Profound Raised $96M to Show You Your AI Score. Here's What to Do When It's Low.
Profound just hit unicorn status tracking AI brand citations. Most brands running the tool for the first time are finding gaps. Here's the earned media playbook for closing them.
Profound raised $96 million this week at a $1 billion valuation. Backed by Lightspeed, Sequoia, and Kleiner Perkins, the AI visibility tracking platform now works with 700+ enterprise companies — including 10% of the Fortune 500 — to show marketers how frequently AI engines cite and recommend their brand.
The technology works. The data is real. And for most of the companies now seeing their scores for the first time, the reaction is the same: lower than expected, and unclear on what to do about it.
Here's the playbook.
First, understand what's actually moving the needle.
Profound CEO James Cadwallader said it plainly to Fortune: "In the future, every company on the planet will care deeply about how AI talks about and surfaces their brand." What the platform shows is your current citation frequency — how often ChatGPT, Gemini, and Perplexity mention your brand in answers about your category. What it doesn't show is what drives those citations at the source.
The source is not your schema markup. It's not your FAQ sections. Those matter, but they're table stakes. AuthorityTech's analysis across 200+ client programs shows that 82–89% of AI-generated answers cite third-party earned media — Forbes, TechCrunch, WSJ, Bloomberg, and authoritative vertical publications — over brand-owned content. Your Profound score is primarily a reflection of your earned media footprint in the publications AI systems weight as authoritative. That's the thing to fix.
The three-move playbook for closing an AI citation gap:
Move 1 — Audit your current earned media footprint.
Before you touch your website, run an honest audit of where your brand has been covered in the last 18 months. Not press releases. Not guest posts on your own blog. Third-party editorial coverage: news articles, profile pieces, product reviews, and analysis that name your company in authoritative publications.
If your coverage is thin — a handful of regional outlets, a few trade publications with low domain authority, or mostly self-published content — that is the gap. Profound's internal data shows that up to 90% of cited sources in AI answers change over time, and different models draw from distinct source sets. Volatility in your score often means you have shallow, inconsistent coverage rather than a deep, persistent earned media footprint.
Move 2 — Prioritize earned media in publications AI systems actually cite.
Not every publication is equal for AI citation purposes. The publications that appear most frequently in AI training data and RAG retrieval are the ones with the highest editorial standards, the longest digital history, and the broadest cross-platform authority: Forbes, TechCrunch, WSJ, Fast Company, Bloomberg, Inc., Wired, and their authoritative vertical equivalents.
One Forbes feature moves the citation needle more than 20 medium-tier trade press mentions. One TechCrunch piece with a named executive quote creates an entity signal that persists across AI training runs and real-time retrieval. Search Engine Land's comprehensive GEO guide confirms that the publications dominating AI citation share in every category are those with the strongest editorial authority and the longest history of AI training data inclusion. Focus your earned media program on the tier of publications where AI engines actually go when assembling answers, not just the publications where you can most easily get placed.
Move 3 — Combine earned placements with citation-optimized owned content.
Every earned media placement needs somewhere to point — an owned content asset that reinforces the entity signal the placement creates. That means a GEO-optimized blog post that defines your brand's core claims with specific data points, structured Key Takeaways that AI engines extract as citation snippets, and FAQ sections that answer the exact questions users are asking in your category.
This is Machine Relations in practice: earned authority (the placements) plus citation architecture (the owned content) plus entity optimization (the structured identity signals) working together. Any single layer underperforms without the others. The brands that move their Profound scores fastest are running all three in parallel.
What this looks like in practice:
A Series B startup selling into enterprise marketing teams opens its Profound dashboard and sees a citation frequency of 3% for its primary category keywords. Competitors are at 15–22%. The gap is real, the category is active, and the brands ranking in AI answers have 12–24 months of consistent Tier 1 earned media coverage that this company doesn't.
The fix: launch a systematic earned media program targeting 2–4 Tier 1 placements per month — company profiles, founder stories, product coverage, executive thought leadership — in publications with domain authority that AI engines weight heavily. Combine those placements with a refreshed blog architecture that uses citation-optimized structure on every key page. Track citation frequency monthly via Profound or equivalent.
AuthorityTech's data shows brands doing this at scale — 12+ citation-optimized pieces per month plus consistent earned media placements — achieve AI visibility gains 200x faster than those relying on occasional content. Conductor's 2026 research confirms the urgency: 32% of digital marketing leaders named GEO their top priority for the year, up from 12% in 2025. The brands at the front of that curve are not just optimizing content — they're building systematic earned authority programs.
If you're looking at a lower-than-expected AI visibility score, the first question to ask is not "how do I optimize my website?" The first question is "where has my brand appeared, in credible third-party editorial coverage, in publications that AI engines actually trust?" The gap in the answer is the gap in the score.
Start with an AI visibility audit to see specifically where your brand's citation authority stands and where the highest-leverage gaps are. The measurement layer is now here. Gartner projects traditional search to decline 25–50% by 2028 while AI search traffic grows at 9.7x year over year. The companies that move fastest on the action layer are the ones that will own their category's AI citation share before competitors figure out what the Profound dashboard is actually telling them.
Frequently Asked Questions
What does a low Profound AI visibility score actually mean?
A low Profound score means AI engines like ChatGPT, Gemini, and Perplexity are not citing your brand when answering questions in your category. The primary driver is usually a thin earned media footprint — insufficient third-party coverage in authoritative publications that AI systems weight as trusted sources. AuthorityTech's analysis shows 82–89% of AI citations come from third-party earned media, not brand-owned content. A low score is a signal to prioritize earned media, not just content optimization.
How quickly can you improve your AI visibility score with an earned media program?
AuthorityTech clients running systematic earned media programs — 2–4 Tier 1 placements per month combined with citation-optimized owned content — typically see measurable score improvements within 60–90 days of placements landing. The compounding effect accelerates: 34% of AI citations in any category go to a single dominant source, so brands that build early earned authority create moats that are increasingly difficult to close. The window to establish category citation dominance is narrowing.
Which publications matter most for improving AI citation frequency?
The publications that move AI citation scores fastest are those with the highest editorial standards and longest digital history: Forbes, TechCrunch, WSJ, Bloomberg, Fast Company, Wired, Inc., and authoritative vertical publications in your specific category. These are the publications that appear most frequently in AI training data and real-time retrieval systems. One placement in a Tier 1 publication with genuine editorial authority produces a more durable AI citation signal than dozens of placements in lower-authority trade press.
What's the difference between tracking AI visibility (Profound) and building it (Machine Relations)?
Profound and similar tools show you your current AI citation frequency — they're the measurement layer. Machine Relations is the full discipline for actually building AI citation authority: combining earned media (Tier 1 placements that AI engines cite as trusted sources) with citation architecture (GEO-optimized owned content) and entity optimization (structured identity signals). Measurement tells you the score. Machine Relations tells you how to improve it. The brands that close citation gaps fastest are running both in parallel — tracking scores while executing earned media programs that move them.