FM02: Targeting safety case

LUNA progress and final receipt

Status: COMPLETE for the authorized read-only FM02 audit. Fresh Google Ads provider evidence was collected for customer 4891907898 and the four assigned failure paths were tested. No campaign, location, schedule, network, keyword, negative, bid, budget, ad, website, GTM, GA4, or conversion mutation was issued.

Corrected goal preserved: nearby people with urgent vehicle-help needs, verifiable phone-widget clicks, and no inferred geography from names. Ads search clicks in this tracker are not phone-widget clicks, connected calls, or proof of a local urgent caller.

Acceptance status

Deliverable A: effective targeting map

The campaign names below are scope identifiers inherited from the supplied C01 receipt only. They are not evidence that a person, city, or route is serviceable. The provider fields, not names, determine what was actually configured.

Provider-wide target map: 89/89 current campaign criteria were ENABLED: 63 DEVICE rows, 21 LANGUAGE rows, one positive PROXIMITY row, and four campaign-level KEYWORD negatives. No active LOCATION row, no location exclusion, and no AD_SCHEDULE row returned. Each campaign has PRESENCE for positive and negative geo target type settings. Device rows are Desktop, Mobile, and Tablet, three per campaign. The 21 language rows all use languageConstants/1024. Networks are Google Search true, Search Partners false, Content false, and Partner Search false on every target campaign.

Budget context: the three current target budget resources total 199,998.56 HUF on the 30.4-day calculation. No budget was changed. The 13 Primary rows read ELIGIBLE and the eight Secondary/Generic rows read LEARNING or equivalent provider status. These status fields are recorded for delivery context, not used to infer geography and not re-verdict FM01 scenarios.

Location option interpretation: PRESENCE/PRESENCE is the provider setting readback, not proof that the actual service area is correctly represented. The only explicit target is the 40 km proximity row. The real service area still requires client/operations confirmation.

Current positive exact keyword inventory

All 86 enabled positive terms are exact match: 20 non-Generic scope campaigns have four each, and the Generic scope campaign has six. There are zero enabled positive PHRASE or BROAD rows, zero enabled target ad-group negative rows, 171 paused or historic non-removed keyword rows, and 33 removed keyword rows. The configured strings below are not user search data and are not used as geography proof.

EXACT | Biatorbágy | Secondary [4]: autómentés biatorbágy; autómentő biatorbágy; biatorbágy autómentés; biatorbágy autómentő
EXACT | Buda | Primary [4]: autómentés buda; autómentő buda; buda autómentés; buda autómentő
EXACT | Budakeszi | Secondary [4]: autómentés budakeszi; autómentő budakeszi; budakeszi autómentés; budakeszi autómentő
EXACT | Budaörs | Primary [4]: autómentés budaörs; autómentő budaörs; budaörs autómentés; budaörs autómentő
EXACT | Csepel | Secondary [4]: autómentés csepel; autómentő csepel; csepel autómentés; csepel autómentő
EXACT | Diósd | Primary [4]: autómentés diósd; autómentő diósd; diósd autómentés; diósd autómentő
EXACT | Generic | Budakeszi 40 km [6]: autómentés; autómentés 0-24; autómentő; azonnali autómentés; non stop autómentés; sos autómentés
EXACT | M0 | Primary [4]: autómentés m0; autómentő m0; m0 autómentés; m0 autómentő
EXACT | M1 | Primary [4]: autómentés m1; autómentő m1; m1 autómentés; m1 autómentő
EXACT | M31 | Primary [4]: autómentés m31; autómentő m31; m31 autómentés; m31 autómentő
EXACT | M5 | Primary [4]: autómentés m5; autómentő m5; m5 autómentés; m5 autómentő
EXACT | M6 | Primary [4]: autómentés m6; autómentő m6; m6 autómentés; m6 autómentő
EXACT | M7 | Primary [4]: autómentés m7; autómentő m7; m7 autómentés; m7 autómentő
EXACT | Piliscsaba | Secondary [4]: autómentés piliscsaba; autómentő piliscsaba; piliscsaba autómentés; piliscsaba autómentő
EXACT | Pilisvörösvár | Secondary [4]: autómentés pilisvörösvár; autómentő pilisvörösvár; pilisvörösvár autómentés; pilisvörösvár autómentő
EXACT | Pomáz | Primary [4]: autómentés pomáz; autómentő pomáz; pomáz autómentés; pomáz autómentő
EXACT | Solymár | Primary [4]: autómentő solymár; autómentés solymár; solymár autómentés; solymár autómentő
EXACT | Szentendre | Primary [4]: autómentés szentendre; autómentő szentendre; szentendre autómentő; szentendre autómentés
EXACT | Törökbálint | Secondary [4]: autómentő törökbálint; autómentés törökbálint; törökbálint autómentő; törökbálint autómentés
EXACT | Zsámbék | Secondary [4]: autómentő zsámbék; autómentés zsámbék; zsámbék autómentés; zsámbék autómentő
EXACT | Érd | Primary [4]: autómentő érd; autómentés érd; érd autómentő; érd autómentés

