AI search is reshaping how B2B brands find the partners who make them visible in ChatGPT, Perplexity, Gemini, and Claude — and Retina Media sells exactly that visibility, which makes its own discoverability the proof of its thesis. Before we run the audit, we need to make sure we're asking the right questions about the right competitors to the right buyers. This document presents what we've learned about Retina Media's market — your job is to tell us what we got right, what we got wrong, and what we missed.
Before we measure how often AI engines cite Retina Media in GEO-consultancy and AI-visibility queries, these three signals tell us whether AI crawlers can reach, parse, and trust the site today.
B2B buyers are increasingly opening their vendor research inside ChatGPT, Perplexity, Gemini, and Claude rather than Google, and the brands that establish citation visibility early gain a compounding first-mover advantage — AI platforms learn to trust and re-cite domains they already surface. Retina Media operates inside this exact shift: it is a founder-led GEO consultancy that sells AI visibility to B2B brands, which means the way AI engines represent Retina itself is both its product and its proof.
This Foundation Review presents three things we need to validate together before the audit runs: the competitive landscape that determines which head-to-head queries we construct, the buyer personas whose search intent shapes how those queries are phrased, and the technical baseline that determines whether AI crawlers can access and parse Retina's content in the first place. Each section below is an input you can confirm or correct — the audit is only as accurate as the assumptions it runs against.
The validation call is a working session with real stakes. It resolves two kinds of decisions: input validation (are the right buyers, competitors, and capabilities in the right tiers?) and engineering triage (which technical fixes can start before the results even come back?). The Pre-Call Checklist at the end of this document captures every one of those decisions in a single place so you can prepare in one pass.
Purpose This is the foundation for your GEO audit. Everything below — the competitors, the buyer personas, the capabilities, and the buyer frustrations — becomes the input that generates the buyer queries we run across AI platforms to measure where Retina Media shows up in the GEO-consultancy and AI-visibility space, and where competitors show up instead.
Your Job Read each section and tell us what we got right, what we got wrong, and what we missed. We've built this from public signals and inference — you know your actual deals, your real buyers, and the names that come up when you win and lose. The purple boxes throughout flag the specific places where your judgment changes the audit.
Confidence Badges Every entity carries a confidence badge. High means it's drawn from concrete site content or strong signals. Medium means it's inferred or sourced from third-party listings and warrants a closer look at the call. Treat medium-confidence items as the first place to spend your validation time.
→ Retina is a solo, founder-led shop (segment: startup), but the case studies are enterprise and scale-up brands — IDVerse, SoFi, Zenreach, Udacity. When a buyer searches, are they looking for a boutique GEO specialist or an enterprise-scale agency? If the former, we weight authority and specialization queries; if the latter, capacity and scale move to center — and broader full-funnel players like Directive and iPullRank become the real head-to-head set.
5 personas: 2 decision-makers, 1 evaluator, 2 influencers — the buying committee whose search intent shapes how every GEO-consultancy query is phrased.
Critical Review Area Personas are the highest-leverage thing to get right — they determine the intent behind every query we generate. If we've mischaracterized who actually evaluates, demos, and signs, the entire query set drifts toward the wrong buyer. Scrutinize the roles, the veto power, and especially the two medium-confidence personas below.
Data Sourcing Note All five personas are LLM-inferred — Retina is a small consultancy with no public G2 or review footprint, so there are no reviewer titles or case-study quotes to source from. Names, roles, seniority, and veto power reflect the standard B2B marketing buying committee; the specific composition for your deals is exactly what we need you to confirm. Treat every persona here as a hypothesis to validate.
→ Diana and Elena (Founder/CEO) both carry veto power. Building on the segment question above: if your deals actually close with the CMO, we lead with board-pressure and ROI-defense queries; if the founder signs, those shift to existential-visibility framing. Which seat signed your last five deals?
