Google Ads API

# Fix list: Google Ads API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/google-ads-api, the October 2026 research run, assessed 8 October 2026. Grade BB, 76.4 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Google Ads API: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). No charge for API calls was found in the docs or the terms, and the access levels and daily operation limits are public without a login (20). Test access comes with enabling the API and test accounts have no billing, so no card is needed (20). A person creates the Cloud project and OAuth credentials, applies for production access and grants the account role (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Security & auth, 71 out of 100, up to 5.1 more on the total

Why it scored 71: OAuth 2.0 with user or service-account flows, revocable grants, and 2-step verification or a passkey for new refresh tokens. There is one scope, `adwords`, described in the discovery document as see, edit, create and delete (22 of 30). The API follows Google Ads user roles, so a read-only role limits a credential, Standard access has a reporting-only permissible use, the MCP server is read-only, and adding, removing or changing users needs a second administrator's approval. Nothing confirms a budget or bid change (15 of 20). The MCP README says only that the server exposes account data to the connected model. No prompt-injection guidance was found (5 of 15). `change_event` returns the user, client type and old and new values for the past 30 days, and the Cloud console shows methods called (12 of 15). security.txt valid to 2030, Google's vulnerability reward programme, and ISO 27001 listed for Google Ads. SOC 2 is listed for Cloud and Workspace, not Ads (17 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 3. Reliability, 82 out of 100, up to 3.6 more on the total

Why it scored 82: Graded on the public API, with the hosted lines. The Ads Status Dashboard lists Google Ads API as its own product with an incidents feed (20). The feed runs from 24 October 2025 to 3 October 2026 and shows no incident on Google Ads or the Google Ads API, so the last 90 days are clean (30). Daily operation limits are published per access level (2,880, 15,000 or unlimited), with 1 request a second on planning services and 10,000 operations per mutate request, but the general per-second limit is a token bucket with no published number (12 of 15). The error guide asks for exponential backoff with jitter on `TRANSIENT_ERROR` and `INTERNAL_ERROR` and names `RESOURCE_TEMPORARILY_EXHAUSTED` for rate limits. No idempotency key was found, though `validate_only` and atomic mutates limit partial writes (10 of 15). No SLA for the API was found (0). The API is generally available at v25. The MCP server is at v0.0.4 (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 4. Schema & documentation, 84 out of 100, up to 2.6 more on the total

Why it scored 84: Google publishes a discovery document for REST (v25, revision 20261005, 175 methods and 1,489 schemas) and protocol buffers for gRPC, which is a machine-readable contract though not OpenAPI (25). No llms.txt. developers.google.com/google-ads/api/llms.txt and developers.google.com/llms.txt both returned 404 on 8 October (0). Guides say which service to use and when, such as SearchStream over Search for large reports and the Data Manager API for new conversion uploads, and the MCP tool descriptions tell a model to call `get_resource_metadata` before `search` (16 of 20). Typed messages with 883 enum definitions in the discovery document, but reports are a GAQL query string (13 of 15). Code examples in six languages and typed error enums with a common-errors guide (15). Major versions in the path, dated release notes and upgrade guides (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Agent ergonomics, 87 out of 100, up to 2.1 more on the total

Why it scored 87: GAQL selects only the fields asked for, with WHERE, ORDER BY and LIMIT. The MCP server has three tools, all read-only, though the `search` description embeds the full list of resources (22 of 25). Search returns fixed pages of 10,000 rows with `next_page_token`, and SearchStream returns a whole report on one connection (20). Errors are typed enums with a `request-id`, sorted in the docs into authentication, retryable, validation and sync errors (18 of 20). No idempotency keys were found. `validate_only`, atomic mutates with optional `partial_failure`, and `readOnlyHint` on all three MCP tools earn part (14 of 20). Six official client libraries. Calls need a customer ID, often a `login-customer-id` header, and knowledge of GAQL (13 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 6. Transparency & trust, 82 out of 100, up to 1.6 more on the total

Made of editorial 67, provenance 96.

Why it scored 82: Closed service under the Google Ads API terms, with Apache-2.0 client libraries and MCP server. The terms page was last updated on 21 February 2019 and still defines the developer token (15 of 30). The Google Ads Data Processing Terms set deletion within a maximum of 180 days and commit to ISO 27001, and the docs state reporting retention of 37 months for granular data and 11 years for monthly and coarser data. The MCP README discloses a usage header added to API calls (22 of 30). A sunset timetable with dates for every version and a dated table of feature deprecations (20). A sub-processor page with entity names and locations exists, but we couldn't confirm which entries apply to Google Ads (10 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Terms of service: read, states 6 of the 7 things a reader expects (9.1 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects, and has 1 clause that costs points (7.3 of 10)

## 7. Maintenance & community, 87 out of 100, up to 1.1 more on the total

Why it scored 87: v25.2 on 23 September 2026, 15 days before the check (30). v25 on 22 July, v25.1 on 19 August and v25.2 on 23 September are three dated releases within 90 days (20). Dated release notes, a support page with a channel for each kind of issue, a Discord channel for the AI tools, and a pull request merged in google-ads-mcp on 29 September. The developer blog returned 403 to our reader, so announcements there weren't read (12 of 15). Six current official client libraries, with google-ads 33.0.0 for Python on the day of v25.2 (15). CI workflows in the Python client and MCP repositories (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: the Google Ads developer blog (ads-developers.googleblog.com returned 403 to our reader), so the notice given before the developer token sunset and before the Customer Match and offline conversion restrictions wasn't read
- Which entries on the Ads sub-processor page apply to Google Ads, since the list names other advertising products against each entity
- The per-second rate limit, which the docs describe as a token bucket that varies with server load
- Whether the API terms will be revised, since the page dated 21 February 2019 still defines the developer token
- The MCP README still lists a developer token step in setup, while the developer guide says the latest server version doesn't need one
- unchecked: the ISO 27001 certificate for Google Ads is linked from the compliance page as a PDF, which we didn't open
- The criteria for Explorer access, since the docs say only that Google may let a project apply and may upgrade it automatically

## Weaknesses

- One OAuth scope, `adwords`, covers reading, editing, creating and deleting. Read-only access depends on the Google Ads user role
- Production accounts need Explorer access (2,880 operations a day) or higher. Basic needs brand verification and Standard a manual audit of about 10 business days
- The per-second rate limit has no published number, since it's a token bucket that varies with server load
- No idempotency keys were found in the reviewed documentation
- The official MCP server is read-only, self-hosted and at v0.0.4, and it isn't in the official MCP registry

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Send `validateOnly: true` on a mutate first. It returns errors without making the change. Batch jobs don't support it
- Create campaigns with status PAUSED and enable them in a separate step, since an enabled campaign with a budget can spend
- Send `login-customer-id` (digits only, no hyphens) when you reach a client account through a manager account
- A failed request still counts against the daily operation quota. Paging with a valid `next_page_token` doesn't
- Wait at least 12 hours between account budget changes on one account, and use the Data Manager API for new Customer Match or offline conversion work

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.