The AI Visibility Audit: How to Score Your Brand Across 6 Pillars

You can’t improve what you don’t measure. This is as true for AI visibility as it is for anything else in marketing. Yet most brands jump straight into tactics — seeding Reddit threads, pitching press, generating reviews — without first understanding where they stand. They’re firing in the dark.

The AI Visibility Audit gives you three things no amount of gut feeling can provide.

According to Joel House, founder of MentionLayer and author of AI for Revenue, "We\'ve run over 400 AI visibility audits at this point, and the single biggest predictor of campaign success is whether the brand started with a proper baseline. Teams that skip the audit waste an average of 3-4 weeks optimizing the wrong pillar."

First, a measurable baseline. Your composite AI visibility score is a single number that represents your brand’s total AI visibility posture. When that number is 27, you know exactly where you stand. When it’s 54 three months later, you know exactly how much progress you’ve made. Without this baseline, you’re asking clients (or your boss) to trust that things are "getting better" without proof. Our AI Visibility Index study includes a 6-dimension self-assessment framework that sorts your brand into one of four tiers — Invisible (0–25), Emerging (26–50), Present (51–75), or Dominant (76–100) — each with its own stage-specific 90-day action plan. That tiered approach is exactly the kind of prioritized diagnostic the 6-pillar audit produces at the pillar level.

Second, a diagnostic breakdown. The six pillar scores tell you *where* your gaps are. A brand scoring 15 on Citations but 65 on Reviews has a completely different action plan than a brand scoring 50 on Citations but 10 on Press. The audit doesn’t just tell you you’re behind — it tells you exactly which pillars need work and which are already strong.

Third, a prioritized action plan. After the audit completes, AI generates a ranked list of recommended actions based on impact-to-effort ratio. Quick wins (high impact, low effort) get prioritized first. This isn’t generic advice — it’s specific to your brand’s scores, your competitive landscape, and your available resources.

The audit is also a powerful sales tool. For agencies, running a free audit for prospects is the single most effective conversion mechanism we’ve found. The PDF report shows the prospect their gaps, compares them to competitors, and presents a specific action plan where every item maps to a service you provide. The audit sells the engagement for you.

Start your free audit to see where your brand stands. A full audit takes about 3-5 minutes to run and costs 50 credits. A quick audit (citations + AI presence + entities) takes 2-3 minutes and costs 20 credits. Run a full audit at onboarding and monthly thereafter. Run a quick audit weekly if you’re in the middle of an active campaign and need faster feedback loops.

The Citations pillar measures your brand’s presence in high-authority forum threads relative to your competitors. This is the most directly actionable pillar because it’s the one you can influence most quickly through citation seeding.

What it scans: The audit runs a SERP scan for your top 20 keywords, searching for Reddit, Quora, and Facebook Group threads ranking on Google’s first two pages. For each discovered thread, it checks the content for your brand name (exact and fuzzy match), your URLs, and your top competitors’ names.

What it produces: A complete inventory of threads where your brand is present, absent, or where competitors are mentioned. You’ll see the total thread count, your mention count vs competitor mention counts, a platform breakdown (Reddit/Quora/Facebook), and a ranked list of top opportunities — high-authority threads where competitors are present but you’re not.

How it’s scored: The formula weights three factors. Brand mention rate (what percentage of relevant threads mention you) accounts for 40% of the pillar score. Competitor dominance ratio (how much of the conversation competitors own vs you) accounts for 30%. Opportunity density (how many threads score above 70 on our relevance scale) accounts for the remaining 30%.

A brand with zero mentions across 142 threads typically scores 15-25 on this pillar. That sounds terrible, but it’s actually the best position to be in from an improvement perspective — 142 threads means 142 opportunities. A brand mentioned in 30% of threads with balanced competitor presence scores 50-65. A brand dominating their threads scores 80-95.

What a low score means: You have a massive gap that competitors are filling. Every thread where a competitor is mentioned and you’re not is a thread feeding AI’s recommendation of your competitor over you. The good news: this is the fastest pillar to improve. Twenty well-placed citation seedings can move this score 15-25 points in 4-6 weeks.

