TIER 01 · DIAGNOSE · AUDIT BEFORE IMPLEMENTATION
Diagnose — measure before you fix.
AI Visibility audits across 5 engines · $250 to $750 · 48 hours to 5 days.
ANSWER · AEO-EXTRACTABLE
The Diagnose tier delivers AI Visibility audits at three depth levels. Pricing transparent · scope locked · deliverables contractually binding.
D1 Quick Check — $250 · 48 hours. Single-engine surface scan. We probe ChatGPT on 5 priority queries · capture raw output · score Citation Stack L1 + L2 health (Access + Structure). For buyers who want a fast read on whether AI Visibility is a problem before investing in the canonical audit. PDF artifact delivered within 48 hours.
D2 Full Snapshot — $500 · 3-5 days. Canonical audit. Most engagements start here. Cross-engine probe across 5 AI engines (ChatGPT · Perplexity · Gemini · Google AI Overviews · Claude) on 20 commercial queries. Full Citation Stack 4-layer health score. Competitor benchmark on 3-5 primary competitors. 10 prioritized findings with severity ranking. Documented PDF audit artifact with raw AI-engine outputs preserved.
D3 Emergency — $750 · 24-48 hour rush. Active-crisis triage. When AI engines are actively misrepresenting your brand · when sudden visibility drop hits · when PR exposure requires immediate cross-engine assessment. Same depth as D2 + crisis-specific recommendation · delivered on rush timeline.
Every Diagnose SKU is Proof Ladder P1 only — deterministic diagnostic. We measure · we report · we name the priority. Fix work happens on the Optimize tier (separate engagement · separate pricing). Continuous monitoring happens on the Dominate tier.
All findings backed by raw AI-engine output. All claims BOX-graded. No opinion without evidence. Most buyers start with D2.
Why Diagnose exists.
The first principle of vAEO is from M6 of the methodology: Accuracy over performance. Measure first · fix second.
Most AEO vendors skip the audit and sell the fix. That sequence costs money: you spend on optimization work without knowing which Citation Stack layer is actually broken. Two-thirds of brands we audit have a primary blocker at L1 Access or L2 Structure — schema misconfigurations, crawler allowlist gaps, SSR rendering issues. A content optimization sprint without that L1 + L2 fix delivers zero AI Visibility uplift.
Diagnose tier is the gate. Three SKUs · three depth levels · same principle: show you what AI engines see · cite · ignore · misrepresent about your brand. Backed by raw output. Cross-checked across 5 engines (single-engine probing is theater · per M3 of the methodology).
«Audit before implementation. You don't fix what you can't see. We make you see it.»
Three SKUs. Pick your depth.
All three SKUs deliver Proof Ladder P1 — deterministic diagnostic. They differ in scope, engine coverage, and timing.
QUICK CHECK
ChatGPT only · 5 queries · L1+L2 health · entry-level surface scan
FULL SNAPSHOT
5 engines · 20 queries · 4-layer health · 10 findings · most engagements start here
EMERGENCY
D2 depth + crisis-specific scope · 1h review call · active brand crisis triage
D2 Full Snapshot — what's actually inside.
D2 is the canonical entry point · most vAEO engagements start here. Worth showing the inside in detail.
Day 1-2 · Cross-engine probe
We probe 5 AI engines (ChatGPT · Perplexity · Gemini · Google AI Overviews · Claude) on 20 commercial queries relevant к your brand category. We capture raw output · structured для cross-engine comparison. Engine-by-engine: did your brand appear · was it cited correctly · was it misrepresented · was a competitor cited in your place.
Day 2-3 · Citation Stack 4-layer assessment
Each layer scored 0-100:
- L1 Access: robots.txt audit · crawler allowlist (GPTBot · ClaudeBot · PerplexityBot · Google-Extended · Bingbot · others) · SSR rendering check · llms.txt manifest · sitemap freshness · Lighthouse performance
- L2 Structure: JSON-LD schema validation (Google Rich Results Test · 0-error requirement) · semantic HTML5 audit · canonical URL integrity · OG + Twitter Card coverage · entity graph consistency
- L3 Semantics: Answer Capsule presence and quality · content sandwich
pattern audit · statistics density · expert quotation presence ·
<dfn>definition discipline · keyword stuffing detection (Princeton GEO -10% penalty avoidance) - L4 Authority: External authoritative source link count · sameAs network completeness · cross-engine entity reinforcement check · BOX-graded claims audit · freshness signal
Day 3-4 · Competitor benchmark
We probe 3-5 of your primary competitors across the same 20 queries · same 5 engines. You receive comparative Share of Model · which competitors dominate which queries · why they're cited (their L1-L4 strengths · gleaned from public observation).
Day 4-5 · Prioritized findings + artifact assembly
10 prioritized findings · severity-ranked (blocker · major · minor) · effort-estimated (quick-win · medium · long-term) · expected-impact-marked (SoM uplift estimate). Documented PDF artifact assembled. Raw AI-engine output preserved as appendix.
Day 5 · Delivery + 30-min review call (optional)
Audit artifact in your inbox by end of day 5. Optional 30-min review call к walk through findings + answer questions. Then you decide: engage Optimize tier · escalate к Dominate retainer · walk away с the diagnostic.
