Guide

Migrating offline conversion imports to the Data Manager API

On June 15, 2026, Google blocked offline conversion imports and enhanced conversions for leads through the Google Ads API. This guide walks through moving an existing upload workflow to the Data Manager API, step by step.

Step 1 — Pick your route

There is no like-for-like replacement; the migration splits into two paths:

	API route	UI route
Who it's for	CRM integrations, scheduled scripts, anything that called `UploadClickConversions`	Marketers uploading CSV files by hand
New destination	`POST https://datamanager.googleapis.com/v1/events:ingest` (JSON)	Data Manager import inside Google Ads (CSV, with column mapping)
What changes	New request shape, RFC 3339 timestamps, numeric conversion action IDs, no developer token	Mostly familiar — the legacy file import is marked "legacy" and Data Manager takes over

Step 2 — Map your fields

The field mapping table covers every field. The three that break most migrations:

Timestamps: conversion_date_time (yyyy-MM-dd HH:mm:ss+00:00) becomes event_timestamp in RFC 3339 (2026-06-20T15:07:01Z). See the timestamp cheatsheet.
Conversion action: no longer a per-row name. It becomes destinations[].product_destination_id — the numeric ID from Goals › Conversions (the ctId parameter in the URL). One request per conversion action.
Click IDs: gclid, gbraid, and wbraid move into the nested ad_identifiers object.

Step 3 — Convert your existing files

The converter tool takes a legacy CSV and produces both a Data Manager API JSON payload and a clean import CSV, validating every row. All processing happens in your browser — conversion data (gclids, order IDs, values) never leaves your machine.

Step 4 — Authenticate and send (API route)

OAuth 2.0 scope: https://www.googleapis.com/auth/datamanager. A Google Ads developer token is no longer required.
Batch up to 2,000 events per request.
Set validate_only: true to dry-run a payload before sending it for real.
Processing is asynchronous — keep the returned request_id to retrieve diagnostics. There is no per-row synchronous result and no partial-failure mode.

Step 5 — Verify

Start with validate_only: true and fix anything it rejects. Common failures are listed in the error codes FAQ.
Send a small live batch, then check the conversion action's status in Google Ads after processing.
Watch for silently dropped data: conversion_environment and external_attribution_data have no Data Manager API equivalent.

Rather not maintain this pipeline at all? Server-side tracking platforms such as Stape or Tracklution stream conversions to Google continuously, so there is no monthly CSV ritual and no format migration next time Google changes the API.