What to do about it: Use the Citation Engine to identify the highest-opportunity threads, generate response variants, and systematically seed your brand into the threads that matter most. Prioritize threads where competitors are mentioned (the gap is most visible to AI), threads ranking in Google positions 1-5 (highest authority), and threads with 20+ comments (highest engagement signal). Reddit is the most important platform for citation seeding given its outsized role in AI recommendations.

The AI Presence pillar is the north-star metric. It directly measures whether AI models recommend your brand when asked buying-intent questions about your category. This is what everything else feeds into — citations, entities, reviews, and press are all inputs that ultimately drive this output.

What it tests: The audit generates 10 buying-intent prompts from your keywords. Questions like "What are the best [category] services?", "Can you recommend a [service type] provider?", and "Compare the top [category] companies." Each prompt is sent to four AI models: ChatGPT, Perplexity, Gemini, and Claude. That’s 40 total tests.

What it measures: For each test, the audit checks whether your brand was mentioned at all (mention), whether it was cited with a link (link citation), whether it was positioned as a recommendation vs just listed (recommendation), which competitors were mentioned, and the AI brand sentiment of any mention (positive, neutral, negative). It also captures the full AI response for audit purposes.

"AI Presence is a lagging indicator by design — it\'s the output that proves all your other work is translating into actual recommendations," says Joel House. "When I see a brand\'s AI Presence score move from 5 to 25 in 90 days, that\'s the compound effect of citation seeding, entity cleanup, and press all hitting the consensus layer simultaneously."

How it’s scored: Overall mention rate (what percentage of tests your brand appeared in) accounts for 50% of the pillar score. Recommendation rate (what percentage of tests positioned your brand as a recommendation, not just a mention) accounts for 30%. Link citation rate (what percentage included your URL) accounts for 20%.

You can do a quick manual version of this test right now — see our 60-second AI visibility test. For benchmarking context, our AI Visibility Index study tested 1,004 businesses across 5 models: the top performer, Asana, scored 91/100 and appeared in all 5 models. The average business appeared in just 0.8 models. Only 4% appeared in all five. The numbers here are usually sobering for brands running their first audit. A brand mentioned in zero out of 40 tests scores 0. A brand mentioned in 2 out of 40 tests scores about 5-10. A brand mentioned in 20 out of 40 and recommended in 10 scores about 55. A market leader mentioned in 35 out of 40 and recommended in 25 scores about 85.

Why it gets 30% weight: This pillar carries the highest weight because it’s the direct measure of what matters: does AI recommend you? All other pillars are inputs — this one is the output. When a potential customer asks ChatGPT "What’s the best [your category]?" and your brand appears in the answer, that’s a high-intent touchpoint you didn’t have before. When it doesn’t appear, you’re invisible in a channel that influences an increasing share of purchase decisions.

What to do about it: AI Presence is a lagging indicator — it improves as the other four pillars improve. You can’t directly "optimize" for AI Presence the way you optimize a Google listing. Instead, improve your citation, entity, review, and press scores, and AI Presence follows. The typical timeline: 4-8 weeks after significant improvements to input pillars, you’ll see AI Presence scores start to move. Track it weekly via the Share of Model metric.

The Entities pillar assesses whether your brand’s identity is consistent and well-represented across the platforms that AI models reference when building their understanding of who you are. Think of it as your brand’s "identity card" in the AI ecosystem.

What it checks: The audit searches Google for your brand name and identifies your presence across key platforms: Google Business Profile, LinkedIn company page, Crunchbase, Wikipedia/Wikidata, industry directories, and social profiles (X, Facebook, Instagram). For each platform found, it extracts the brand description. It also checks your website for structured data: Organization schema, Product schema, FAQ schema, BreadcrumbList schema.

