Google Ads location-performance audits for agencies: interpret presence results and approve precise targeting changes

Define the client's geographic question

A location-performance review should begin with the client's actual business concern. The agency may need to understand where recorded value comes from, investigate acquisition cost outside a service area or prepare a campaign for a new region. Those questions require different evidence. A country performance table alone does not establish the correct target list for every campaign.

Write the intended decision and selected scope before collecting additional data. A useful brief might ask whether the selected campaign's geographic reach still matches the client's delivery coverage. Another might simply ask which locations contributed most conversion value in a completed period. Preserve that distinction rather than turning every geographic report into an exclusion proposal.

Identify the business owner of the service-area facts. An agency can inspect advertising settings, but it should not invent which regions accept bookings or receive deliveries. The client may have recently changed its fulfilment policy, and the account's historical activity does not prove that policy is still current.

The deliverable should name the observed performance, relevant evidence limits and exact next decision. It can recommend further campaign investigation, retain a deliberate targeting policy or prepare a supported location change. A useful review does not need to promise recovered spend or perfect geographic accuracy.

Agree on the geographic level in the brief as well. A client asking about countries should receive a country comparison, while a regional expansion question may need finer evidence. This prevents the agency from presenting a convenient mixed table and leaving the owner to interpret its inconsistent scale. The selected level should follow the decision, and any necessary change in level should be stated plainly before the report is used to justify a live setting.

Distinguish presence, interest and configured targets

Google's geographic reporting reference distinguishes actual user-location and area-of-interest rows. The current ppctor account reader selects the presence type. Its performance population therefore should not be described as a combined list of every place a user expressed interest in.

A configured target is another concept. It identifies a geographic criterion chosen for a campaign, while the presence report describes selected recorded activity. A country in the performance table is not by itself proof that the campaign explicitly targets that country as an individual criterion.

Location-targeting mode is also separate. Reading presence-type performance does not establish that every campaign's positive targeting mode is set to Presence. To answer the configuration question, inspect the actual campaign field rather than infer it from this reporting filter.

The agency should keep these concepts visible in the explanation. Reported presence, configured location list and positive targeting mode each answer a different question. Combining their labels can make a legitimate report look like a targeting error and can lead to the wrong live edit.

A useful client question is whether the concern is about observed activity or intended eligibility. If the owner says a region should no longer receive advertising, inspect the relevant current setting. If the owner asks where last month's conversions were reported, preserve that historical reporting question. The same place name can appear in both discussions while referring to different evidence, so conversational context should retain the purpose as well as the name.

Understand the current location reader

The current account helper requests geographic country rows for the selected interval and filters them to reported presence. It collects impressions, clicks, cost, conversions and conversion value through the shared metric aggregation. These base metrics support the informational account location table.

The reader resolves returned country criterion references into names through a separate geographic-constant lookup. If no country keys are returned, it produces an empty location collection. It does not manufacture a row for every possible country or every campaign target.

When US country activity is present, the helper requests a further regional breakdown and can substitute those returned rows for the country row. Other countries remain at country level in this compact path. That asymmetry matters when the client compares the displayed locations.

The informational card does not score the account or offer a direct performance-based fix. Its purpose is visibility into the selected geographic population. A follow-up configuration change needs its own exact target, relevant current setting evidence and supported preparation path.

Preserve the reporting interval and population

The helper applies the requested date segment with the configured account time zone. State the actual interval when explaining the result. A completed rolling window, current partial day and calendar month are different reporting populations even when a brief Slack request uses informal wording.

The account location query does not filter the report to today's enabled campaigns. Historical activity can therefore remain relevant even when a campaign is now paused. Current status and period activity should be recorded separately when a proposed edit relies on either fact.

Use the same scope for comparisons across periods or tools. A campaign-specific report cannot be compared directly with the account table as though both contain identical activity. If a difference matters, reconcile the period, geographic level and conversion metric before diagnosing a reporting defect.

