Why ZoomInfo Is Taking Apollo to the Mattresses

There’s always a moment, in every maturing tech category, where the story stops being about features and starts being about power.

That’s what this ZoomInfo–Apollo lawsuit is really about.

On its face, it’s a patent case. ZoomInfo says Apollo is infringing two of its patents on how modern GTM platforms crawl the web, turn unstructured text into structured “feature matrices,” and generate engagement recommendations based on similarities to existing customers. Apollo says: come on — this is just collecting, analyzing, and displaying information on a computer. You can’t patent that.

On December 15, a federal judge in Delaware said, in effect:
“You might be right in the end, Apollo. But ZoomInfo’s claims are plausible enough that we’re going to trial.”

Apollo’s motion to dismiss was denied. The patents live to fight another day. And the rest of the GTM intelligence industry just got dragged a little closer to the line of fire.

This isn’t ZoomInfo “winning.” It’s ZoomInfo climbing into the ring and making sure everyone sees who’s holding the gloves.

The Godfather problem in GTM

The title of this piece isn’t an accident. In The Godfather, “going to the mattresses” is what you do when diplomacy ends and the street war begins. You burn political capital. You lock into a long fight. You accept there’s no going back to how things were.

That’s roughly where we are in B2B data and GTM intelligence.

For a decade, the model was simple:

• Scrape the web,
• Normalize the data,
• Sell access to a giant database as if it were a fact of nature.

ZoomInfo was the canonical version of that model. Apollo, the faster, cheaper, more product-led remix. Dozens of smaller players layered AI copywriting, sequencing, “intent,” and “signals” on top. Everyone talked about coverage, contacts, company count. The whole thing felt like a race to see who could hoard the most records and ship the most Chrome extensions.

Then a few things happened at once:

• Privacy law got teeth.
• AI made “we enrich your CRM” feel like table stakes.
• And platforms like Apollo started to look less like scrappy upstarts and more like existential threats.

When incumbents feel their moat narrowing, they do three things:

1. Raise prices.
2. Raise switching costs.
3. Reach for the law.

ZoomInfo is now on step three.

What the lawsuit actually says

ZoomInfo is asserting two patents:

• US 10,380,609 – web crawling for leads generation and engagement recommendations
• US 11,392,964 – predictive analytics for leads generation and engagement recommendations

If you strip out the legalese, the patents describe a workflow that will feel very familiar to anyone building or buying GTM tools:

1. Compare target accounts to your existing customers.
   • Measure “fitness, engagement, and intent” characteristics for each.
   • Find similarities.
2. Crawl the web in a “guided” way.
   • Use a trained classifier to categorize web pages, hyperlinks, and link structures using content and code.
   • Use that categorization to decide where to crawl next — theoretically skipping irrelevant pages and saving time and storage.
3. Build feature matrices and recommend who to talk to.
   • Turn what you find into a feature matrix for each target.
   • Compare those matrices to the ones for existing customers.
   • Generate engagement recommendations based on the similarities.

ZoomInfo’s story is: this is not just generic database work. This is a specific technical improvement in how web crawlers and ML models behave — less noise, less compute, better targeting. And, they say, Apollo’s platform (Apollo, Apollo AI, Sales Automation, Labs) uses substantially the same pattern.

Apollo’s story is simpler:

This is just “collecting, analyzing, and displaying information on a computer.” That’s an abstract idea. Under Alice and Mayo, you can’t patent abstractions plus a generic computer.

The judge’s answer is subtle, but important:

• Step 1: Yes, this is an abstract idea – collecting, analyzing, and displaying data. On that front, Apollo is right.
• Step 2: But ZoomInfo has plausibly alleged that it’s implemented in a way that improves the underlying computer process (how crawling is guided, how models are regularized, how storage is reduced). That’s enough, at this early stage, to count as a potential “inventive concept.”

At the motion-to-dismiss phase, plausibility is the bar. The court doesn’t decide whether ZoomInfo is actually right. It decides whether ZoomInfo’s argument deserves evidence and cross-examination.