→ Marcus evaluates editorial fit; Tom (SEO) evaluates technical fit. In a GEO purchase, which one actually gates the shortlist — because that decides whether we weight content-quality queries or crawlability/technical queries in the evaluation cluster.
→ Priya is medium-confidence and rated an influencer. Does demand-gen actually sit in your GEO buying committee, or is GEO owned entirely by content and SEO? If demand-gen isn't in the room, we drop her pipeline/lead-volume queries and reallocate to the founder and CMO clusters.
→ Tom is your only high-technical evaluator. Does he champion GEO as the heir to SEO, or defend the existing SEO program? If he's a skeptic, we add defensive queries comparing GEO against traditional SEO ROI and crawlability depth (where iPullRank is strong).
→ Elena is medium-confidence as a distinct buyer. At Retina's enterprise and scale-up clients (SoFi, Udacity), does the founder actually engage, or is this a pure marketing-org purchase? If founders aren't buyers at that size, we merge her into Diana and drop the founder-authority query cluster.
Missing Personas? These roles sometimes appear in GEO-consultancy deals — do they show up in yours? Head of Brand / Communications (if accuracy of how AI describes the company is owned by brand rather than growth); PR / Comms lead (if executive-authority and thought-leadership programs are bought out of comms, not marketing); RevOps / Marketing Ops (if attribution and measurement is owned by ops and they gate the "prove the ROI" conversation). Each new role that genuinely shows up warrants its own query cluster. Who else is in the room when you win?
5 primary + 4 secondary competitors across two GEO market segments — full-service agencies and GEO monitoring platforms.
Why Tiers Matter Tier assignments decide which competitors get dedicated head-to-head queries ("Retina Media vs. Animalz," "best GEO agency for B2B SaaS") versus broad category-awareness queries. Four of the five primaries — Directive, Skale, iPullRank, and First Page Sage — are medium-confidence, drawn from third-party GEO-agency roundup listings rather than your won/lost deals. Getting these tiers right determines which ~30–40 head-to-head queries test direct competitive differentiation rather than general category visibility.
→ Three things to confirm: (1) Do the four medium-confidence primaries — Directive, Skale, iPullRank, First Page Sage — actually show up when you win or lose? They came from agency roundup listings, not your deals; any that don't appear should drop to secondary, pulling ~6–8 head-to-head queries each out of the differentiation set. (2) Profound sits in secondary as a "buy-vs-hire" tool — if your buyers genuinely weigh a self-serve monitoring platform against hiring you, Profound (or adjacent tools like Evertune / AthenaHQ) should move to primary, which reshapes the head-to-head set toward tool-vs-consultant queries. (3) Is anyone here irrelevant, or is a name missing entirely?
11 buyer-level capabilities mapped and rated outside-in, from strong to weak — these determine which capability queries the audit runs.
Show me exactly where my brand does and doesn't show up when buyers ask AI — run real buyer queries and tell me how invisible we are.
Tell me who AI cites in my category across ChatGPT, Perplexity, Gemini, and Claude — and how that differs platform to platform.
Don't just hand me a report — actually build the content that gets us cited and keep shipping it every month.
Make sure AI describes us accurately and compellingly — sharpen our story so the answers reflect how we actually win.
Position our founder and execs as the quotable, trusted voices AI surfaces in our space.
Connect AI visibility to actual deals — show me this moved pipeline, not just citation counts.
Make sure AI crawlers can actually reach and parse our content — fix the technical plumbing that blocks ingestion.
Do you have a repeatable, proven system for this — or are you making it up as you go?
I need to watch our AI visibility move over time, not just get a one-time snapshot audit.
Can a boutique shop actually handle our volume, or will we outgrow them and need a bigger agency?
Is there a dashboard my team can log into and run ourselves, or is everything dependent on the consultant?
Prioritize the Strengths The audit tests all 11 capabilities, but competitive differentiation queries will emphasize 3. Six are rated Strong — which of these best represents where Retina Media actually wins deals?