Configuration inspection time has its own role. A current target list does not reconstruct the exact list used for every historical click. A review of past reach needs appropriate historical evidence when the current setup changed during the interval.

Review the US regional substitution

The helper requests US region-segmented presence metrics when it has a US country row. If relevant regional rows are returned, it removes that country row and inserts the regional collection. It does not display the country total and its regional parts together as independent additive locations.

The regional query uses the provider's region segment. The current implementation resolves names but does not separately select and verify every returned constant's target type. The compact interface calls this a state breakdown; do not extend that label into a claim that the helper independently validated a complete fifty-state inventory.

Rows without a usable returned region key are skipped by aggregation. The regional collection also keeps rows with positive cost or impressions. A nonempty result can therefore be a selected collection without proving that every US country metric has been preserved in the replacement.

If the optional US regional read returns no retained rows or its request throws, the account reader keeps the already measured country row. This best-effort fallback preserves known country evidence without claiming state detail was available. A failure of the primary country read still leaves that report unavailable; the optional fallback cannot invent country metrics that were never obtained.

Establish coverage before using the total

The performance table calculates its All row from the displayed rows. When a US country row has been replaced by a selected regional collection, that total depends on the collection's coverage. It should not automatically be declared identical to a separate complete account total.

A missing regional key can prevent some activity from appearing in the selected breakdown. The current reader does not reconcile every regional sum with the original US country metrics before substitution. The agency should establish that relationship when an exact total or complete geographic distribution is required.

This limitation does not make every returned regional metric unusable. It changes the scope of the conclusion. The agency can describe a selected regional row accurately while avoiding an unsupported statement that the collection contains all US activity or every possible state.

Keep country and regional reporting levels explicit in a client export or conversation. A total based on the displayed mix is a useful summary of that mix. A requested full-account reconciliation needs the appropriate underlying evidence and cannot be solved merely by removing a caveat from the headline.

For an exact reconciliation, compare the relevant base metrics under the same period and location-type filter before calculating ratios. A cost total, conversion total and conversion-value total can each need coverage assessment. Matching one of them does not automatically establish every other measure. Keep the result of that assessment attached to the selected report so the next operator does not treat a partial geographic collection as a permanently certified account total.

Handle unresolved geographic names

The geographic-name lookup is a display enrichment. It uses returned references to request readable names. A missing name does not automatically invalidate a successfully returned metric row, but it does leave the client-facing geographic identity unresolved.

The current helper can fall back to a Country or Region label with a raw reference suffix. Treat that as a current presentation limitation rather than a desirable client output. The agency should resolve the actual location name before using the row as an ordinary approval target.

Country names and similarly named subregions can be ambiguous in conversation. A live proposal needs the actual geographic target, with canonical context where relevant. The assistant should not choose one based solely on a short label that fits several possible places.

Keep technical references behind the scenes once the identity is resolved. The client needs a readable country, state or intended radius, together with the affected campaign. Internal geo IDs are useful for precise preparation, but they should not become the ordinary report's explanatory text.

Canonical context is particularly useful when a short name is shared by several targets. The owner can review the intended country and region without inspecting provider references. Once the selection is established, retain it for follow-up preparation rather than asking the same ambiguity question repeatedly. A later request for another similarly named place remains a new identity question; previous conversation context should help resolve it without substituting an unrelated target.

Interpret value, ROAS, CPA and volume

The current table uses account classification to choose its additional metrics. Ecommerce presentation includes conversion value and ROAS; the other presentation includes conversion rate and CPA. Both retain clicks, impressions, cost and conversions as base context.

The All row sums base metrics before calculating ratios. It does not average location ROAS or CPA values. The agency should use the same weighted arithmetic when explaining a geographic combination so a small location does not receive the same influence as a much larger one merely because both have one row.

The summary identifies the location that drove the most value or conversions by absolute volume. That is different from the highest return or lowest acquisition cost. A large country and a small region can have very different scale, and the summary does not measure the marginal effect of additional spending.

