Google Ads for local services: three checks that connect advertising with real enquiries

Choose the three local-service checks

Local-service advertising has a practical boundary: the business must be able to help the person who responds. A campaign can record contacts while attracting enquiries outside the service area, promoting unavailable work or directing callers to a route nobody answers. Reviewing only the headline acquisition cost leaves those differences unexplained. Begin with the connection between the advertised promise and the service the team can actually provide.

Three checks give that review a useful order. Confirm the service geography and the configured targeting. Identify exactly which calls or forms the account counts. Test the contact route and confirm who handles it. These are connected responsibilities, but each needs its own evidence. A location setting describes advertising configuration; a call metric describes a measurement; the business's response process describes operations.

This guide concerns ordinary Google Ads account review for businesses serving local customers. It does not treat Google's separate Local Services Ads product as interchangeable with the advertising resources described here. The review principles are practical, while the application claims refer to the current Google Ads reporting and approval workflows.

Write down the area the business serves

A business's service area can be more nuanced than a circle around its office. Travel time, appointment length, staffing and the type of work can affect where it accepts customers. A mobile technician may cover a different area for urgent work than for planned installations. Write these distinctions down before deciding whether a location row is useful or irrelevant.

Separate the public offer from internal exceptions. If the website says the company serves an entire region but the team routinely rejects requests from its outer towns, advertising settings alone will not resolve the inconsistency. Confirm what should be advertised and update the service description where needed. The account reviewer should not invent an operational boundary from a campaign name or a street address.

Include how a person can use the service. A shop that customers visit has a different geography from a service dispatched to their homes. A specialist consultation may support remote enquiries even when the business has a local office. People arranging work for another property can have legitimate interest in an area where they are not currently located. Those facts influence the targeting decision and prevent simplistic exclusions.

Keep the confirmed area available to the advertising owner and the person handling enquiries. When coverage changes, record the date and the affected offer. A campaign can then be reviewed against a clear business rule instead of alternating between broad targeting and unexplained restrictions whenever the office receives an unsuitable request.

Read the configured location targets

Ask for the selected campaign's current location settings when reviewing coverage. The configured targets, exclusions and any applicable radii describe what the account is trying to reach. They are separate from a performance table showing where activity was reported. A familiar campaign name does not prove the expected targets are still in place, especially after changes made by several people or tools.

Inspect the actual target scope before requesting an edit. A regional campaign can contain more than one area, and removing one target does not necessarily create the entire intended service boundary. A radius may also include places the business does not serve. The proposal should state which existing target changes and what is added or removed, with units clear when a distance is involved.

The current edit catalogue includes supported campaign location changes and a Presence setting action. Support remains specific to the selected resource and setting. The existence of a broad catalogue is not permission to claim that every campaign type uses identical location resources or that every provider operation can be executed through the same builder.

Read current settings first, then formulate the requested state. That avoids a blind “fix the location” instruction whose desired outcome is unspecified. For detailed distinctions between configuration and performance, see the location-review guide. The local-service question is whether the resulting target fits the service offer, not merely whether the API accepted a location.

Understand presence and interest

Google's advanced location documentation distinguishes people likely to be in or regularly in a location from a wider presence-or-interest population. That difference can matter when a service requires the customer to be in a particular area. It can also matter when a customer legitimately arranges local work while located elsewhere. Review the business need before choosing the setting.

A strict local appointment offer may call for a presence-focused review. A property service taking requests from owners who live elsewhere may have a reason to consider interest in the area. Neither case justifies a universal instruction for every business with a local address. Identify the campaign, explain the customer situation and inspect the current state before preparing a supported change.

Location targeting uses provider signals and is not guaranteed to identify every person's location perfectly. A Presence setting should therefore not be described as a complete defence against every out-of-area enquiry. The contact process can still ask for the service location and explain coverage clearly. That operational check complements targeting without pretending the account report provides a verified address for every visitor.