Anchor example — Barcelona Boat Rental
The Barcelona Boat Rental case is a D2 Snapshot delivery. Pre-audit measurement (2026-04-23 · cross-engine probe · 20 commercial queries · 5 engines): BBR Share of Model 19% · dominant competitor Click&Boat 44% · BBR cited zero times in Perplexity for top 4 commercial queries. That artifact informed the full Citation Stack repair sequence currently in flight.
«Five days · five engines · twenty queries · ten findings. Decision-ready audit artifact · not a deck.»
Which SKU for your situation.
One screen · pick your row.
| Your situation | SKU | Price | Timeline |
|---|---|---|---|
| You want a fast surface read before committing to canonical audit | D1 Quick Check | $250 | 48h |
| You don't know if AI engines mention your brand at all | D2 Full Snapshot | $500 | 3-5d |
| You need decision-ready audit before authorizing fix budget | D2 Full Snapshot | $500 | 3-5d |
| Your brand was just mentioned wrongly in a viral ChatGPT thread | D3 Emergency | $750 | 24-48h |
| Your AI Visibility dropped in last 14 days · need triage | D3 Emergency | $750 | 24-48h |
| CMO board meeting tomorrow · audit required by then | D3 Emergency | $750 | 24-48h |
| You already have a credible audit · ready to fix | → Optimize tier | ||
| You're an agency selling AI Visibility к your clients | → Partnership Track | ||
«Pick the row that matches your week · not the one that matches your ambition.»
What Diagnose delivers · and what it doesn't.
Explicit Proof Ladder framing per Methodology M9:
Diagnose tier = Proof Ladder P1 only. Deterministic. Today.
You get:
- Verified measurement across 5 AI engines (cross-checked · reproducible · raw output preserved)
- Citation Stack 4-layer health score (0-100 deterministic)
- Prioritized findings (severity · effort · expected impact)
- Documented PDF artifact (yours · audit-able · contract-ready)
- Recommended next-step SKU sequence (Optimize · Dominate · neither)
You don't get:
- Mechanical fixes. Diagnose is assessment · not implementation. Fix work happens on Optimize tier (separate engagement · separate pricing). Anyone selling «free audit with implementation included» is bundling the audit price into hidden fix-tier markup.
- 30/60/90 day re-measurement. That's Proof Ladder P3 · happens on retainer (Dominate tier). One-off Diagnose = single-point-in-time snapshot.
- Business outcome attribution. Proof Ladder P4 · downstream business impact · not promised by any honest AEO vendor.
- Endless «discovery deck» upsell. Pricing locked · scope locked · timeline locked. We deliver what we promised · then you decide.
«We name what we deliver · we name what we don't. Anyone bundling 'free audit' is hiding the price somewhere downstream.»
Citation Stack coverage per SKU.
Each Diagnose SKU assesses a specific scope of the Citation Stack model.
| L1 Access | L2 Structure | L3 Semantics | L4 Authority | |
|---|---|---|---|---|
| D1 Quick Check | ✓ assess | ✓ assess | — | — |
| D2 Full Snapshot | ✓ assess | ✓ assess | ✓ assess | ✓ assess |
| D3 Emergency | ✓ assess | ✓ assess | ✓ assess | ✓ assess |
D1 = L1 + L2 only. ChatGPT-only surface scan · sufficient для entry-level read · NOT sufficient для cross-engine assessment or content/authority work.
D2 + D3 = full Citation Stack 4-layer assessment. Both cover all four layers · across all 5 engines. D3 differs от D2 only in timing (24-48h rush vs 3-5d standard) and crisis-specific scope.
«Four layers · all required. D2 covers all four. D1 covers the two that block everything else.»
Frequently asked.
Can we skip Diagnose and go straight to Optimize?
Yes · if you already have a credible cross-engine audit from elsewhere. We need the data · not the audit vendor's name. If your audit covers Citation Stack L1-L4 health scores · cross-engine measurement · prioritized findings · we accept it as the assessment input. If it doesn't · we run D2 first. No exception.
What data do you need from us upfront?
Minimum: 5-20 priority queries (we help shape them if needed) · 3-5 primary competitor names · access к GSC (read-only) · access к robots.txt + sitemap.xml (public URLs · no credentials). For D3 Emergency: brief on the active crisis · timeline · scope of misrepresentation observed.
Do you need access к our site analytics?
Google Search Console read-only access · helpful для understanding current organic baseline · not strictly required. We don't need Google Analytics · we don't need CMS access · we don't need staging environment access. Diagnose work is observation-based on public surface.
What AI engines do you audit?
D1 = ChatGPT only. D2 + D3 = ChatGPT · Perplexity · Gemini · Google AI Overviews · Claude. We can add Brave · Mistral · Bing Chat · other engines on request для Dominate retainers · but those are not part of standard Diagnose scope.
How reproducible are the results?
Citation Stack 4-layer health scores (L1 + L2) are deterministic — we re-run the same audit on the same site and get the same numbers. L3 + L4 cross-engine measurement is reproducible within probabilistic bands — AI engines reweight constantly · output can shift 10-15% between probes 7 days apart. We preserve raw output as audit artifact so you can audit our audit.
Most engagements start с D2. Pricing locked at $500. Scope locked at 5 engines · 20 queries · 4 layers · 10 findings. Timeline locked at 3-5 days.
Then you decide what's next.
«$500 · five days · cross-engine probe · decision-ready artifact. Then you decide.»