Use actual currency context rather than inferring it from the compact dollar symbol. Keep ordinary numeric presentation to at most two decimal places, and retain undefined denominators as such. A location without spend does not have a defined ROAS simply because the table needs a comparison cell.

Keep conversion definitions explicit

The geographic reader selects standard conversions and conversion value. It does not separately enumerate every contributing action in the compact table. Do not rename the value purchase-only revenue or the count qualified leads without establishing the requested conversion scope.

A business can have several primary conversion actions, including actions with assigned monetary values. The client's intended aggregate should determine the report. The agency should not silently remove legitimate actions or turn a simple geographic comparison into an unsolicited measurement-repair project.

If the request names a particular conversion or asks for comparison with a specific Analytics event, use the corresponding scoped evidence. The country table's general conversion total cannot be relabelled as that named event merely because its number appears plausible.

Downstream sales quality remains another question. A higher conversion rate does not establish that the location produced more profitable customers. The agency can assign a follow-up business-data review where needed, while preserving the actual Ads metric as the current evidence.

Check the client's actual service area

Geographic eligibility should reflect the client's current operations. Confirm where the business accepts bookings, fulfils orders or supports customers. A region with recorded conversions can still be outside a newly changed service policy, while a low-volume region can remain commercially important.

Do not assume the service area from billing information, headquarters or website language. Those details can differ from the intended advertising market. The client supplies the business constraint; the agency verifies how the campaign's actual targeting relates to it.

Several offers in one account can have different geographic rules. An online course and an in-person workshop need not share the same reach. This is another reason to investigate the contributing campaigns before preparing a broad location change from an account summary.

Ask the client to distinguish a commercial preference from an operational restriction. A region that is less profitable can still accept customers, while an area outside booking coverage may require a direct eligibility decision regardless of a favourable historical ratio. These situations should not share an automatic exclusion rule. The agency can recommend a performance investigation for the first and prepare the exact approved service-area correction for the second when the required settings are supported.

Record a clear exception when the business deliberately advertises beyond a narrow physical service area. The decision may be valid for its actual offer. Future reviewers should not repeatedly exclude that reach simply because one generic account policy suggests a smaller region.

Compare geographic levels fairly

The compact collection can mix whole countries with US regional rows. A country-versus-state comparison is not a comparison of equally sized markets. Absolute value and spend should be interpreted with that scale difference in view.

Efficiency ratios remain useful where their underlying metrics are defined, but they do not remove every population difference. Campaign mix, customer intent and available volume can vary by location. A smaller high-return row does not automatically show that a larger region should lose all reach.

When the client asks for a country ranking, obtain a country-level population rather than treating regional substitutions as country counterparts. When it asks for US states, establish the relevant regional coverage. Preserve the requested level instead of accepting whatever mixed display is easiest to reuse.

Use additional geographic detail only when it helps the decision. A broad market question may not need city or postal detail. A local service-area question may need it. The agency should avoid both insufficient scope and an unnecessary full hierarchy scan.

Separate geographic granularity from display length. A short Slack reply can summarise a properly scoped country analysis without hiding that it used country data. A detailed regional request may need a readable selected table rather than dozens of narrative sentences. The formatting choice should preserve the analytical population, not change it. If the client needs the complete requested collection, provide the appropriate supported report surface and retain the same definitions used in the conversational summary.

Investigate campaign contributions

Account locations can contain different campaign contributions. Brand search, prospecting and distinct offers may distribute differently across regions. A geographic row therefore does not isolate the causal effect of one target criterion.

A focused follow-up can ask which selected campaigns contributed to the concern or inspect one named campaign's geographic results. Preserve the original period and metric unless the user changes them. That keeps the additional evidence connected to the actual decision.

Current target settings help establish what an edit would change, but they are not a complete explanation of historical location performance. A campaign's current Presence field and target list should be read as configuration facts, separate from the period's recorded activity.

One campaign can need a targeting correction while the rest of the account remains appropriate. The proposal should name that scope and exact effect. Do not convert a broad informational table into a silent account-wide location rewrite.