Overlap result: 86 unique normalized enabled exact strings across 86 rows, zero identical exact-string overlaps between target campaigns. This deterministic inventory check does not prove that every unobserved semantic variant can route.

Location exclusions, negatives, and shared controls

The four current campaign-level negatives are all ENABLED EXACT rows in the Generic scope: autómentő adok veszek, autómentő elado, autómentő webshop, and elado automento. No target location exclusion, no enabled target ad-group negative, no shared negative set, no shared criterion, and no target campaign shared-set association was returned. These are configuration findings only and no negative was added or removed.

Deliverable B: demand-routing proof

The complete 14-day evidence window is 2026-09-03 through 2026-09-16 in Europe/Budapest. The 2026-09-17 completeness probe returned 17 impressions, 2 search clicks, and 1,342.84 HUF, so that partial day is excluded from the comparison. The split at 2026-09-15 separates the earlier delivery history from the C01 exact reconciliation.

Search-term evidence: the read-only search_term_view query returned 213 rows and 197 distinct aggregate rows in the Generic scope plus 16 rows across nine non-Generic scopes. Ten target campaigns had any returned search-term row and 11 had none in this window. Every returned row used ADVERTISER_PROVIDED_KEYWORD as the match source and SEARCH as the network. The post-reconciliation slice returned 16 rows, 41 impressions, one Ads search click, and 1,076.35 HUF across seven scopes. Raw search-term text is deliberately not reproduced because the contract prohibits exposing user-level search data. A missing row is sparse evidence, not proof of zero demand.

Where demand can route: current enabled exact terms provide a controlled route for 86 configured strings. In the two-day post-reconciliation slice, seven target scopes produced search-term rows and one produced an Ads search click. Where it is blocked: the checked negative and overlap paths found no current target ad-group negative, shared negative, duplicate exact string, or current positive PHRASE/BROAD criterion that blocks a matching configured term. Where evidence is too sparse: 11 scopes had no search-term row in the full window and the current two-day local/route-labelled delivery was only 38 impressions and one Ads click. No low-volume result is classified as zero demand or as proof of a wrong service area.

Historical match-type boundary: the full-window Generic search-term aggregate contains EXACT 36, NEAR_EXACT 18, PHRASE 138, and NEAR_PHRASE 5 rows, while its current static criteria are six enabled positive EXACT rows and no enabled phrase/broad rows. The phrase and near-phrase rows are historical delivery before the 2026-09-15 reconciliation; they are not used to claim a current phrase keyword remains enabled. Post-reconciliation Generic rows are EXACT 3 and NEAR_EXACT 4.

Deliverable C: scenario verdicts, thresholds, owners, and reversible responses

Owners are role-based so no unsupported staffing or 24/7 assumption is introduced: the Ads owner owns the readback and any separately authorized account change; the client operations owner confirms the real service area and answer window; the measurement owner maintains the unique Google-Ads-originated phone-widget ledger and a privacy-safe call-outcome log. All responses below are recommendations for a later authorized decision, not actions taken in this read-only audit.

