simulationPPC Auditoraccount checks, findings and a review score on invented Google Ads fixtures
PPC Auditor
A scripted audit on invented Google Ads accounts. Check counts, scores and playback timing belong to these fixtures; this demonstration does not query an account or call a model. Atlas’s auditor is in beta.
- 01Campaigns organized by product line
- 02Ad groups contain 10–20 keywords max
- 03SKAGs on top-performing keywords
- 04Brand vs non-brand campaigns separated
- 05Search and Display in distinct campaigns
- 06Remarketing campaigns configured
- 07Campaign naming convention is consistent
- 08Ad schedule configured for business hours
- 09Geographic targeting matches service area
- 10Bid strategy aligned with campaign goals
- 11Target CPA on conversion campaigns
- 12Target ROAS on revenue campaigns
- 13Manual CPC where automation is weak
- 14Device bid adjustments applied
- 15Location bid adjustments applied
- 16Audience bid adjustments configured
- 17Smart bidding learning period respected
- 18Portfolio bid strategies applied correctly
- 19Negative lists on all campaigns
- 20Campaign-level negatives in place
- 21Broad match has phrase/exact variants
- 22Long-tail keywords in every ad group
- 23Exact match keywords cover core terms
- 24Phrase match used for intent variants
- 25Search term report reviewed within 30 days
- 26Irrelevant search terms added as negatives
- 27Competitor keywords in isolated ad group
- 28Quality scores reviewed and actioned
- 29Responsive Search Ads in all ad groups
- 30At least 3 ad variations per ad group
- 31Dynamic keyword insertion used well
- 32Sitelink extensions on all campaigns
- 33Callout extensions configured
- 34Call extensions on phone campaigns
- 35Structured snippet extensions present
- 36Price extensions on product campaigns
- 37Promotion extensions for active offers
- 38Ad copy tone matches brand voice
- 39Account average Quality Score ≥ 7/10
- 40Landing page experience above average
- 41Expected CTR rated Above Average
- 42Ad relevance rated Above Average
- 43Low QS keywords (≤3) paused or reworked
- 44QS improvement actions documented
- 45Ad copy tightly matches keyword themes
- 46Landing page load time under 3 seconds
- 47Mobile landing page experience optimized
- 48Landing pages match ad message and offer
- 49Primary CTA visible above the fold
- 50Page load time under 3 seconds
- 51Mobile-responsive design confirmed
- 52Trust signals present on landing pages
- 53Form friction minimized (≤5 fields)
- 54Conversion path clearly defined
- 55A/B testing active on key landing pages
- 56UTM parameters properly configured
- 57Customer match lists uploaded
- 58Similar audiences enabled where applicable
- 59In-market audiences applied
- 60Remarketing lists for search configured
- 61Audience exclusions configured
- 62Demographics targeting optimized
- 63Life events audiences tested
- 64Custom intent audiences created
- 65Observation vs targeting used correctly
- 66Daily budgets set per campaign
- 67Pacing within ±10% of monthly target
- 68No budget caps hit during peak hours
- 69Shared budgets used where appropriate
- 70Monthly spend forecast within target range
- 71Budget split reflects campaign priority
- 72Seasonality adjustments planned
- 73Emergency budget controls documented
- 74Return on ad spend meeting targets
- 75Google Ads conversion tracking installed
- 76Google Analytics linked to Ads account
- 77Conversion actions match business goals
- 78GA4 conversions imported into Google Ads
- 79Phone call tracking configured
- 80Offline conversion imports set up
- 81Cross-device conversion tracking enabled
- 82View-through conversions configured
- 83Conversion window fits the sales cycle
- 84Auction insights report reviewed monthly
- 85Impression share above 60% on core terms
- 86Competitor ad copy analyzed
- 87Competitor landing pages benchmarked
- 88IS loss to budget tracked and actioned
- 89IS loss to rank tracked and actioned
- 90Keyword gap analysis completed
- 91Competitive bid strategy documented
- 92Market trend monitoring automated
One decision
Two modes. Quality is the default on the accounts we keep and runs everything. Stealth runs the hard checks on prospects we have not signed, so a pitch audit never became a second workflow.
What broke
The first version wrote a projected conversion lift into every report. Nobody could defend the number in a room, so the rule became: report what you checked, score it, and let the planner decide.
Scripted demonstration · invented data
The case study has the implementation and the limits of the evidence.