Respect the location evidence boundary

Google's geographic-targeting guidance explains that location uses several signals and does not guarantee perfect accuracy. The agency should not present a reported presence row as an independently verified residence, delivery address or exact physical coordinate.

The location report is also not a customer-address list. It provides advertising performance in its selected geographic segmentation. A requested comparison with actual shipping or booking destinations requires the relevant business data and a deliberate compatible scope.

A location outside the expected service area deserves investigation without automatically proving a broken setting. Establish the actual campaign target list, positive targeting mode, relevant exclusions and report population. These observations can explain different aspects of the concern.

Keep the client conclusion precise. The agency can state what the selected report shows and what configuration was read. It should not promise that a new Presence setting will certify every visitor's address or eliminate every possible geographic discrepancy.

Distinguish unavailable and empty reports

The audit's report phase can show loading while the location read is pending. Its completed failure mapping records an unavailable location report separately from other successful account reports. A failed geographic request should not silently become a confident table of zero outcomes.

A successful empty collection causes the informational location card to be hidden. It does not prove that every possible target recorded no activity. The helper presents its returned geographic population rather than a manufactured complete list of locations.

Name-resolution failure and metric-read failure also differ. A label can be unresolved while metrics arrived, whereas a failed report leaves the performance observation unavailable. The agency should resolve the relevant condition before ranking or preparing a setting change.

Zero denominators are not provider failures. CPA without conversions and ROAS without cost have no defined ratio. State the condition instead of repeatedly querying until a numeric-looking answer appears. Only changed actual evidence can support a different calculation.

Follow an invented agency regional review

This case is invented. Brook Vale Media manages Oak Harbour Lessons. The client asks whether its Local Workshops campaign reaches places outside its newly agreed service area. The agency begins with the selected account location view but does not treat that mixed collection as the campaign's configured target list.

The specialist notes that the US row was replaced by regional data and establishes the requested geographic coverage before presenting a complete distribution. The client confirms its actual booking area, while the agency obtains the named campaign's relevant configuration.

The campaign's target list and positive targeting mode become two separate observations. Oak Harbour Lessons requests a supported update to its intended location criteria and a separate Presence change. The proposal states both logical effects rather than claiming one automatically performs the other.

The location helper preserves existing exclusions and other retained positive criteria unless the requested operations explicitly remove them. The agency displays the exact additions and removals in human-readable steps. No provider write occurs while the owner reviews the pending proposal.

After confirmation, the execution flow checks the declared resulting configuration. The receipt establishes the supported target-list and mode changes. It does not rewrite historical geographic activity or guarantee that every future visitor has an independently verified address.

The handoff retains the original service-area concern, report scope, selected campaign, exact approved effects and technical outcome. Any later performance review uses its own period and metric. The invented example demonstrates an accountable targeting decision rather than a promised improvement.

Brook Vale Media also checks the scope of its completion message. It names Local Workshops and the approved changes, leaving other campaigns outside the claim. When Oak Harbour Lessons later asks about an online offer, the specialist establishes that offer's different market brief instead of copying the local service-area decision. The first receipt remains accurate because it described the actual campaign effect. This gives the client continuity without allowing one successful edit to become an unsupported account-wide targeting conclusion.

Separate Presence settings and target lists

The current convenience catalogue has a campaign action that sets positive geographic targeting to Presence. This changes the declared mode field. It does not itself add a country, remove a radius or replace the campaign's existing positive location criteria.

The location-list action is different. It prepares selected positive location and proximity additions or removals and retains the Presence setting. Combining the two actions can be appropriate when explicitly requested, but their effects should remain separate in the approval layout.

Do not infer either setting from the account performance filter. The report intentionally requests presence-type rows, regardless of what current mode evidence has been read. A live configuration answer needs the relevant campaign field and criteria.

Actual campaign eligibility remains necessary. The existence of a convenience action does not certify every possible campaign and provider condition. Current validation should establish that the prepared mode request is applicable before the owner can receive a trustworthy execution result.

