A full teardown of two go-to-market machines aimed at the same market — a $5B private challenger (Clay) and a public incumbent (ZoomInfo) being repriced by AI in real time.

What’s actually AI-native, what’s theater, where the gap between the investor story and the GTM motion is widest, and the three moves I’d hand Clay before its own moat starts to leak.

In May 2025, ZoomInfo did something no billion-dollar software company does on a whim.

It changed its stock ticker.

ZI became GTM.

The press release called the three letters “the soul of ZoomInfo.”

Twelve months later, almost to the day, ZoomInfo cut about 600 people — roughly 20% of the company — shut its Israel office, and told investors the reason was AI.

Same twelve months, different company:

Clay crossed $100M in ARR, raised a $100M Series C from Alphabet’s CapitalG at a $3.1B valuation, and by January 2026 was doing an employee tender at a $5B valuation while signing a lease in Manhattan to add 498 jobs.

One company renamed itself after go-to-market.

The other rewrote what go-to-market is.

I want to pull both apart and show you why one motion compounds while the other decays — and, because the interesting version of this story isn’t “the startup is winning” (it is), where Clay is quietly more fragile than its valuation suggests.

The number I keep staring at is ZoomInfo’s revenue-growth arc:

+57% in FY21, +46% in FY22, +13% in FY23, then -2% in FY24.

That’s not a company that got disrupted overnight.

That’s a company watching its core asset get cheaper every single quarter while it told the market a story about AI.

Clay’s curve points the other way, and I’ll spend the rest of this teardown on why.

Quick aside on the funding, because the shape tells the story on its own:

Seed in 2017, then years of relative quiet, then Series A, B, C, and a $5B secondary all packed into roughly 18 months.

Slow, slow, slow, vertical.

And it isn’t the product doing that.

One widely-shared teardown of Clay’s moat called the product “literally just a spreadsheet.”

It’s the motion wrapped around the spreadsheet.

Six dimensions, loosely graded, plus a composite for how AI-native each company’s own go-to-market actually is — not how AI-native its marketing claims to be.

The grades are the hook.

The rest of this teardown is the evidence.

Clay’s defensibility isn’t a feature.

It’s a flywheel that took six years to spin up.

And it has four moving parts.

In 2023 Clay started posting roles for a “GTM Engineer,” a function that didn’t exist.

Today there are hundreds of GTM Engineer postings at Cursor, Webflow, Notion, and beyond.

Every one of those reqs is a person whose career now depends on Clay fluency.

Naming the job created the buyer.

Read Clay’s careers page and you find no SDRs, no BDRs, no AEs.

You find two flavors of GTM Engineer:

The company dogfoods its product as its own pipeline engine.

Clay runs:

This looks like brand fluff until you see the number underneath it:

Accounts without an internal Clay champion are 60–80% more likely to churn at renewal.

Clay maintains 8,000+ programmatic “dossier” pages and refreshes them continuously using its own Claygent AI system.

That’s content engineering.

And it costs almost nothing per page to keep alive.

That single reframe explains:

Clay sells capability.

And capable people like to show off.

Meanwhile, Sasha “Ice” Petrov posted back-to-back 40-point games, vaulting himself into the spotlight. With upsets shaking up the standings, analysts are split on who’s truly leading the pack—and with half the season left, it’s anyone’s game.

ZoomInfo’s motion is the textbook version of the thing Clay routed around.

Start with careers.

ZoomInfo is hiring:

This is a labor-led pipeline.

More bookings means more reps.

It’s exactly the cost structure AI is built to compress.

The company also sells through:

That is a motion that monetizes lock-in, not love.

The Trustpilot score reflects it.

Copilot didn’t ship until February 2024.

The billing model only shifted when the old model stopped fitting the market.

Recovery to 90% NRR is still structural decline.

The company isn’t growing one half.

It’s amputating the other.

Workflow systems compound.

Contact databases commoditize.

That’s the difference.

Which is exactly why the important analysis is where it’s exposed.

Clay created a highly-paid technical priesthood.

That’s a fantastic wedge.

It’s also a ceiling.

Users increasingly complain about:

The anxiety is becoming part of the product experience.

Clay owns no underlying data.

Its moat is orchestration, workflows, and community.

That’s powerful.

But it’s also vulnerable if foundation models absorb the orchestration layer directly.

Kill the learning tax.

Protect the highest-LTV segment.

Expose Clay workflows as callable agents and MCP endpoints.

Meet users where the AI work already happens.

The lesson here isn’t: Clay good. ZoomInfo bad.

Both companies are betting on the same future.

The difference is:

Clay built a motion that compounds.

ZoomInfo built a motion that needed more people every quarter to hold flat.

Then AI made people the most expensive way to grow.

Both companies are betting on the same future.

The difference is:

Clay built a motion that compounds.

ZoomInfo built a motion that needed more people every quarter to hold flat.

Then AI made people the most expensive way to grow.

If you’re choosing what to copy:

Copy the motion, not the product.

The spreadsheet is replicable.

The flywheel took six years.

And if you’re Clay, the thing to fear isn’t ZoomInfo.

It’s becoming ZoomInfo.

A great asset wrapped in a billing experience customers learn to resent.

The 30 days start now.