Profound's Data Marketing Play Teardown: The Data Behind AI Search

Most brands publish content-shaped oatmeal. Profound turned AI search data into rocket fuel.

I obsess over how marketing teams turn weird, defensible data into unfair marketing advantages. Profound does this in spades.

The core play: they turned AI search visibility into a refreshable scoreboard, a quality-over-quantity coaching layer that educates, and a megaphone that fills rooms. It's one of the strongest data marketing engines running right now in the AI space , and frankly, it’s the kind of thing that happens when you hire data storytellers in your first handful of hires.

Here’s how Profound pulls this off, across three pillars:

• Data: The raw numbers that power the engine.
• Insights: Analysis + narrative that turn numbers into actionable takeaways.
• Distribution: The smart moves that get the story in front of the right audience.

What Profound Is (In One Breath)

Profound (tryprofound.com) shows brands how visible they are inside AI answers, then helps them create the right content to improve that visibility. Founded by James Cadwallader. Backed by Sequoia Capital, Kleiner Perkins, Khosla Ventures, Saga VC, and South Park Commons. Their client roster includes logos like Figma, MongoDB, and Ramp.

This research is based on me poking around, some Ahrefs usage (weird, right?), LinkedIn and a demo call.

My Framework to Judge a Data Play

When I evaluate data marketing, I look for three things: Data. Insight. Distribution.

Or, if you like metaphors: Scoreboard. Coach. Megaphone.

Let me show you how Profound nails all three.

Data: The Layer That Trains the Market to Come Back

The Profound Index

Here’s where things get interesting. The Profound Index is:

• Public. No gates, no forms, no “request a demo” nonsense
• Weekly. Fresh data that keeps people coming back
• Comprehensive. Currently covers US ChatGPT across 12 industries, with more promised

The dashboard ships with everything you’d want:

• A brand visibility leaderboard per industry
• Week-over-week change highlights (because movement = story)
• Top-three company charts in each segment so your eyes lock on what’s shifting
• A browsable keywords module and Prompt Volumes that surface the most-prompted terms

That last bit? Quiet masterstroke . The open clicks and prompts tease the paid product without giving away the crown jewels. You see what buyers actually ask, not what we hope they ask. It’s editorial that sells by implication.

Scale and Access

The index rides on uniquely hard-to-get clickstream and conversation data. Here’s how I rate the general data/market match + their free marketing data offering:

Data Scorecard:

• Frequency: High (weekly)
• Access: High (live, ungated, self-serve)
• Uniqueness: High (clickstream and conversation corpus are rare)
• Commercial orientation: High (public value equals product value)
• Prospect brands: High (names targets without using their private data)
• Coverage today: US ChatGPT, 12 industries, more coming
• Methodology: Published on page (transparency builds trust). I like how they credit the folks behind it too. Classy move.
• Programmatic SEO: Low (for now) without any long-tail pages across segments, types of companies or models.
• Prospect targeting: High because the index names brands without using those brands' proprietary data. Translation: You can attract exactly the companies you want to sell to without any risky data-sharing arrangements.

Honest Traffic Note

It’s not really working yet.

The index is still young on backlinks and organic pull. Ahrefs shows barely any backlinks yet, with research posts (keep reading) doing more link work than the index itself. But here’s the thing— the data itself is the moat; distribution can be tuned .

What’s remarkable: Profound launched their site in June 2024 but already hit a domain authority of 71 according to Ahrefs. That’s the power of quality research driving real backlinks.

Humans behind the numbers: Nick Lafferty (Growth Marketing), Siddharth Chandrappa (Data Scientist), Ali Vaghar (Head of Data), Ralfi Berk (Product Manager), Josh Prunty (R&D). As I said, the methodology is published with full transparency. Naming builders compounds credibility.

Insight: The Research Layer That Is Killing It

Research That Travels

Profound’s Resource Hub leans on analyst-written studies, not just product cheerleading. They don’t create a ton of content—there are 19 posts filed under research since June 2025, and 10 of them link to LinkedIn articles, which I love. Two of them are e-books, which is a nice retro throwback.

That said, they are getting a ton of traction with their research. For example, three of the research posts alone rank in Profound’s top five backlinked articles:

AI Platform Citation Patterns
Which sources do ChatGPT, Google AI Overviews, and Perplexity cite, and how do those patterns differ. This is the evidence base for GEO (Generative Engine Optimization) that proves it’s not just SEO with a new hat.

The Surprising Gap Between ChatGPT and Google
Quantifies the canyon between AI answers and classic SERPs. If your plan is only SEO, you’re already behind.

AI Search Intent Study
What 50M+ prompts reveal about how people actually use these systems, including how often they want a direct answer, not a click-out.