The answer was yes. Which means:

• Discovery is coming.
• Claim construction is coming.
• And this isn’t going away quietly.

Why now? Follow the incentives.

If you look at the timing, ZoomInfo’s decision to go to the mattresses isn’t mysterious.

• Its own risk profile has been rising — securities suits, right-of-publicity and privacy settlements, pressure from investors to defend margins and demonstrate “defensible IP.”
• Apollo has become the default alternative in a lot of GTM teams, with a more aggressive product cadence and a friendlier price point.

In that light, the lawsuit serves at least three functions:

1. Buying time.
   Even the act of litigating slows a rival down. Leadership gets pulled into strategy sessions. Product roadmap meetings include outside counsel. Potential partners ask, “So… what happens if you lose?” That’s time ZoomInfo can use to reposition itself as less of a static data broker and more of an “AI GTM infrastructure” player.
2. Sending a signal.
   ZoomInfo isn’t just claiming ownership of a dataset. It’s claiming ownership of a method:

“We own the way modern ABM platforms match lookalikes, crawl the web, and generate engagement recommendations.”

1. Whether that claim ultimately holds is a legal question. But the signaling effect is immediate. Every smaller vendor using language like “smart crawler,” “ML-guided scoring,” or “AI-recommended next best account” is now reading two patents a lot more closely.
2. Raising the barrier to entry.
   If you can’t win on product alone, you can win by making the cost of entry intolerably high for everyone else. That’s what strong, litigated patents do — not just to your current rival, but to the next wave of founders.

This is classic incumbent behavior in a sector that’s grown faster than its guardrails.

The second front: privacy and legitimacy

The interesting twist is that this patent fight is happening at the same time that both ZoomInfo and Apollo are being challenged on another axis entirely: whether their underlying data collection model is legitimate at all.

• ZoomInfo recently agreed to a multi-state settlement north of $29 million over using individuals’ names and profiles in marketing without consent.
• Apollo has faced class actions under Illinois’ Right of Publicity Act for allegedly repackaging residents’ identities into a commercial subscription product.

So on one axis, vendors are arguing:

“This scraping and scoring pipeline is ours. You can’t copy it.”

On another axis, regulators and plaintiffs are asking:

“Does anyone have the right to be running this pipeline at all, in this form, with this level of consent?”

That’s the real tension: incumbents are trying to lock down a model the law is simultaneously tightening up.

It’s an odd moment. The same workflow that’s being defended as “novel and inventive” in a patent case is being questioned as “invasive and exploitative” in privacy cases.

The courts won’t resolve that tension neatly. They’ll push on one side here, another side there. But for GTM leaders, the pattern is clear:

• The age of “we scraped it, therefore we sell it” is ending.
• The age of “who consented, who benefits, and who owns the method?” is beginning.

What this means if you’re buying GTM data, not building it

If you’re a CMO, VP of Demand Gen, RevOps, or Sales leader, you don’t need to become a patent lawyer. But you do have to internalize one thing:

Technical risk is no longer just “does the API work?”
It’s “could this vendor’s core method be enjoined, taxed by license, or fundamentally altered in court?”

Practically, that means a few shifts.

1. You now evaluate legal durability alongside feature sets

• Is there active IP litigation around how this vendor crawls, scores, or recommends?
• Is the vendor’s model purely scrape-based, or can it still deliver value if certain types of collection are restricted or reinterpreted?
• How dependent are they on one “secret sauce” method that someone else now claims to own?

If the answer is “very,” you’re buying a bigger risk than the UI suggests.

2. “We scrape bigger and faster” is a brittle value prop

Between:

• Patent fences (like the ones ZoomInfo is trying to enforce), and
• Privacy and publicity lawsuits around scraped profiles,

the “we just have more emails” pitch is starting to look like a liability, not an advantage.

What looks more robust?

• Deep, first-party signal integration (product usage, in-app behavior, tickets, contract events).
• Custom, customer-specific modeling rather than one-size-fits-all scoring.
• Consent-aware data flows you could explain to your legal team without breaking into a sweat.