• AI Visibility Audit & Query Simulation
• Cross-Platform Citation & Share-of-Answer Analysis
• GEO Content Production & Execution
• Strategic Positioning, Messaging & Narrative
• Executive Thought Leadership & Authority Programs
• Proprietary GEO Methodology & Frameworks
→ Two ratings to pressure-test: Delivery Scale & Team Capacity and Self-Serve Tooling are rated Weak — honest for a solo shop, but is scale actually a deal-loss reason against Skale or Directive, or do your buyers value the senior-strategist model enough that it's a non-issue? And Technical Crawlability Audit is rated Moderate — accurate next to iPullRank's engineering depth, or underrated? These ratings decide where we play defense in capability queries. Anything missing, or any two capabilities we should merge?
10 pain points: 6 high, 4 medium severity — the buyer frustrations whose language becomes the audit's query phrasing.
→ Severity and language to check: is "competitors becoming the default AI-cited authority" really high-severity for your buyers, or is the sharper version "a competitor became the default answer and I can't dislodge them"? And three pains we'd expect in GEO deals but didn't capture — do they come up? Miscategorization ("AI recommends us for the wrong use case, not just with wrong facts"); prior-agency burn ("we already paid an SEO/content agency and saw nothing — why is this different?"); internal blame ("leadership thinks the marketing team dropped the ball on AI"). Each one that's real becomes its own query phrasing.
What our technical analysis found about whether AI crawlers can reach, parse, and extract Retina Media's content. These are engineering and content-operations fixes — not content strategy, which comes after the audit measures citation visibility.
Start Here — No Critical Blockers, Two High-Severity Items The good news first: robots.txt is permissive and AI crawlers can reach the site. The two items worth engineering and content attention now are the near-empty splash homepage (Marketing — add crawlable body content and a full nav/footer) and the stale case studies and aging blog (Content — institute a refresh cadence). One verification item is worth front-loading: confirm the homepage's sparse render is an intentional splash and not client-side rendering hiding content from crawlers — re-fetch it with JavaScript disabled. If the homepage is CSR-dependent rather than intentionally minimal, that finding's severity rises sharply.
What we found: The root URL (retina.media/, which serves /home) renders as a minimal splash page: a single H1 ("When buyers ask AI who to trust, we make sure it's you."), an H2 tagline, ~20–25 words of body text, and exactly one outbound link ("SEE HOW WE DO IT" → /geo). No primary navigation, footer, service descriptions, or internal link graph is present in the rendered output. Deeper pages (/geo, /cited, the blog) render full navigation and body content, so this is an intentional splash design, not a site-wide rendering failure.
Why it matters: The homepage is the single most-crawled URL on any domain and the most common landing point for AI crawlers resolving a brand entity. A crawler that fetches retina.media/ receives almost no extractable text to associate with the brand and only one internal link to follow, starving crawlers of the link graph they use to discover and weight the rest of the site. For a GEO consultancy whose entire value proposition is being correctly represented in AI answers, a content-free homepage undercuts the brand's own entity signal at the most important crawl entry point.
Recommended fix: Add substantive, crawlable body content to the homepage — a concise plain-language description of what Retina Media does and who it serves, the core service offering, and a full primary navigation + footer link block (GEO, About, Cited, Blog, Portfolio, Contact) so crawlers can reach the rest of the site directly from the root. Keep the splash aesthetic if desired, but ensure real text and links are present in the served HTML, not only behind interaction.
What we found: Every customer/portfolio case study carries a sitemap lastmod older than one year: /idverse (2025-04-29), /zenreach (2025-04-09), /sofi (2024-05-30), and /southern-company (2020-04-06). Across the 34 content-marketing pages (30 blog posts + 4 case studies), category freshness averages 0.39: 8 pages updated within 90 days, but 15 pages (44%) carry date signals older than 180 days and 4 are confirmed older than 365 days. The product/service pages are healthier (avg 0.68; /geo, /cited, /home all updated 2026-03-16).