Failure-path receipt and reproducibility

Read-only queries were issued through GoogleAdsService.search only. No mutate service, paid-ad click, real call, synthetic conversion, credential output, caller detail, or public publication was used. The target ID set was taken from the supplied C01 receipt and rechecked against provider campaign IDs before analysis.

TARGET_IDS = 16650800567, 23974835198, 23974835201, 23974835204, 23974835207, 23974835210, 23974835213, 23974835216, 23979397390, 24246214449, 24246214452, 24246214455, 24246214458, 24257365147, 24257365150, 24257365153, 24257365156, 24257365159, 24257365162, 24257365165, 24257365168

Q1 identity
SELECT customer.resource_name, customer.id, customer.descriptive_name,
       customer.currency_code, customer.time_zone, customer.status,
       customer.manager, customer.test_account, customer.auto_tagging_enabled
FROM customer
LIMIT 1

Q2 campaign settings
SELECT campaign.resource_name, campaign.id, campaign.name, campaign.status,
       campaign.advertising_channel_type, campaign.advertising_channel_sub_type,
       campaign.bidding_strategy_type, campaign.bidding_strategy,
       campaign.manual_cpc.enhanced_cpc_enabled,
       campaign.target_spend.target_spend_micros,
       campaign.target_spend.cpc_bid_ceiling_micros,
       campaign.campaign_budget,
       campaign.network_settings.target_google_search,
       campaign.network_settings.target_search_network,
       campaign.network_settings.target_content_network,
       campaign.network_settings.target_partner_search_network,
       campaign.geo_target_type_setting.positive_geo_target_type,
       campaign.geo_target_type_setting.negative_geo_target_type,
       campaign.primary_status, campaign.primary_status_reasons,
       campaign.serving_status, campaign.bidding_strategy_system_status,
       campaign.start_date_time, campaign.end_date_time
FROM campaign
WHERE campaign.id IN (TARGET_IDS)

Q3 campaign criteria, including locations, schedules, devices, languages, and negatives
SELECT campaign.resource_name, campaign.id, campaign.name,
       campaign_criterion.resource_name, campaign_criterion.criterion_id,
       campaign_criterion.display_name, campaign_criterion.status,
       campaign_criterion.negative, campaign_criterion.type,
       campaign_criterion.location.geo_target_constant,
       campaign_criterion.language.language_constant,
       campaign_criterion.keyword.text, campaign_criterion.keyword.match_type,
       campaign_criterion.proximity.radius, campaign_criterion.proximity.radius_units,
       campaign_criterion.proximity.address.postal_code,
       campaign_criterion.proximity.address.province_code,
       campaign_criterion.proximity.address.province_name,
       campaign_criterion.proximity.address.country_code,
       campaign_criterion.proximity.address.street_address,
       campaign_criterion.proximity.address.city_name,
       campaign_criterion.proximity.geo_point.longitude_in_micro_degrees,
       campaign_criterion.proximity.geo_point.latitude_in_micro_degrees,
       campaign_criterion.ad_schedule.day_of_week,
       campaign_criterion.ad_schedule.start_hour,
       campaign_criterion.ad_schedule.start_minute,
       campaign_criterion.ad_schedule.end_hour,
       campaign_criterion.ad_schedule.end_minute,
       campaign_criterion.device.type
FROM campaign_criterion
WHERE campaign.id IN (TARGET_IDS)

Q4 target ad-group criteria
SELECT campaign.resource_name, campaign.id, campaign.name,
       ad_group.resource_name, ad_group.id, ad_group.name, ad_group.status,
       ad_group_criterion.resource_name, ad_group_criterion.criterion_id,
       ad_group_criterion.status, ad_group_criterion.negative,
       ad_group_criterion.type, ad_group_criterion.primary_status,
       ad_group_criterion.primary_status_reasons,
       ad_group_criterion.keyword.text, ad_group_criterion.keyword.match_type
FROM ad_group_criterion
WHERE campaign.id IN (TARGET_IDS)