Prepare supported location additions and removals

The current location helper reads existing nonremoved positive location and proximity criteria for the exact campaign. Removal references must belong to that selected population. It does not accept an arbitrary criterion from another campaign merely because the user supplied a plausible geographic name.

Requested geographic additions resolve enabled provider targets and use their readable canonical or ordinary names. Existing retained location references are checked so the helper does not add the same reference again unnecessarily. Target identity matters more than a short ambiguous place label.

The action verifies the intended resulting positive geographic set and proximity collection. Its description explicitly retains location exclusions and Presence settings. The word change should not be interpreted as clearing every existing target when the actual requested operations are additive.

A replacement request therefore needs explicit removals as well as additions. Show the complete intended effect before approval. If the current tool cannot prepare a requested negative location edit through this path, select a genuinely supported alternative instead of pretending the positive-criterion helper performs it.

Review what remains after removal as carefully as what is added. Removing one criterion can leave other positive criteria in place, including a broader retained area or radius. The current helper preserves those retained entries rather than interpreting an informal place name as permission to clear the whole list. The owner needs to see the actual intended result so a small removal is not mistaken for a complete geographic restriction.

Review address radii and store prerequisites

The address-radius helper has its own input shape: address, positive radius and stated miles or kilometres. Its local limit is at most 500 in the chosen units. That is the current application policy, not a universal statement about every provider campaign's radius eligibility.

The helper stores the supplied street-address form and recognises a trailing two-letter country field in its supported preparation. It does not promise a complete automatic address-normalisation or independently verified geocoding service. The client should supply enough geographic context for the intended target.

A radius around a linked business location uses a separate workflow. The catalogue distinguishes creation of a location-group prerequisite from adding a radius around an existing populated group. Provider population is asynchronous, so the assistant should not silently promise that creating the prerequisite makes an immediate later target ready.

Keep these prerequisites visible when relevant. An ordinary country-target update does not require store-location setup. Conversely, a store-radius request should not be replaced with an unrelated address radius merely to avoid its actual dependency.

State the unit with every radius in the approval description. The same numeric distance in miles and kilometres represents a different intended reach. The client's brief should determine the unit, and the resulting criterion should preserve it. A readable name for the centre also matters; a distance without a sufficiently clear address is not a reviewable business decision. Provider validation and configuration readback remain separate from verifying that the chosen centre matches the client's actual premises.

Present readable owner-approved effects

Name the campaign and actual geographic effects in ordinary language. Additions, removals, radius values and a Presence change should be clear without internal criterion IDs. The owner should not have to infer scope from a long technical request body.

Show each logical action as a visible step in the supported confirmation. If the complete geographic redesign needs narrower proposals for safe review, explain that requirement before writing. Do not silently execute only part of the intended redesign and report the whole brief as completed. The approved scope should remain clear in the result.

The pending message states that no changes have been made yet and the proposal expires in fifteen minutes. Only the connected owner's confirmation permits the live edit. A report, explanation or geographic brainstorming request does not itself authorise a mutation.

Current validation checks applicability before execution. Result verification then checks the declared requested state. These distinct safeguards should not be confused with the performance analysis that motivated the change. The proposal remains an exact configuration decision.

Verify settings and reconcile uncertainty

The final executed-and-verified receipt establishes the declared resulting fields and relevant relationships. It does not certify perfect location accuracy, service fulfilment or improved return. Keep the technical success prominent while preserving the separate business review.

A visible change-in-progress state acknowledges confirmation while provider work continues. A delayed Slack notification is different from a failed mutation. The owner should not need to click again simply because the result takes several seconds.

If the edit was sent but complete readback is unavailable, reconcile the exact current campaign and durable execution record before another attempt. Some settings may already have changed. Preserve separate outcomes when one logical step was verified and another remains uncertain.

A later restoration request is a new supported edit requiring current identity and exact intended state. Do not promise universal rollback from a vague instruction to undo the location review. The known earlier configuration can inform the proposal without authorising blind replay.