Why it matters: AI engines, ChatGPT in particular, concentrate citations on recently updated content. Ahrefs found AI-cited content is ~25.7% fresher on average than non-cited content across platforms (August 2025), and ConvertMate found 76.4% of ChatGPT's most-cited pages had been updated within the last 30 days. Case studies are prime evidence assets for vendor-evaluation queries, and stale, undated proof points are deprioritized relative to competitors' fresher content. The aging long tail of the blog steadily falls out of the dominant citation window even though the underlying analysis remains strong.
Recommended fix: Institute a refresh cadence for evergreen, commercially important pages: re-date and lightly update the case studies with current outcomes and a visible "last updated" date, and prioritize refreshing the highest-value evergreen blog posts (the GEO white papers, atomic-content, how-b2b-buyers-build-shortlists) on a rolling schedule so flagship assets stay inside the 90-day window. This is a recurring content-operations commitment, not a one-time fix.
What we found: Two substantive blog posts render with an H1 only and no H2/H3 subheadings, presenting as a single uninterrupted narrative: "The Next Trillion-Dollar AI Opportunity is GTM" (~1,200 words) and "The AI That Threatened to Expose an Affair Explains Itself" (~1,200 words). Separately, the case studies use generic, non-descriptive section labels ("THE STORY," "THE EXECUTIONS," "THE ASK," "THE RESULTS") that carry no standalone meaning, and the homepage exposes only an H1 plus a single H2.
Why it matters: LLMs use heading structure to segment a page into self-contained, citable passages, and descriptive noun-phrase headings double as passage labels that help a model decide what a section answers. A long post with no subheadings is far harder to excerpt cleanly, and generic labels like "THE STORY" give a model no signal about what the passage contains. The bulk of the blog already does this well (descriptive H2/H3 hierarchies), so these are isolated outliers worth correcting rather than a systemic problem.
Recommended fix: Add descriptive H2/H3 subheadings to the two unstructured posts, breaking them into labeled passages. On the case studies, replace generic section labels with descriptive headings that name the client, the problem, and the measurable outcome (e.g., "IDVerse: 49% organic traffic lift in year one").
What we found: The case study and portfolio pages are very thin: /southern-company (~110 words), /sofi (~180 words), /zenreach (~220 words), /idverse (~260 words), and the /portfolio index (~100 words). Content-depth scores for these pages fall below 0.5; /southern-company and /portfolio fall below 0.4. Most consist of a short brand description plus a one-line statement of work and a credits list, with little specific, claim-bearing detail (IDVerse is the exception, carrying concrete metrics).
Why it matters: Thin pages give an LLM nothing specific to cite when answering buyer questions about proof, outcomes, or a vendor's track record. A 110–200 word case study that states what was done but not what resulted, with what numbers, cannot support a citation in an evaluation query. These are commercially important pages (social proof for a consultancy), so the depth gap directly limits how the brand's track record can surface in AI answers.
Recommended fix: Expand each B2B case study into a structured, outcome-led narrative: the client's situation, the specific work performed, and quantified results with dates. Treat the recent B2B engagements (IDVerse, Zenreach/Adentro, SoFi) as priorities and consider retiring or de-emphasizing the older creative-only portfolio pieces that no longer reflect the GEO positioning.
The following items could not be assessed through our analysis method (rendered markdown). We recommend your engineering team verify these manually before the validation call.
What to check: Our analysis fetches rendered page content as markdown, not raw HTML, so JSON-LD schema blocks are not visible to this method. We could not confirm the presence or correctness of structured data (Organization, Article, FAQPage, Book, BreadcrumbList) on any page. Squarespace injects some baseline schema by default, but page-appropriate, fully populated schema cannot be verified from our output. High-value opportunities worth confirming: FAQPage on /cited and /geo, Article (with author and dated publish/modify fields) on every blog post, Book on /cited and /dwell-in-a-place-of-yes, and Organization/Person for the founder's authority entity.