Q5 campaign performance, complete-day window
SELECT campaign.id, campaign.name, segments.date,
       metrics.impressions, metrics.clicks, metrics.cost_micros,
       metrics.average_cpc, metrics.conversions, metrics.all_conversions,
       metrics.search_impression_share,
       metrics.search_budget_lost_impression_share,
       metrics.search_rank_lost_impression_share
FROM campaign
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q6 keyword performance, complete-day window
SELECT campaign.id, campaign.name, ad_group.id, ad_group.name,
       ad_group_criterion.criterion_id,
       ad_group_criterion.keyword.text, ad_group_criterion.keyword.match_type,
       segments.date, metrics.impressions, metrics.clicks,
       metrics.cost_micros, metrics.average_cpc, metrics.conversions,
       metrics.all_conversions
FROM keyword_view
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q7 search-term evidence, complete-day window
SELECT campaign.id, campaign.name, ad_group.id, ad_group.name,
       search_term_view.search_term,
       segments.date, segments.day_of_week, segments.device,
       segments.ad_network_type,
       segments.keyword.info.text,
       segments.search_term_match_type,
       segments.search_term_match_source,
       metrics.impressions, metrics.clicks, metrics.cost_micros,
       metrics.average_cpc
FROM search_term_view
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q8 user location delivery aggregate
SELECT campaign.id, campaign.name,
       user_location_view.country_criterion_id,
       user_location_view.targeting_location,
       user_location_view.resource_name,
       segments.date, segments.day_of_week, segments.device,
       segments.ad_network_type,
       segments.geo_target_city, segments.geo_target_district,
       segments.geo_target_most_specific_location,
       metrics.impressions, metrics.clicks, metrics.cost_micros,
       metrics.average_cpc
FROM user_location_view
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q9 geographic delivery aggregate
SELECT campaign.id, campaign.name,
       geographic_view.country_criterion_id,
       geographic_view.location_type,
       segments.date, segments.day_of_week, segments.device,
       segments.ad_network_type,
       segments.geo_target_city, segments.geo_target_district,
       segments.geo_target_most_specific_location,
       metrics.impressions, metrics.clicks, metrics.cost_micros,
       metrics.average_cpc
FROM geographic_view
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q10 location-view delivery aggregate
SELECT campaign.id, campaign.name, location_view.resource_name,
       segments.date, metrics.impressions, metrics.clicks,
       metrics.cost_micros, metrics.average_cpc
FROM location_view
WHERE campaign.id IN (TARGET_IDS)
  AND segments.date BETWEEN '2026-09-03' AND '2026-09-16'

Q11 shared negative controls
SELECT shared_set.resource_name, shared_set.id, shared_set.name,
       shared_set.type, shared_set.status, shared_set.member_count
FROM shared_set

SELECT shared_set.resource_name, shared_set.id, shared_set.name,
       shared_criterion.resource_name, shared_criterion.criterion_id,
       shared_criterion.keyword.text, shared_criterion.keyword.match_type,
       shared_criterion.type
FROM shared_criterion

SELECT campaign.resource_name, campaign.id, campaign.name,
       campaign_shared_set.resource_name, campaign_shared_set.shared_set,
       campaign_shared_set.status
FROM campaign_shared_set
WHERE campaign.id IN (TARGET_IDS)

Q12 partial-day completeness probe
SELECT segments.date, metrics.impressions, metrics.clicks, metrics.cost_micros
FROM customer
WHERE segments.date BETWEEN '2026-09-17' AND '2026-09-17'

Validation note: final HTML parsing and section-boundary checks passed. The exact report and this tracker remain private local artifacts. The report's ## Report section contains the same substantive receipt, scope, query row counts, failure-path results, thresholds, and ownership boundary. No public review-host publication was performed because the contract expressly prohibits publishing private evidence.