After a change, evaluate the relevant outcomes using a stable period and definition. Lower traffic is not automatically better targeting, and a smaller lead total is not automatically a failure. Ask whether the offer now reaches the intended population and whether the measured contacts and confirmed business observations support the decision.

Interpret geographic performance carefully

A location-performance table can help identify where spend and selected conversions were reported. Name the geographic view and the period so the reader knows what the rows represent. Do not substitute user-location performance for configured targets or present a provider location signal as a complete record of the eventual service address.

Use conditions to select candidates when the question is clear. A request for locations with spend above an agreed threshold and no primary lead conversions can produce a focused review population. It does not establish that every location should be excluded. Small samples, a slow enquiry process and different service economics can all affect the decision. The report should distinguish the measured filter from the recommendation.

Ask the handling team about confirmed unsuitable service areas using aggregate observations. Those facts can help evaluate a targeting change without copying personal contact details into Slack. A customer contacting the business from one place may need service in another, so relate the operational evidence to the actual work requested rather than assuming the two locations are identical.

Keep missing coverage visible. An unavailable field or provider read is not a zero-conversion location. A preview is not the full matching count. Weighted acquisition ratios should use their compatible components rather than averaging location ratios equally. These reporting habits make the geographical review useful even when the data is sparse or the service boundary is complicated.

Define the contact being counted

A local business may accept telephone calls, enquiry forms and bookings. Those are separate routes, and each can have several implementations. Decide which route the review concerns before requesting a headline lead count. A combined primary total can be useful for account monitoring, but it should not hide the difference between a submitted appointment request and a click on a telephone link.

Write the success condition for the selected action. A form may count when submission succeeds. A booking may count when a confirmation is issued. A telephone action may count according to its particular call measurement. Ask the tracking owner if the trigger is unclear. Naming an action “Customer enquiry” does not establish that every event corresponds to a received and handled request.

The current broad Leads measure includes the primary lead categories together. Calls can be reported separately, and a named conversion action can be selected when the business needs one specific contact route. All primary Conversions is broader still. The report should retain these definitions rather than changing the denominator to whichever total produces the most attractive acquisition cost.

Also separate contacts from unique people. One person may submit a form and call later. Their actions can be recorded separately by the configured account. The reporting tools add the selected actions; they do not repair overlap by guessing. A person-level customer count requires evidence and a definition that actually support it.

Inspect the call-conversion meaning

Google documents several call-conversion measurements, including calls from ads, website calls, telephone-link clicks and imported outcomes. These do not observe the same thing. Some setups use a minimum duration; a mobile telephone-link click does not establish the call itself. Inspect the selected action before translating its count into conversations with potential customers.

For a local-service review, primary Calls can be a useful measure, and primary call conversion value can also be recorded. Explain what that value represents. A value assigned to a call is not necessarily a payment received for a completed job. An imported outcome has the meaning supplied by the business's import process. The count and value should remain accurately labelled.

Call duration can be relevant to the configured measurement without proving suitability. A long call can concern an unsupported service, and a short call can be a returning customer confirming an appointment. The assistant should not claim to know the conversation's quality from a category name. Ask the telephone or service owner for a scoped summary when suitability is the decision.

Keep actual handling questions separate from the measured total. Were calls answered? Was the number correct? Did routing reach the right team? Did the business have capacity to help? Those questions may require a test or operational evidence outside the advertising account. A reported call conversion cannot certify the entire route from dialling to completed work.

Keep primary and secondary actions clear

Inspect the configured roles of the actions that contribute to the review. Primary form and call actions can coexist intentionally. A lower-intent event may be retained for observation. The lead-generation conversion-setup audit lists these roles and asks for manual review because local businesses have different contact models. Its review state does not prove that the account must adopt a single universal setup.

The default aggregate includes primary actions and excludes secondary ones. An explicitly requested action can be examined on its own. Make that distinction clear when a colleague compares the answer with a broader provider total. Two totals can disagree because their action populations differ, not because the application failed to count the same population correctly.