What it measures: Platform presence (are you on the platforms AI references?), description consistency (do all platforms describe your brand the same way?), information completeness (are all fields filled out?), schema markup coverage (is your website providing structured entity data?), and Knowledge Panel existence (does Google display a Knowledge Panel for your brand?).

How it’s scored: Platform coverage (percentage of key platforms where you’re present) accounts for 25%. Consistency score (how well your descriptions align across platforms) accounts for 35%. Schema completeness accounts for 20%. Knowledge Panel bonus adds 20 points if you have one.

Inconsistencies are the silent killer here. When your LinkedIn says "marketing platform" and your Google Business Profile says "advertising agency" and your Crunchbase says "ad tech company," AI models receive conflicting signals about what your brand actually is. The result: AI either picks the most common description (which might not be the one you want) or doesn’t mention you at all because it can’t form a clear consensus.

What to do about it: This is one of the fastest pillars to improve and often the cheapest. Audit every platform where your brand appears and align the descriptions. Use the same core language: same business category, same value proposition, same founding details. Add Organization schema markup to your website at minimum. Apply for a Google Knowledge Panel if you don’t have one. For a deeper dive into entity strategy, see our guide to entity SEO and knowledge graphs. Most brands can improve their entity authority score by 20-30 points in a single focused week of updates.

The Reviews pillar measures your brand’s presence, volume, rating, and recency across review platforms that AI models cite when answering comparison and buying-intent queries. Reviews are a critical trust signal — AI models treat verified user reviews as strong evidence of product quality.

What it scans: The audit checks for your brand on Google Reviews, Trustpilot, G2, Capterra, Yelp, and relevant industry-specific review sites. For each platform, it collects: total review count, average rating, most recent review date, and review velocity (reviews per month over the last 6 months). It runs the same scan for your top three competitors to provide comparison data.

How it’s scored: The formula weighs five factors. Platform coverage (percentage of key review platforms where you have presence) accounts for 20%. Rating score (average rating divided by 5, scaled to 100) accounts for 25%. Volume score (logarithmic scale: 0 reviews = 0, 10 reviews = 30, 50 reviews = 60, 200+ reviews = 100) accounts for 25%. Recency score (have you received reviews in the last 90 days?) accounts for 15%. Velocity versus your top competitor accounts for the final 15%.

The logarithmic scale for volume is important. Going from 0 to 10 reviews is a bigger score jump than going from 100 to 200. This reflects reality: having zero reviews on a platform is dramatically worse for AI visibility than having a modest number. The first 10-20 reviews on each major platform are the highest-leverage investments.

The competitor comparison is eye-opening. When we audit B2B brands, we frequently find review gaps of 10:1 or worse against the top competitor. One client had 23 total reviews across all platforms. Their top competitor had 4,200. That’s not a gap you close overnight, but it’s a gap you need to see to understand why AI recommends them and not you.

What to do about it: Start with the platforms that AI cites most frequently for your industry. For B2B SaaS, that’s G2 and Capterra. For local services, it’s Google Reviews. For e-commerce, it’s Trustpilot. Get your first 15-20 reviews on each priority platform. Build a review request process into your customer journey. Track review velocity weekly. Respond to every review (this signals active management to AI models). Focus on platforms where your competitors are strong and you’re absent — those are the exact platforms AI models are pulling comparison data from.

The Press pillar measures your brand’s earned media footprint — the third-party mentions, news articles, guest posts, podcast appearances, and thought leadership that AI models use as authority signals. Remember: 90% of LLM citations come from earned media. This pillar may only carry 15% of the composite score weight, but it punches well above that in terms of influence on AI recommendations.

What it scans: The audit searches Google News for your brand name (last 12 months), searches Google web for third-party mentions (excluding your own domain), and searches for your founder/CEO name plus brand name (thought leadership). Each mention is classified by type: news article, press release, guest post, podcast, award or list feature, or social mention.

What it measures: Total mention count, mention type distribution, publication authority (estimated by Google position and domain reputation), link ratio (percentage of mentions that include a backlink), unlinked mentions (brand name without link — these are outreach opportunities), mention velocity (mentions per month), and thought leadership presence.

