How to check your own Google Ads account

Before you audit the account

This is a review you can do without changing a campaign. Use read-only access, record what you find, and agree on any edits with the person responsible for the account.

Open the correct account and note its currency, time zone and reporting dates. Choose a period long enough to include your normal sales cycle. Keep those dates consistent across exports; recent leads may still be waiting for a quote or decision.

A billing email does not establish campaign access. Ask an account administrator to invite your Google login if needed. Google’s access-level guide distinguishes viewing reports, editing campaigns and managing users.

Check 01

What counts as a conversion?

In Goals → Conversions → Summary, list the conversion actions. Write down what actually triggers each one: a submitted form, a call, a purchase, a page view or an imported sales outcome.

Then inspect the goals selected by each campaign. A useful event can still be the wrong bidding objective. Opening a contact page, for example, does not establish that someone sent an inquiry.

Record the action name, source, goal, optimization setting and business meaning together. Use Google’s primary and secondary conversion guidance to interpret the settings. Do not change a goal just because its label looks unfamiliar.

Check 02

Is one result counted correctly?

Compare the intended action with its trigger and the record created in your business. A form confirmation, the email it generates and the CRM entry can all describe the same inquiry. They are not automatically three leads.

Check whether Google Ads tracking and a GA4 import represent the same action. Having Google Tag Manager and a Google tag on the website is not, by itself, proof of double counting. The evidence is which events fire, which conversion actions receive them and which goals use those actions.

For purchases, inspect order IDs and values. Google documents transaction-ID deduplication for repeated conversions within the same conversion action. It does not mean separate conversion actions automatically deduplicate each other.

If a test submission or purchase is needed, arrange an approved test with the site and account owner. Record the expected event and the observed result. Treat a dashboard discrepancy as something to investigate, not immediate proof that the tag is broken.

Check 03

Which searches are you paying for?

Open Campaigns → Insights and reports → Search terms. Sort by cost and compare the visible queries with your actual offer, service area and customer intent.

Separate clearly irrelevant searches from relevant searches with no recorded outcome yet. A plumber that does not sell parts has a different reason to question “wholesale pipe fittings” than “water heater replacement near me.” The latter needs an outcome and landing-page check before a verdict.

The report omits some queries for privacy and volume reasons. Its match-type column describes the relationship between the query and the keyword; it is not simply a copy of the keyword’s setting. See Google’s search terms report documentation.

Keep a proposed negative-keyword list separately. Before applying it, check whether it would exclude a service you do sell. Broad match alone is not evidence of waste.

Check 04

Does bidding match the outcome?

For each campaign, record the bid strategy, budget, selected goals and any target. Then ask whether the conversion data represents the business result you want more of.

Google began changing these strategy names in June 2026: “Maximize conversions with a target CPA” becomes “Target CPA,” and the corresponding value strategy becomes “Target ROAS.” Older labels may still appear in records. See Google’s bidding guide.

A missing target is not automatically a fault. A ROAS target is not a required control for a conversion-volume strategy. Record the intended outcome, the reliability of its measurement and the business constraint before recommending a change.

Check 05

Can the visitor take the promised next step?

Open the final URLs from the account, including relevant mobile pages. Does each page offer the advertised service or product? Are location limits, availability and the next action clear?

• For a service inquiry, check the phone link, form and who receives the request.
• For a purchase, check product availability, the cart and the checkout path.
• For a booked call or appointment, check that the scheduling step works and records the booking.

Record broken links and mismatched promises with the exact ad and destination. A functioning form does not establish that someone answers it. Compare the handoff with your inquiry record. Our Google Ads DIY guides cover the related setup decisions.

Check 06 · If used

What is Performance Max promoting?

Review campaign goals, asset groups, promoted products or services, and the pages visitors can reach. Look for a specific mismatch: an unavailable product, an unrelated destination or a goal that rewards the wrong action.

A single asset group is not automatically a problem. Google recommends grouping assets around a theme, and audience signals are optional. Judge whether the structure fits the actual offer. See Google’s asset-group guidance.

Performance Max also has a search terms report. Use the available detail alongside landing-page and business-outcome records; it does not make every placement or customer journey visible. Skip this check if the account does not use Performance Max.

Turn observations into a useful handoff

Copy this record for each finding. Keep the observation separate from your interpretation, and give any proposed change an owner.

Where and when

Account, campaign, reporting dates and the exact report or setting.

What you observed

The value, query, event or destination. Attach a private screenshot or export.

Why it matters

The business outcome it may affect, plus anything still unknown.

Next verification

The record, test or person needed to confirm the explanation.

Proposed change

What to change, who approves it and how the result will be checked.

Prioritize confirmed measurement failures and broken destinations before drawing conclusions from cost per conversion. Keep unresolved questions marked as unresolved. This review should leave a clearer decision, not a longer list of assumptions.