That’s the pivot point: from static, prebuilt data lakes to bespoke, insight-driven systems built around your motion.

3. You architect for supplier fragility

If this case does nothing else, it should convince GTM and RevOps teams to:

• Avoid building your routing, scoring, and segmentation logic so tightly around a single vendor’s proprietary schema that you can’t switch if you need to.
• Introduce intermediate layers — CDPs, warehouses, or internal models — that allow you to swap sources without rewriting the logic of your go-to-market.

In other words: expect the ground to move, and design your stack so it doesn’t break every time it does.

The broader question: what are we actually trying to optimize?

Ezra Klein likes to ask a simple question in complex systems: what is this structure optimizing for?

ZoomInfo vs. Apollo is not just a fight over who owns a particular set of method claims. It’s a proxy war over what the GTM data ecosystem optimizes for in its next phase.

The first phase optimized for volume:

• Count of contacts
• Count of accounts
• Count of “signals,” however noisy

The second phase — the one we’re now stumbling into — will be decided by three different optimizations:

1. Precision
   Who can surface the right accounts at the right time, with real, explainable signals — not just “this IP hit your blog twice”?
2. Legitimacy
   Whose data collection practices can survive not just the letter of GDPR/CCPA/LGPD, but the spirit? Who can look regulators, customers, and end-users in the eye and say: “You knew what you were signing up for. Here’s how you can opt out”?
3. Defensibility of method, not just data
   Who builds systems so materially better — technically and ethically — that they’re worth protecting, not just with marketing copy but with IP you’re proud to enforce?

Right now, ZoomInfo’s move looks like an incumbent trying to use the law to freeze the board. Apollo’s defense looks like the usual “this is all just software” argument challengers make when they’ve run close to the line.

But the more interesting story is what happens next:

• Do we see more vendors retreat into generic, lowest-common-denominator functionality to avoid infringement?
• Or do we see a split — with one camp doubling down on “scrape and score,” and another building genuinely new ways of modeling buyer readiness using higher-quality, more consent-driven signals?

If it’s the latter, this lawsuit may end up being, ironically, good for the ecosystem. Not because one side wins, but because everyone is forced to stop treating the GTM data pipeline as a natural resource and start treating it as a design space with real tradeoffs.

Why “going to the mattresses” might backfire

There’s one more irony worth naming.

When an incumbent sues a challenger, they’re not just talking to the court. They’re talking to:

• Developers choosing which APIs to build on.
• Revenue leaders choosing which stack to standardize on.
• Regulators deciding which models feel extractive versus enabling.

ZoomInfo is telling all of them:

“We consider this workflow — guided crawling + feature matrices + engagement recommendations — ours.”

That might work in court. It might even, in some form, be right.

But if the market hears:

“We’re going to war to defend the old scrape-and-score paradigm,”

and, at the same time, sees regulators tightening around that paradigm, it could push the most interesting builders and the most sophisticated buyers somewhere else entirely:

• Toward more custom, model-driven approaches built on their own data.
• Toward smaller, more specialized signal providers whose differentiation isn’t quantity but relevance.
• Toward architectures where data partners are interchangeable pipes into a proprietary intelligence layer, not the intelligence layer itself.

In other words, ZoomInfo might win the right to tax a shrinking slice of the value chain.

The ending we don’t see yet

We don’t know how this case ends.

• A jury could find the patents invalid.
• Apollo could design around the claims and move on.
• The parties could settle quietly, with a license, a press release about “partnership,” and a few redacted pages everyone pretends not to read too closely.

But the effect on how serious GTM teams think should be clearer already:

• Patent risk is no longer abstract. It’s part of your buying criteria.
• Scraping and scoring, by itself, is no longer a stable foundation.
• The future of B2B data will belong less to whoever owns the biggest static lake, and more to whoever can turn custom signals, consented data, and adaptable models into something GTM teams can actually trust.

ZoomInfo has gone to the mattresses to defend one version of that future. Apollo is fighting to keep building another.

The smart move, if you’re on the buying side, is not to pick a side in that war — it’s to build a GTM system that doesn’t care who wins.