Skip to main content

Field guide / 12 checks / Australia

Google Ads audit checklist for service businesses

Before increasing spend, check what the account counts as a result, which searches it pays for and what happens after the enquiry. Use this with your business owner, adviser or Google Ads provider to agree what needs attention.

By Ajay Dabhi · Published and reviewed 16 September 2026

01 / Before you start

Bring the account and the business records together.

Start with permission to inspect the account, a complete reporting period, the services and areas the business actually covers, and the records used to judge enquiry quality. If something is unavailable, mark it unknown and assign the next investigation.

Use the worksheet’s status column for pass, needs attention, unknown or not applicable. Record your evidence before making changes. A pass means you have checked the relevant condition; it is not a guarantee of future performance.

This guide focuses on service-business lead generation, especially Search campaigns. Shopping feeds, app campaigns and a full Performance Max asset or placement audit need additional checks.

Keep the completed worksheet private to the review team. Use internal reference numbers rather than customer names, contact details or account credentials.

02 / Review and record

The 12-point Google Ads audit.

0 of 12 checks reviewed

Tick a check after reviewing it. This tracks your reading, not the account’s health. Ticks reset when you leave or reload; save findings in the downloadable worksheet.

Check 01

Confirm business ownership and access

Where to look
Account access, linked accounts and billing settings, with the business owner.
Evidence to record
Who has administrator access, who owns the billing relationship, which providers are linked and who can recover access. Keep personal details out of shared copies.
How to decide
The business should retain control and a clear handover route. Resolve unexplained access or billing blocks with the owner before planning a launch.

Check 02

Write down the reporting baseline

Where to look
Campaign reports, date selector, account time zone and change history.
Evidence to record
The exact period, currency, active campaigns, spend, conversion definitions and material changes. Exclude a partial current day and allow for delayed conversions when comparing periods.
How to decide
Compare like with like. If goals, bids and pages changed together, record the uncertainty instead of crediting one edit for the result.

Check 03

Check what each conversion really measures

Where to look
Goals → Conversions, then each campaign’s selected goals, including custom goals.
Evidence to record
For each action: source, trigger, primary/secondary status, count setting, value and campaigns using it. Separate a page view, phone click, enquiry, booking and paid job.
How to decide
Use a deliberate, validated business outcome for bidding. Secondary actions can still be used by custom goals. Document the goal map before changing it; do not switch every setting at once.

Check 04

Test the call and form measurement

Where to look
The actual mobile landing page, tag diagnostics, call reports and the receiving business system.
Evidence to record
With the owner’s agreement, record controlled test timestamps and the expected events: click only, successful submission, connected call and repeat interaction. Check duplicates, receipt and classification.
How to decide
A phone tap does not prove a connected call. A tag firing does not prove a valid enquiry. Repair missing or duplicate events and retest before treating the conversion count as reliable.

Check 05

Read the searches you paid for

Where to look
The search terms report for the same campaign and date range, with cost and outcome columns.
Evidence to record
Examples of useful service enquiries, wrong services, job searches, product research and places you cannot serve. Match to lead quality when possible. The report does not disclose every query.
How to decide
Keep relevant intent, investigate ambiguous intent and exclude clearly unsuitable searches. A query with no conversions in a small sample is not automatically waste.

Check 06

Inspect negative keywords and their scope

Where to look
Account, campaign and ad-group negatives, plus the shared lists actually attached to campaigns.
Evidence to record
The excluded term, match type, applicable campaigns and a real unwanted search it addresses. Check that useful service or booking searches are not blocked.
How to decide
Add a specific exclusion for a demonstrated mismatch. Test its effect on useful searches. Do not copy an old or generic list wholesale; negative match behaviour differs from positive keywords.

Check 07

Match geography and hours to delivery

Where to look
Campaign location targets, exclusions, advanced location options, geographic reports and ad schedules.
Evidence to record
Where customers can be served, where enquiries originated, the selected presence/interest setting, call-answering hours and account time zone.
How to decide
For on-site services, assess presence-based targeting against the real service area. Interest-based targeting may suit other buying journeys. Google’s location signals are estimates; settings alone do not guarantee serviceable leads.

Check 08

Separate spend by campaign and network

Where to look
Campaign settings and performance reports segmented by network, device and campaign type where available.
Evidence to record
Where spend went, the bidding strategy and targets, the goal used and qualified outcomes. Keep Search, search partners, Display and Performance Max scope explicit.
How to decide
Investigate a costly segment using its lead quality and evidence volume. Do not raise budget solely to improve impression share or accept an automated recommendation without a business reason.

Check 09

Match the search, ad and service

Where to look
Keywords, responsive search ads, assets and their final URLs.
Evidence to record
One real search → matching ad promise → destination for each priority service. Check current claims, location wording, disapprovals and outdated sitelinks.
How to decide
Repair a broken promise or destination. Give materially different services a suitable message and page. Keep price, availability, qualifications and results claims tied to current evidence.