Present useful geographic tables and charts

A geographic table compares selected locations within a period. A daily chart answers a trend question and needs daily evidence. The assistant should not manufacture individual dates from a country total or treat a mixed country-region collection as a daily time series.

Keep geographic level visible in the presentation. A country comparison should use country data; a regional comparison should establish its selected regional population. The client should not have to discover that one row is a whole country while another is only one US region.

Slack native chart composition supports a readable maximum of three selected metrics or segments. Longer time series must also respect the twenty-point-per-series limit. Use a stated coarser breakdown when daily detail was not requested, and explain a necessary split when the user explicitly requires more daily points.

Readable names, dates and precision matter more than technical resource references. Keep current-year date labels short and ordinary metrics at no more than two decimal places. If the geographic identity remains unresolved, establish it before presenting a confident target-specific conclusion.

Continue the location discussion in Slack

A focused opening can ask which reported locations produced the highest return in a selected period. The assistant should preserve the metric and report scope, then state any mixed-level or coverage limitation. Most value and highest ROAS are different rankings.

A follow-up about a named campaign narrows the population. Retain the conversation's relevant period and business context, but obtain the appropriate campaign evidence for its new claim. Reusing the account summary is helpful for context, not proof of the campaign's exact target list.

Conceptual explanations and creative market planning can remain conversational when the required facts are already established. Actual current targets and measured performance need relevant provider reads. Avoid redundant calls just to restate a returned number or explain the difference between a country and a radius.

The final client message should name the observation, decision and known result. Keep user mentions before the final feedback text and controls. Progress messages should be visible without a user mention, and raw IDs, technical payloads and undefined parents should remain outside the ordinary reply.

Use the established conversation to reduce repetition while preserving factual boundaries. A follow-up such as apply that setting can refer to the precise previously described campaign change if no ambiguity remains. The assistant still needs current preparation checks, but it should not ask the owner to reconstruct the entire service-area brief. If that phrase could instead mean a report filter or a different campaign effect, ask the narrow missing question before preparing a live mutation. Context helps resolve the decision; it does not replace exact approval.

Leave a precise agency handoff

Record the client's service-area brief, selected period, reporting source and geographic level. Include the current target-list and mode evidence only where it was actually read. Another operator should know which observations support the recommendation.

Keep verified configuration, geographic reporting and fulfilment operations as separate milestones. A successful campaign edit should not close an unresolved delivery-policy question. A later report discrepancy should be investigated with its actual scope rather than used to erase the established technical result.

Retain deliberate geographic exceptions with their business owner and reason. Future reviewers should not repeat an unsupported exclusion proposal because a broad account table looks uneven. The actual campaign objective can explain a valid difference in reach.

Give the next specialist one concrete continuation point. It might be a regional coverage reconciliation, a named campaign configuration read or a later performance assessment. Naming that remaining observation is more useful than leaving a generic investigate-locations note. It keeps provider activity tied to a real question and lets the client understand which part of the review is finished and which part still requires evidence.

Use the agency audit workflow for broader context, the device-performance review for another reporting dimension and the approval workflow for precise live decisions. Each keeps its own evidence boundary.

Answer common location-review questions

Does this report list every configured campaign target?

No. It reports the selected geographic presence population.

Does a presence-filtered report prove Presence targeting mode?

No. The campaign's actual mode field needs its own read.

Does the informational card change the audit score?

No. It provides performance visibility without a scored fix.

Does the US regional replacement prove complete country coverage?

No. The current helper does not reconcile every replacement collection to the country total.

What happens when optional US regional detail fails?

The account reader keeps the already measured US country row. That preserves country evidence, not state detail. Failure of the primary country read remains an unavailable report.

Does adding a country also set Presence mode?

No. The current location-list action retains that separate setting.

Does the positive location helper edit exclusions?

No. Its declared path retains location exclusions.

Does verified targeting certify every visitor's address?

No. It establishes the requested configuration, not independent address verification.