AI marketing reporting: verify numbers before commentary
Give an AI reporting tool a defined export and calculation rules, then reconcile its numbers before reviewing its prose. Keep clicks, sessions, orders and revenue distinct; report refunds and attribution limits, and ask the model to identify unknown causes instead of inventing explanations.
Read article
Search Console AI report: what impressions tell you
The dedicated Search Console generative AI report measures visibility through impressions. Use it to inspect pages and markets under consistent filters. Keep overall search clicks, referral sessions and purchases in separate records; they cannot supply a dedicated AI click-through rate without a corresponding AI click metric.
Read articleShopify custom app logs: investigate incidents without exposing customer data
Shopify's Developer Dashboard now shows filterable logs and health metrics for custom apps. Learn a safe diagnostic method that avoids exposing customer data.
Read articleAI search visibility: distinguish brand mentions, citations and clicks
A practical guide to separating brand mentions, citations and clicks in AI search, with a dated sampling protocol, tool variability checks and a diagnostic worksheet.
Read article
Llms.txt and Google AI features: what the documentation actually requires
Google Search ignores llms.txt. Separate its documentary role from AI feature eligibility and access controls before deciding what to change on your site.
Read articleAI Brief Expands to French: What Advertisers Can Prepare During Closed Beta
Google is extending the closed AI Brief beta to French.
Read article