Do not declare multiple primary actions a duplicate error without implementation evidence. A form and a call may be distinct events the business wants to measure. If two implementations record the same submission unintentionally, that needs a tracking investigation and a justified correction. Reporting the configured sum is separate from changing what the account considers primary.

When evaluating bidding, also inspect relevant campaign-goal configuration rather than treating the primary label as the whole setup. For the performance question, keep the selected actions stable across periods. If the owner changes a role, record the timing and explain how it affects the comparison instead of presenting the new aggregate as an unqualified improvement.

Choose the corresponding website event

Analytics can provide a website-side view of a contact route when the relevant event is implemented. Ask for observed events if the name is unknown, then confirm which one describes the intended form or booking. A name can suggest a likely match, but it is not proof of the trigger. A generic contact event may cover more routes than the particular advertisement being reviewed.

The reporting framework can compare a named Ads action with a selected GA4 event, or the primary Leads measure with the confirmed corresponding event. Preserve the requested period and traffic scope. An all-traffic website count can include enquiries beyond Google Ads attribution. A difference between those sources does not establish a failure by itself.

A telephone-link event deserves especially careful labelling. It can describe contact intent without proving a completed call. Comparing that event with a recorded call measure may be informative if the distinction is explained; renaming both “answered calls” would misrepresent the evidence. Let the user select the business question and keep the source meanings visible.

If the requested event data is unavailable, say what could not be returned. Do not use an unrelated key-event total to make the comparison look complete. The observed list is period-bound, so absence from it does not certify that the site has never sent the event. A tracking owner can then investigate the specific route with the right context.

Separate enquiry volume from commercial suitability

A recorded enquiry can be genuine and still unsuitable for the business. The caller may need work outside the service area, request something the team does not offer or require a date with no availability. An advertising count alone cannot classify every enquiry on those terms. Bring the handling team's verified observations into the review when commercial suitability matters.

Keep that context structured. Ask whether a recurring mismatch occurred, what service it concerned and over which period. Aggregate categories are often enough to identify the next task without publishing names, addresses or customer messages. A small set of confirmed observations is more useful than a sweeping claim that the campaign produces bad leads.

Recorded imported outcomes can add another view when implemented. They remain conversion actions in the account, not unrestricted access to the service database. Offline, call and lead descriptions can overlap, so their totals should not be added as independent groups. Ask for the named outcome when the review concerns a particular recorded stage.

The next decision should address the actual mismatch. Wrong service requests may need clearer copy or reviewed exclusions. Out-of-area work may need a targeting and offer review. Unanswered calls need the telephone owner. By assigning the problem accurately, the team avoids repeatedly changing advertising settings to solve an operational issue that those settings cannot fix.

Test the live contact route

Open the advertised landing page and follow the route a new customer would use. Test the relevant device experience, the form, any booking transition and the displayed telephone option. A page that loads is only the beginning of that test. The visitor must be able to understand the offer, complete the intended contact and receive an accurate confirmation.

Coordinate a test enquiry with the business so the team recognises it. Confirm that submission reaches the correct destination and that the chosen service is preserved. A form can appear successful while sending the request to an old mailbox. A booking service can accept an appointment the team cannot fulfil. These are practical problems for the route owner, even when advertising eligibility is unaffected.

Check telephone information with the same care. The visible number, call link and routing should lead to the intended business response. Do not assume that a number works because it appears in an asset. The advertising measurement and the actual handling test establish different facts. When something fails, identify the route and observed behaviour rather than declaring the entire account broken.

The current Links command reports Google's published destination findings; it does not submit the business's forms or certify live call routing. Pair that provider evidence with a browser and contact test when needed. A new website problem can precede a provider finding, while an older policy finding can need separate review after the page appears to work.

Align the offer with response hours

Response hours influence the experience promised by local advertising. A business offering immediate telephone help needs a route capable of handling that expectation. A business accepting asynchronous quotation requests can make a different promise. State the actual process on the destination and in relevant copy rather than allowing “contact us now” to imply service availability the team cannot provide.

