Google is changing how advertisers send first-party conversion and audience data across its advertising ecosystem, with Data Manager API becoming a new infrastructure layer for workflows such as offline conversions and Customer Match. In this webinar, Stape will explain what Data Manager API is, what has changed in Google’s approach, and show how to start using it through Stape for conversion and audience data.
Speakers:
Dan Murovtsev, Product Manager @ Stape
1. Google Data Manager API: what changed and why it matters
See where Data Manager API fits into Google’s first-party data infrastructure, what has changed in 2026, and why advertisers should start getting familiar with it now.
2. What can you do with Data Manager API today?
Explore practical capabilities such as uploading offline conversions, sending additional conversion data, working with Customer Match audiences, and sending data to Google platforms through a unified API.
3. How Data Manager API works with Stape
See how Stape simplifies the connection to Data Manager API and learn how Stape tags and apps can be used to send conversion events and audience data.
4. Data Manager API in practice: live demo
Follow a practical setup in Stape, see conversion and audience data being sent through Data Manager API, and learn how to identify and troubleshoot common implementation issues.
5. Q&A
Bring your questions about Google Data Manager API, offline conversions, and Customer Match.
Click the button below to get the webinar presentation.
The Data Manager API can already send data to GA4, and the Stape conversion events tag includes Google Analytics as a destination.
However, this functionality is at a very early stage and limited in what it does, with intermittent success so far. It is worth testing, but it is not yet fully reliable.
The regular Google Ads tag runs on proprietary code and is designed to work only with incoming gtag requests. Google can tell when an event was not generated by gtag in the browser, so the standard tag does not work properly with other incoming events. A partial signal may still be sent, but this is not a supported use case.
The Data Manager API tag, like the Google Ads Offline tag before it, is built for sending conversions based on incoming events that don't come from Google, such as webhooks.
The Data Manager API currently has no endpoint for retracting a conversion. The only option is adjusting the value through the multi-source conversion mechanism, so conversion adjustments can only be partially replaced for now.
For retractions and refunds, the Google Ads Offline API with the Conversion Adjustments tag is still available. Although this API is restricted for new users, it remains accessible through the Stape connection.
Yes. Measurement Protocol is officially in maintenance mode and no longer receives updates. The Data Manager API is already positioned as its official alternative and is expected to fully replace it eventually.
Whether it will become a full-fledged alternative remains to be seen.
No, tracking specialists are not being replaced. Stape itself is actively hiring software engineers.
AI is a tool, and it still requires context and underlying skills. Asking an LLM to set up tracking without understanding what it does, or without the ability to verify the result, does not produce reliable work. Tracking skills remain valuable; the work simply involves fewer manual clicks.
Users with a specific lead gen use case, template or platform in mind are encouraged to submit a request on the Stape feature request board (feature.stape.io). Stape already has more than 150 templates in the gallery and is happy to build new ones that don't exist yet.
There is no ready recommendation for this setup yet. Native connections for importing data are available for some platforms, such as Meta.
The Stape team is open to exploring this case. Users are encouraged to send a support ticket describing the use case and the goal they want to achieve.
For popular CRMs, Stape apps can send the webhooks and handle the setup. Most other CRMs also have a native way to send webhooks, either through the interface or with some code. Their marketplaces may also offer apps for this. Checking the CRM's own capabilities first is recommended before turning to tools like Zapier, since all that's needed is an HTTP request sent to the server container.
An existing container can be adapted without any problems. The Data Manager connection and the required tags can simply be added to it.
No specific background is required. Anyone who is interested in tracking and enjoys the work can become an expert. The webinar host, for example, previously worked in aircraft ground handling.
Such setups have not yet been seen in real practice, but they should be possible if data from these sources can be routed to server GTM. This should work for Databricks and Snowflake; BigQuery is less certain.
Batches are supported but limited: approximately up to 50 users per audience request and up to 10 conversions per outgoing event. The Stape team is open to discussing specific cases and limitations.
Yes, up to 90 days, depending on the conversion settings.
These are different use cases. Offline conversions should be sent through Data Manager, while online conversions should be sent the usual way.
Yes, capturing multi-source interactions directly supports the GA4 data-driven attribution model. However, there are specific technical limits on how non-Google data is ingested and evaluated.
It is the same field. The Data Manager API adjusts the value of the existing event.
No, Data Manager is not related to the Enricher power-up.
Only the gclid can be sent, as it is the most valuable of the three identifiers.
Stape plans to add these parameters to the logs in the near future. Own credentials can also be used with Stape hosting.
Yes, this may lead to over-reporting.
Multi-source conversion adjustments only apply after the conversion action's 14-day trial period. Data Manager hits sent during this period are not processed for that conversion action.
Yes, that is correct, provided the same transaction ID as in the originally reported conversion is sent.
In this example, the end account is under an MCC account, and the connection was made with an account that has access to the MCC. This is why the IDs are different. Without an MCC, the same ID is entered in both fields.
Comments