Recommended action: Verify structured data with Google's Rich Results Test, view-source, or Screaming Frog with JS rendering. Confirm Article schema (with datePublished/dateModified and author) on blog posts, FAQPage on /cited and /geo, and Book schema on the book pages; add or correct any that are missing or under-populated.
What to check: Because our method reads rendered markdown rather than raw HTML, we cannot assess meta descriptions, Open Graph / social-card tags, canonical URLs, or meta robots directives, and we cannot definitively rule out client-side rendering on individual pages. The deeper pages (/geo, /cited, the blog) returned full body content, indicating server-rendered content rather than a site-wide CSR problem — but the homepage's near-empty render should be re-checked with JavaScript disabled to confirm the splash is intentional rather than CSR-dependent.
Recommended action: Spot-check meta descriptions and OG tags with a social-preview tool or view-source, confirm the canonical tag strategy for the root vs. /home (both resolve), and verify the homepage and a sample of pages render their content with JavaScript disabled (e.g., Screaming Frog comparing raw vs. rendered HTML).
Method Note Schema coverage could not be scored — our analysis reads rendered markdown, which doesn't expose JSON-LD blocks, so all 40 pages are unscored on that metric (see the Manual Verification Checklist above). Structural-reference pages also lack scorable date signals (2 pages unscored for freshness). Everything else reflects a full 40-of-40 page sample; this is not a partial crawl.
Why Now GEO visibility is a time-sensitive, compounding opportunity:
• AI search adoption is accelerating — 94% of B2B buyers now use LLMs during the buying process (6sense, November 2025), and buyer discovery patterns are shifting quarter over quarter.
• Early citations compound: domains AI platforms learn to trust now get cited more frequently as training and retrieval data accumulate.
• Competitors who establish GEO visibility first create a structural disadvantage for late movers in the same category.
• GEO consulting is still early-innings in its own AI optimization — acting now means competing against inaction, not against entrenched strategies.
The full audit will measure Retina Media's citation visibility across real buyer queries in the GEO and AI-visibility space — from "best GEO agency for B2B SaaS" and "Retina Media vs. Animalz" to "who can get us cited in ChatGPT" and "how do I measure AI visibility." You'll see exactly which queries return answers that name your competitors but not Retina — and what it would take to appear in them. Fixing the Layer 1 items now (homepage content, case-study freshness, schema) improves the baseline before we even start measuring, so the audit captures the lift rather than the gap.
A 45–60 minute working session to walk through this document — confirm the buyers, competitors, capabilities, and pain points, and resolve the open decisions in the Pre-Call Checklist.
We translate the validated foundation into a buyer query set and run it across the selected AI platforms — ChatGPT, Perplexity, Gemini, and Claude — capturing who gets cited and who doesn't.
A visibility analysis, competitive positioning breakdown, and a prioritized three-layer action plan — the content and technical work ranked by which gaps actually cost you citations.
Start Now — Before the Call Three technical items your team can begin immediately, no audit results required: (1) Engineering: verify the homepage renders without JavaScript to confirm the splash is intentional rather than CSR hiding content from crawlers; (2) Engineering: verify JSON-LD schema (Article with author/dates on blog posts, FAQPage on /cited and /geo, Book and Organization) using Google's Rich Results Test; (3) Marketing/Content: add crawlable body content and a full nav/footer to the homepage, and begin re-dating the four case studies. These don't depend on the rest of the audit and will improve your baseline visibility before we even measure it. Note: robots.txt is already confirmed permissive — no crawler-blocking action is needed there.
Two jobs before we meet. The questions on the left require your judgment — no one knows your business better than you. The engineering and content tasks on the right don't require the call at all.