Review campaign scheduling when timing is part of the question. Google's ad-schedule guidance explains that scheduled eligibility uses the account time zone. The office's local clock, the audience's location and the account setting may need reconciliation. An instruction such as “during opening hours” is incomplete until the days, times and applicable zone are understood.

Do not infer the best schedule from one weak hour row. Small samples, delayed conversions and different contact routes can affect the pattern. A visitor may submit a form outside office hours for later response. Review the offer and handling process alongside measured activity, then formulate a bounded change if the evidence justifies it.

Record exceptions such as holiday closures or temporary staffing changes. These need practical communication and sometimes a campaign review, but the assistant should not guess them from a performance decline. Confirm the dates and scope with the business. The next report can then distinguish the operational interruption from an unexplained advertising movement.

Keep each service promise specific

A local business may sell urgent repair, routine maintenance and larger planned work. Those offers can have different lead values, response times and service areas. Keep their differences understandable in the account review. A generic total can monitor overall contact volume, but it may hide which offer changed or which team needs to respond.

Read the advertisement and destination together. An ad for emergency assistance should not land on a page that only accepts appointments next month without explaining the transition. A low starting-price message should make its applicable scope clear. The product or service owner supplies these facts; the assistant can help express them without inventing availability, guarantees or pricing.

Campaign and ad-group structure should support the meaningful decision. Separating offers can make budget and contact outcomes easier to interpret. Excessive fragmentation can also leave each row with little evidence. The right review asks whether the structure preserves the important business differences, not whether it follows a universal arrangement for every local advertiser.

When comparing offers, select their corresponding actions and report periods consistently. Cost per emergency call and cost per installation quotation describe different acquisition tasks. The business can consider both when planning investment without pretending they should have the same acceptable ratio or that the cheaper immediate contact must represent more valuable work.

Inspect intent and exclusions

Use available traffic evidence to examine whether the offer attracts the intended request. A service campaign may receive informational searches, employment enquiries or requests for unsupported work. Those are candidates to investigate, not facts to assume. Read the returned search examples and relate them to the business's verified offer before recommending an exclusion.

A keyword filter can prioritise that review. Ask for keywords with meaningful spend and no selected primary outcome over an appropriate completed period. The result is a measured candidate population. It does not automatically prove that each keyword is irrelevant or that the business should pause it. A slow sales process, a different contact route or a limited sample can require further evidence.

The negative-keyword audit checks presence through supported campaign, group or attached-list sources. It does not certify the suitability of every exclusion. A term such as “commercial” can be essential for one service and irrelevant for another. Generic negative lists should not replace the actual offer and traffic review.

For a supported addition, review the exact text and target before confirmation. Shared relationships can affect more than the campaign being discussed. Preserve ambiguous terms for human review, and distinguish brainstormed candidates from demonstrated waste. The negative-keyword workflow provides action-specific detail while the local-service review supplies the commercial reason for the change.

Review availability before increasing demand

A low advertising acquisition cost can invite more spend, but operational capacity still matters. The business may have no appointment slots, a temporarily reduced team or an unavailable specialist. Confirm the constraint before increasing the volume of requests. Advertising data cannot establish the current service calendar unless the relevant evidence is actually supplied.

Ask the owner what can be handled and which offer is affected. A shortage for one type of work may not apply to every campaign. A business can still accept planned enquiries while urgent availability is limited. Communicate the distinction rather than treating a capacity problem as a reason to pause all advertising automatically.

Use the destination to set accurate expectations. If the business will respond later, say so. If a particular service is temporarily unavailable, do not continue inviting immediate booking for it without clarification. Creative assistance can help phrase the message, but the facts must come from the people responsible for delivery.

Document any supported campaign change and its reason. A capacity-related pause is different from a performance-related pause, and the condition for re-enabling it should be clear. The assistant should not infer that an asset can resume merely because yesterday's conversion ratio improved. Confirm that the operational constraint has actually changed before proposing the next action.