Check 10

Follow the mobile enquiry path

Where to look
The live final URL and each redirect, on a phone as well as desktop.
Evidence to record
Service and area clarity, credible proof, visible call/enquiry action, usable fields, loading problems and whether the request reaches the right person. Include the thank-you state.
How to decide
A broken destination or contact path takes priority. Improve the specific obstacle a buyer encounters, then repeat the complete check. A good-looking desktop page is not sufficient evidence.

Check 11

Reconcile enquiries with business outcomes

Where to look
Call and form records alongside the CRM, booking system and completed-job records.
Evidence to record
Unique enquiry, source where known, service, suitability, response, booking, completed paid work and loss reason. Keep unknown attribution visible and remove duplicates.
How to decide
Calculate cost per qualified enquiry and customer only when their definitions and records support it. Agree who returns this feedback and how often; do not label every platform conversion a sale.

Check 12

Turn the findings into a short action plan

Where to look
Your completed worksheet, reviewed with the business owner and the person making changes.
Evidence to record
For every issue: source/date, consequence, priority, recommendation, owner, due date, acceptance check and retest result. Record a baseline before implementation.
How to decide
Fix loss of control, broken enquiry paths and unreliable measurement first. Then address demonstrated waste and message gaps. Test one clear hypothesis at a time where practical; judge the result after enough relevant evidence has accumulated.

Field note / Account review

A phone click was set as a primary conversion.

In a September 2026 review of a service-business account, I found a website conversion action whose source was phone-number clicks. Its optimisation setting was primary. The review also lacked verified paid-job or revenue feedback.

That is a useful place to start an audit: establish exactly what the account calls a result before deciding whether it needs more spend.

Evidence boundaryAnonymised configuration finding from a real account review. Client details are withheld. This example reports the finding and recommendation; it does not claim a completed repair, verified calls or improved results. It is separate from the 153-lead case study.

01 / OBSERVED

The event and the business outcome differed.

The source measured a tap on a phone number. That tap alone did not establish that a call connected, became a suitable enquiry or turned into paid work.

02 / INTERPRETATION

The report could overstate useful enquiries.

A primary action can influence bidding when its goal is used by the campaign. A phone-click count is therefore insufficient evidence for a claim about qualified leads or profitability.

03 / RECOMMENDATION

Define the goal before changing the bidding.

Document the campaign’s selected goals. Keep phone clicks as a diagnostic signal, establish a reliable call or enquiry measure, and connect it to qualification and paid-job records. Check custom goals too: they can use secondary actions for bidding.

04 / ACCEPTANCE CHECK

Prove the whole path.

The owner and measurement lead should agree controlled tests: a tap without a call, a connected call and an enquiry with a recorded outcome. Confirm each is counted only as intended, record the change date and reconcile the reporting before judging performance.

03 / From finding to action

Leave with an owner and a retest.

The worksheet contains all twelve checks, blank fields for your findings and a separate example row. The example records a recommendation from the review above; its repair and results remain unverified.

P0 / Fix or contain first
Loss of control, a broken enquiry path or unreliable measurement that makes spending decisions unsafe.
P1 / Next agreed change
Demonstrated spend mismatch, service-area problems or an ad-to-page gap with a clear business consequence.
P2 / Planned investigation
A hypothesis, improvement or small sample that needs more evidence. Record when you will revisit it.
Download the free audit worksheet

At the next review, reopen unresolved findings, record what changed and repeat the acceptance checks. Compare qualified enquiries and paid work using the same definitions. A lower platform cost per conversion alone does not establish improvement.

04 / Sources and method

Check the underlying rules.

Google’s documentation below was checked on 16 September 2026. Interface labels can change. The audit order and priority framework are my recommendations; the worked example is an anonymised first-party review finding.

05 / Practical questions

Using the checklist.

Can I use this with my current agency or adviser?

Yes. Share the page and use the worksheet together. Ask for the evidence behind a finding, who will make the change and how you will know it worked. You do not need to change provider to use the process.

Is this a free audit service?

The checklist and blank worksheet are free resources with no signup. They do not include an inspection of your account. If you want me to review an account or implement changes, we agree the scope and price first.

What date range should I audit?

Choose a complete period that contains enough activity for the question, record its dates and allow for conversion delay. Use a comparable earlier period where useful. Extend the review for a low-volume account, and annotate changes in goals, services, budgets or seasonality.

What should I fix first?

Start with loss of account control, an unavailable destination, a broken enquiry path or measurement you cannot trust. Then prioritise demonstrated waste and message mismatch. Low traffic or a low optimisation score alone does not identify the right fix.

06 / Next move

Need a second pair of eyes on the findings?

I review Google Ads accounts and the enquiry path for service businesses. We agree the scope before work begins. I am based in Melbourne, with on-site visits by arrangement and remote work across Australia.

Discuss an account review →