The layout of the full audit report. Numbers are anonymised from a Dutch SME audit in June 2026. Each locked module card below spells out what it covers. Pick a plan to see them populated for your brand.
Where this brand appears, it appears first (position #1 across all engines). But uneven presence (0–71%) plus zero third-party corroboration means most customers asking AI for this category never see it at all.
Average position #1 across every engine where mentioned. The job to be done is volume of presence, not quality of presence. Recent locale improvements lifted Gemini from 34% to 67%; everything else still sits in the 20–40% band.
Engines synthesise across multiple sources. With only the brand's own domain to draw on, they hedge or stay silent. This is the root cause behind the uneven presence AND the cluster of AI mistakes. Without independent sources to triangulate, the engine is improvising.
Engines are repeating contradictory facts the brand publishes about itself: the supported-store count appears as 12 / 13 / 14 across different pages, and one competitor is in-or-out depending on the source. Locking one canonical fact source collapses 36 of 38 flagged claims at once.
Counted only on questions that don't name any brand, so every mention here is earned.
Average position when mentioned: #1 across all engines. When it lands, it lands first.
Each cell = how often that engine mentioned the brand on that topic. Deeper coral = mentioned more often. Red-outlined zeros = content gaps.
| Topic | Gemini | Claude | Perplexity |
|---|---|---|---|
| Price comparison | 84 | 60 | 55 |
| Supported supermarkets | 71 | 44 | 50 |
| App reliability | 38 | 22 | 18 |
| Privacy | 12 | 8 | 5 |
| Family / multi-user | 0 | 0 | 0 |
Read: Family / multi-user has zero coverage across every engine. That's a content gap.
Each of these is populated from your data: your questions, your AI engines, the sites AI trusts about you, your fix plan. The preview titles below tell you exactly what they cover.
Same layout as the sample above. Populated with your questions, your AI engines, your fix plan. Each audit runs in about 90 minutes.