Use the business's acquisition economics

Acquisition ratios need a stable denominator and a business interpretation. Spend divided by primary leads is cost per recorded lead. Spend divided by primary calls is cost per recorded call. A named form action produces a narrower ratio. Show the components and use the measure that fits the decision rather than switching to a broader conversion total to produce a lower cost.

Different services can support different acquisition costs. A recurring commercial contract and a small one-off appointment have different economics. The account tool cannot determine contribution margin from the conversion name. Ask the business for the relevant context and keep it distinguishable from the advertising metrics returned by the provider.

Recorded values also require explanation. A call or form can have an assigned value without collecting payment. The broad primary conversion-value measure can include non-purchase actions. A value-to-spend ratio based on those assignments is not automatically revenue from completed jobs. Label it according to what the setup records.

A verified zero denominator makes the corresponding acquisition ratio undefined when spend is positive. An unavailable read does not establish that zero. Weighted totals should use compatible summed components rather than an equal average of campaign ratios. These simple rules make the economics discussion reliable enough for the owner to decide what additional evidence is needed.

Work through an invented account review

This illustration uses fictional names and figures, not customer results. Willow Service Team advertises routine maintenance and urgent assistance. Over fourteen completed days, Maintenance spends 600 currency units and records 20 selected form leads. Urgent Assistance spends 900 and records 30 primary call conversions. Both ratios happen to be 30, but they describe different contact routes and service promises.

The reviewer first confirms the actions and their triggers. The form count represents accepted submissions. The call action follows its configured call measurement; it is not automatically a count of answered suitable jobs. The office supplies an aggregate observation that some urgent callers requested work outside the confirmed area. That context prompts a targeted location review.

The current campaign targets and location option are inspected. Separately, a contact test finds that after-hours routing does not explain when a response is available. The team assigns the routing issue to its telephone owner and reviews whether the campaign schedule matches the immediate-response offer. The advertising ratio alone did not establish either problem.

A supported location proposal is prepared only after the desired scope is agreed. Its target and exact change are reviewed before confirmation. The follow-up retains the same form and call definitions, records when routing changed and checks the relevant completed period. This creates a reproducible review without claiming that one setting guarantees every enquiry will be inside the service boundary.

Ask focused account questions in Slack

Use Slack to ask the question you have defined. “Show primary calls and spend for Urgent Assistance over the last fourteen completed days” names the route, campaign and period. “Show the location settings for Maintenance” asks for configuration rather than performance. “Compare Contact Form with our confirmed website event” selects a specific measurement correspondence.

The reporting framework discovers available fields, validates the requested plan and retrieves evidence from the relevant source. It can preserve an explicit population in a follow-up. A new period or breakdown needs an appropriate read if the existing dataset does not contain it. A request for a daily graph should not be answered by inventing daily values from an account total.

Ask for creative work separately when that is the task. The assistant can suggest clearer service-area wording, headlines or contact expectations from the supplied business context. It does not need an unsolicited research call simply to brainstorm. Measured keyword history, current settings and live edits are different requests that require the appropriate tool evidence.

When a necessary definition is missing, a short clarification is useful. The assistant should ask which campaign, contact action or website event is intended rather than substituting a convenient broader total. See plain-English account questions for practical request structure. The advantage is an accessible route to specific evidence, not a claim that the bot knows every operational fact.

Use reports and monitoring for the next decision

A recurring report should use a stable contact definition and a period that fits the business's activity. An account-wide primary total can flag a broad change; a named service action can support a more specific review. Keep the dates and scope visible so the owner can distinguish a current completed window from a partial day or a previous month.

Choose a chart when movement over time matters and a table when comparing campaigns is the main decision. Native charts have bounded point and series counts. Longer periods can use a coarser breakdown when daily detail was not specifically requested. Reusing returned data for a requested visual is appropriate when the necessary points are already present.