This feeds distribution

These reports earn real attention. Media coverage touched places like Bloomberg and Axios. A recurring subplot: the rise, fall, and rise of Reddit citations inside AI answers. Reddit even surfaced the topic on an earnings call.

Pro tip for anyone running PR on data: if your dataset can quantify a public company's role in a new behavior, editors will care . As Reddit wins from AI, so too does Profound.

Editorial Packaging That Works

Posts are concise and actionable. The sidebar pattern is excellent: What changed, How visibility shifted, Why it matters. Bylines are prominent, and the site often links out to the analyst’s LinkedIn post. That builds both the person and the brand.

Cadence and mix: Roughly weekly posting overall since launch. About 73% of posts are company or product news. Research and data pieces account for about 10% of total posts, with around one data-driven article per month. Less can be more when it's excellent .

All About The Bylines: A lot of the LinkedIn content and research is written by Josh Blyskal, who worked at HubSpot in AI/ML before. He also speaks at their events and has pulled in ~14k followers. He was Profound’s second employee, which really sheds light on this go-to-market play. In general, Profound does a nice job of crediting a diverse range of researchers and writers in their posts, which I really like.

The faces are part of the product. This reminds me of the expertise amplification approach I wrote about—putting real humans behind the insights.

Here’s what’s brilliant: Profound links to analyst's LinkedIn posts in their resource feed . Great symbiotic relationship between digital and social that amplifies both channels.

Distribution: The Loop That Turns Interest Into Pipeline

LinkedIn First

I found Profound through LinkedIn, then through their events. The feed strategy is person-led: researchers and the founder post first, the brand account amplifies. That maps to how LinkedIn currently rewards expert content from people. The result is deeper comments and better dwell time.

In Person: ZeroClick Events

Invite-only events in key hubs that fill fast and stay practical because the research authors get on stage. The last two:

• New York, October 8, 2024, at Ramp HQ
• London, November 19, 2024, at Outernet

Speakers included Reddit, Pullrank, WPP, and Ramp. I applied to the New York event and got rejected. Still fanboying. That's what demand looks like .

Email (Surprisingly Quiet)

Here’s an easy win they’re missing: I signed up for events and had contact, but didn’t receive research updates by email. Start a monthly Research Roundup with a single delta, one chart, and one action. Let people opt into category-specific alerts.

Sales Enablement

This is where you feel the orchestration. I sat through a pitch from Ron, a rep who could have taught the research. Slides were packed with data from the public studies, and he followed up with fresh slices. That's the loop. Public index and research seed the narrative, IRL talks echo it, sales closes with it .

Media

From what I can see, media pickup is good but PR pressure hasn’t been slammed to max. A steady weekly or biweekly pitch around the leaderboard movers would add backlinks without diluting voice. For now, it looks like mostly product coverage together with some happy-go-lucky pickups when there is serious news they can hijack.

The Verdict

What I Love

Quality over quantity. When you find the right data, you’re golden. The sheer amount of data shared for free is generous and smart. The commercial alignment is tight—nearly everyone who follows the data is either a buyer or a channel.

Real repurposing is happening across ZeroClick events, sales, and LinkedIn. Josh, the second hire, is a researcher who knows how to package insights. That wasn’t an accident. This playbook was chosen on purpose .

The results speak for themselves: Fortune 100 prospects fighting to join their owned events, massive social visibility, and media pickups in Bloomberg and Axios . All from turning unique data into a content engine.

This connects to what I wrote about in creating standout content—when you have genuinely valuable data, the content writes itself.

What I Would Do Differently

• Add categories faster, as coverage permits
• Push research updates by email, ideally segmented by industry.
• Test short webinars tied to major releases
• Turn up PR for the index movers. Spin up a monthly email with brief analyses and changing data trends to industry reporters.
• Explore a light programmatic layer for long-tail demand (but I also respect the current restraint—it keeps the signal clean)

Why The Engine Works

Scoreboard builds habit. Coach rewires mental models. Megaphone turns attention into meetings .

Most importantly, the public data and the product outcome are the same thing. No messaging gap. That’s the unlock.

This reminds me of the trust-building approach I analyzed with the New York Times—when your public value directly demonstrates your private value, everything clicks.

How to Steal This Play (Without Stealing the Soul)

Want to build your own data marketing engine? Here’s the playbook:

1. Pick one flagship metric that buyers care about. For Profound, it’s rankings.
2. Publish it on a reliable cadence (weekly is gold standard - things rarely change more frequently)
3. Put the methods on the page (transparency builds trust)
4. Lead with the delta and a time window
5. For every chart, write a single action
6. Create a Resource Hub with one killer study each month or quarter
7. Let a person post first (brand amplifies after)
8. Pair it with a local event where the researcher presents insights in 18 minutes
9. Train sales to lead with the same three stats

Done right, your scoreboard becomes someone else's weekly habit .

Connect with me on LinkedIn for more stuff like this.