"Unlinked mentions are the most overlooked asset in any brand\'s AI visibility arsenal," says Joel House. "Every unlinked mention is a warm outreach opportunity — someone already wrote about you, and converting that to a linked mention strengthens both your traditional SEO and your AI consensus signals."

How it’s scored: Mention volume (logarithmic scale) accounts for 20%. Authority score (weighted by publication quality) accounts for 25%. Diversity score (variety of mention types — news, guest posts, podcasts, etc.) accounts for 20%. Recency (freshness of mentions) accounts for 15%. Link ratio (percentage with backlinks) accounts for 10%. Thought leadership (founder/team personal mentions) accounts for 10%.

The authority scoring is particularly important. One mention in TechCrunch or Forbes is worth more than fifty mentions in low-authority blogs. AI models weight publication authority heavily when deciding which brands to recommend. A brand featured in three tier-one publications will outperform a brand with 30 mentions in sites nobody has heard of.

What to do about it: Press takes the longest to build but has the most durable impact. Start with the Digital PR strategy for the AI era. Identify unlinked mentions (brand references without links) and reach out for link additions. Pitch guest posts and thought leadership articles to industry publications. Launch a podcast appearance strategy. Build a press kit that makes journalists’ lives easy. Target 2-3 high-authority placements per month. Track mention velocity over time. Every press mention is a permanent addition to the consensus layer that AI models reference indefinitely. For the business case behind these investments, see our ROI of AI visibility analysis.

The composite AI Visibility Score combines all six pillars with their respective weights: AI Presence (30%) + Entities (25%) + Reviews (15%) + On-Page (10%) + Citations (10%) + Press (10%) = Composite Score (0-100). Here’s how to interpret your number.

15-35: Early Stage. Most brands land here on their first audit. You have significant gaps across multiple pillars. AI models either don’t know you exist or barely mention you. This is actually the best place to start from because the improvement trajectory is steepest — small investments yield large score gains. Focus on quick wins: entity cleanup (instant improvement), initial citation seeding (2-4 week impact), and getting your first reviews on key platforms (2-3 week impact).

35-55: Developing. You have presence in some areas but significant gaps in others. AI models might mention you occasionally but don’t recommend you consistently. This is where prioritization matters most. Your audit action plan will show which pillars are holding you back. Often it’s one or two pillars dragging the composite score down while others are relatively healthy. Target those weak pillars aggressively.

55-75: Competitive. You’re in the game. AI models mention you regularly and recommend you some of the time. Your competitors are still ahead in some areas. At this stage, the focus shifts from filling gaps to building advantages. Increase citation seeding volume, pursue higher-authority press placements, build review velocity, and start targeting platform-specific optimizations for each AI engine.

75+: Leader. You’re among the top brands AI recommends in your category. AI models mention you frequently and recommend you strongly. At this stage, the focus is on defense: maintaining review velocity, continuing press cadence, monitoring competitor activity, and ensuring entity consistency as your brand evolves.

The action plan structure. After the audit completes, the system generates 5-8 prioritized actions ranked by impact-to-effort ratio. Each action specifies which pillar it addresses, what specifically to do, expected impact (high/medium/low), effort required (high/medium/low), estimated timeline, and which MentionLayer module handles it. Quick wins always come first — actions like "Fix entity inconsistencies across 3 platforms" (high impact, low effort, done in a day) before "Launch press campaign targeting 5 industry publications" (high impact, high effort, 4-week timeline). See how it works inside MentionLayer.

Monthly re-auditing is how you prove ROI. Run the full audit monthly. Store each audit as a snapshot. The dashboard shows your composite score trending over time, with each pillar score tracked independently. This trend line is the single most powerful proof of progress for clients and stakeholders. "Your AI Visibility Score went from 27 to 58 in 90 days" is a concrete, measurable result that justifies continued investment. The 90-Day Playbook outlines the typical trajectory and milestones you should expect.