Monitoring should prompt a useful investigation. A sudden absence of recorded calls can justify checking measurement and the contact route. A spend threshold can identify exposure. Neither signal explains the cause by itself. Unavailable data should retain its unknown state rather than being translated into a verified zero that triggers an unsupported conclusion.

Pacer monitors spend against its configured current period and uses completed account-time-zone data. It does not change live campaign budgets merely because a projection differs from the target. Saved Reporting and Alerts have their own configurations. Use the pacing guide and alerts guide to keep those jobs distinct from the business's decision about service capacity.

Review a supported edit before it is sent

Translate a confirmed finding into a specific request. “Improve local leads” is not a complete operation. The intended change may involve targeting, an exclusion, an asset status or clearer creative. It may instead belong to the website or telephone owner. The supported advertising tool should prepare only the operation that the available evidence and business decision justify.

The conversational workflow prepares a proposal before the live edit. Read the exact target name and any relevant parent context. Review each operation, particularly when the request contains dependencies or several steps. Internal resource identifiers are not needed to make the business decision, but the readable scope must make clear which campaign, group or asset changes.

Confirmation is the authorisation for that reviewed execution. The proposal expires after fifteen minutes, and relevant settings and authorisation are checked again. A visible result distinguishes verified completion from acknowledgement or incomplete verification. When an outcome is uncertain, inspect the account before repeating a creation or other operation that could produce an unwanted duplicate.

Record the reason and follow-up condition. A capacity pause, a scope correction and a performance test have different success criteria. The confirmation workflow describes the implementation in detail. For a local business, the additional discipline is to keep the operational promise aligned with the advertising state after the change.

Keep a repeatable local-service routine

Use a short routine that someone can actually repeat. Confirm the service area and any temporary constraints. Verify which primary calls and forms the selected report counts. Check the live contact route when a change or a suspicious movement calls for it. Then review the relevant traffic and acquisition metrics with those definitions in place.

Keep a decision record with the period, action set, important observations, owner and next review date. Note changes to service coverage, response hours and tracking. Those facts help the next person interpret a movement without reconstructing the whole business context from a long thread.

Assign each finding to the system that can resolve it. A targeting mismatch belongs to the advertising review. A failed form belongs to the website owner. Unanswered calls belong to the handling process. Unknown conversion data needs a measurement investigation. Several problems can coexist without proving that the account requires a sweeping rebuild.

Make the handoff specific enough to close

Each task should say what observation will establish that it is complete. A form-routing correction can be closed after the agreed test reaches the intended inbox and the confirmation behaves correctly. A targeting change can be checked against the requested current settings. A performance test needs its defined outcome and review window. These are different completion criteria, so a generic “fixed” message is insufficient for all three.

For a temporary closure, record who confirms reopening and which advertising scope is affected. For a coverage change, keep the revised area alongside the date it took effect. For a measurement repair, ask the tracking owner to document the trigger and whether historical figures remain comparable. The next account report can then use the correct context instead of rediscovering the same unresolved question.

The aim is a clear path from relevant interest to a handled enquiry. A useful report makes the evidence and limits readable, and a useful edit changes only the agreed scope. Keeping those responsibilities connected prevents the team from treating every contact as a completed customer or every weak day as a reason to rewrite the campaign.

Local-service advertising questions

Is this a guide to Google's Local Services Ads product?

No. It describes ordinary Google Ads account review for businesses serving local customers and the current ppctor workflows.

Does Presence guarantee that every enquiry is in the service area?

No. Provider location signals are imperfect, and the enquiry's actual service location still needs confirmation.

Is every telephone-link click an answered call?

No. Inspect the call-conversion setup and keep clicks, recorded calls and handling outcomes distinct.

Can a primary call have recorded value?

Yes. Explain whether that value is assigned or supplied through a recorded outcome; it is not automatically collected payment.

Does a clean destination check prove the form works?

No. Google's published destination findings and a live submission test establish different facts.

Should every location without conversions be excluded?

No. Review the service scope, period, volume and selected outcome before proposing an exclusion.