Google Ads agency red flags include withheld account access, reports built on vanity metrics, and conversion tracking that cannot be verified inside the account.
The warning signs below are checkable. Most of them live inside the Google Ads account itself, which means a business owner or marketing lead does not need technical training to confirm whether something is wrong. The checks focus on ownership, reporting, campaign hygiene, and contract terms.
Google Ads Agency Red Flags. What to Check First
Start with ownership and access, because every other check depends on being able to see the account. An agency that controls the login controls the evidence.
- Confirm the Google Ads account is owned by the business, not the agency, and that the business holds admin-level access.
- Open Change History and look at the dates of the most recent edits.
- Check whether conversion tracking is configured and whether the recorded conversions match real enquiries or sales.
- Review the search terms report for irrelevant queries that should have been excluded.
- Check how many keywords run on broad match and whether negative keyword lists exist.
- Compare the reported numbers against the account's own figures for the same period.
- Read the contract for notice periods, exit clauses, and how fees are calculated.
Each item produces a yes-or-no answer rather than an opinion. That matters when a concern has to be raised with an agency or escalated internally.
Account Access and Ownership Signals
The clearest single signal is an agency that will not grant admin access to the advertiser's own account. Common explanations include proprietary technology, internal policy, or a claim that access would expose trade secrets. None of those explanations change the underlying position: the advertiser is paying for the account, and the account history is the advertiser's record of what was done with that money.
A related pattern is the agency-owned account. The advertiser's campaigns run inside an account created and controlled by the agency, with the advertiser added as a limited user. If the relationship ends, the campaign history, conversion data, and audience signals stay behind. Rebuilding from scratch costs time and loses the learning the account accumulated.
Shared tracking setups create a different problem. When one conversion tag or pixel is used across multiple client accounts, the data feeding optimisation decisions is mixed. Attribution becomes harder to trust, and a business cannot tell which of its own conversions came from which source.
Reporting That Hides Real Performance
Reports become a red flag when the numbers shown cannot be reconciled with the account. Impressions, clicks, and average cost per click are easy to display and say little about whether spending produced enquiries. A report that leads with those figures while omitting cost per conversion, conversion volume, and search term quality is describing activity rather than results.
Several specific reporting habits deserve scrutiny. Counting view-through conversions alongside click conversions inflates apparent performance. Treating add-to-cart, begin-checkout, or page views as conversions makes a campaign look productive when no purchase occurred. Reporting brand and non-brand performance as one blended figure hides whether growth came from people already searching for the business name or from new demand.
Static PDF reports with no live dashboard access are a structural problem rather than a stylistic one. If the only view of performance is a document the agency prepares, the advertiser cannot independently check anything between reporting cycles.
Campaign Management Warning Signs
Change History answers the question of whether anyone is working on the account. A long gap between edits suggests the account is running on defaults. That is not automatically harmful, but it means optimisation decisions are not being made.
Keyword hygiene is the next area. Broad match keywords let Google match searches beyond the literal terms, which can be useful with enough conversion data and sensible bidding. Without negative keywords, broad match also lets spend flow to queries with no commercial intent. An account with broad match keywords and a thin or missing negative keyword list is spending without a filter.
Campaign structure reveals intent as well. Running Search and Display campaigns together, or relying entirely on Performance Max with no asset customisation, makes it difficult to see which placements and creatives produced results. Geographic targeting that does not match where the business actually serves customers is another checkable issue, particularly for businesses whose customers are local.
Conversion tracking integrity sits underneath all of this. If tracking is broken, missing, or counting the wrong action, every optimisation decision built on it is unreliable. This is worth verifying directly rather than accepting a summary.
Contract and Pricing Patterns to Question
Percentage-of-spend pricing creates a structural conflict. When the agency's fee rises with ad spend, the incentive to increase spend does not automatically align with the incentive to improve return. That does not make the model unusable, but it makes the absence of a cap or a performance clause worth questioning.
Long contracts with painful exit terms are a common pattern. Notice periods measured in months, fees charged on remaining contract value, or clauses that restrict the advertiser from working with another provider deserve a careful read before signing. Guarantees of specific rankings or specific results before the agency has reviewed the account and the business are another signal, because the inputs that determine outcomes are not fully under any agency's control.
Unclear billing is a related issue. Management fees, ad spend, and any platform or tooling charges should appear as separate, identifiable lines. When they are bundled into one figure, the advertiser cannot tell what portion of the payment buys media and what portion buys management.
How to Verify Concerns Inside the Account
Verification works best as a sequence rather than a single audit. Confirm access first, then read the account's own record of changes, then compare that record against what has been reported. Discrepancies between the two are the most useful finding, because they are specific and difficult to dispute.
Documentation matters at this stage. Screenshots of Change History, exported search term reports, and saved copies of reports received from the agency create a record that supports a direct conversation. A conversation framed around specific account evidence is more productive than one framed around general dissatisfaction.
If the concerns are confirmed and the relationship ends, the transition itself needs planning. Access should be secured before notice is served, and any overlap period should be used to transfer historical data and confirm that tracking continues to fire correctly. Losing conversion history during a handover damages the next provider's ability to optimise from day one.
For businesses in Malaysia evaluating a prospective agency, the same checks apply before signing. Asking for admin access, a defined reporting format, and clear contract terms during the selection process surfaces most of these issues early. Blackstone Intelligence, a Kuching-based consultancy operated by Blackstone Consultancy Sdn Bhd, works across SEO, web systems, and AI automation for Malaysian businesses, and its published case work includes local search projects for Sinar Saredah Sdn Bhd and Eyonic Sdn Bhd.
The practical test is whether the account can be inspected. Where access, change history, and conversion data are all available and consistent with what has been reported, the relationship is functioning on verifiable terms. Where any of those three is blocked, the absence of evidence is itself the finding.