection quality indicator shows High before you continue.
Once quality is High, the Custom attribution option unlocks in Ads Manager.
How the test is structured
You'll run a 2-cell test: one Test campaign using Custom Attribution and one BAU campaign using Meta's standard attribution. To keep the comparison clean:
Creatives and targeting must match across both campaigns.
Once the test is live, make no manual changes to either campaign for the full duration. If something has to change, alert your Meta contact first.
Use a consistent naming convention so both campaigns are easy to identify:
Test cell:
Meta PBA test – TEST {campaign name}BAU cell:
Meta PBA test – BAU {campaign name}
Step 1: Build your Test and BAU campaigns
In Meta Ads Manager, create two Sales-objective campaigns, one Test, one BAU. The easiest way is to duplicate an existing reference ad set into each new campaign so the setups are identical.
BAU (control): Create a new Sales objective campaign and duplicate your reference ad set into it. Name it
Meta PBA test – BAU {campaign name}.Test: Repeat the same step and name it
Meta PBA test – TEST {campaign name}.Configure both campaigns to meet these eligibility criteria:
Campaign objective: Sales
Budget strategy: Campaign budget
Bid strategy: Highest volume or Highest value
Conversion location: Website (or Website + App)
Performance goal: Maximize number of conversions or Maximize value of conversions
Dataset: Select an eligible dataset (look for the green Custom attribution label)
Conversion event: Purchase (currently the only supported event)
Attribution model:
Test campaign → Custom
BAU campaign → Standard
Custom attribution source (Test only): If you're using Web only, this auto-populates. If you use both Web and App datasets, select the source for each.
Save and pause both campaigns for QA — don't launch yet.
Reminder: Double-check that creatives and targeting are identical between Test and BAU before moving on.
Step 2: Set up the study in Meta's Experiments tool
Meta's Conversion Lift (CLS) tool is used here as a lightweight way to split your audience into the two cells. Note that this is a workaround to create an A/B test with a minimal holdout, not a fully powered lift study (more on that below).
Open Meta's Experiments tool and click Create a Lift Test.
Give the study a clear name, e.g.
Custom Attribution vs BAU – Q3 2026.Create two cells:
Cell 1 (Test):
Meta PBA test – TEST {campaign name}Cell 2 (BAU):
Meta PBA test – BAU {campaign name}For more complex designs (e.g. 3-cell), add cells as needed.
Assign campaigns to each cell by specifying the relevant Ad Sources.
Schedule:
Start date: one day before your campaign start date.
End date: one day after your campaign end date.
Holdout: Set the control (holdout) group to the minimum allowed — e.g. 2% for a standard 2-cell study (1% per cell). A smaller holdout maximizes your test audience, which matters most when using CLS to split traffic.
Select Purchase as the primary objective. Add other KPIs (e.g. ROAS) if you want to track them.
Important — don't use the Lift UI to pick a winner. Because this setup uses a minimal holdout, the study isn't statistically powered and is very unlikely to reach significance. Treat the Lift results as a mechanism for splitting the audience, not as your decision-making data. Determine the winning campaign using your Triple Whale source of truth.
Step 3: Launch and monitor
Launch both campaigns and keep an eye on the following throughout the test:
Integration health: Confirm your CAPI integration and campaign quality stay stable in Events Manager and Ads Manager.
Match key coverage: Make sure it stays consistent before and during the test.
No manual changes: Leave Test and BAU campaigns untouched for the full duration.
Step 4: Review results in Triple Whale
Custom Attribution campaigns show a Custom label in Ads Manager, and their Results column reflects your externally attributed (Triple Whale) data rather than Meta's default attribution.
Even so, evaluate the Test vs. BAU result in Triple Whale, for two reasons:
Attribution methodology differs. In Ads Manager, Custom Attribution may show lower KPIs than a Standard campaign, because Custom reflects cross-publisher attribution while Meta's standard reporting is single-source. That's expected, not a sign of underperformance.
Data timing differs. There can be processing delays or differences between when Meta receives attributed data and when Triple Whale reports it.
Comparing both campaigns in Triple Whale, on the same attribution basis, gives you the cleanest read on whether Custom Attribution is driving real, incremental performance.