top of page
Won by refusing to be positioned as 'another data enrichment vendor' and instead building the horizontal orchestration layer that lets any GTM team stitch together 150+ data providers and AI agents into one workflow — becoming infrastructure other AI-native sales tools now get built on top of.
1
MODEL
BUSINESS MODEL
SaaS, API Platform
model bm
HOW THEY BUILT IT
- Originally built as a horizontal, general-purpose workflow/programming platform, not a sales tool — the founders spent years building a flexible data model and integration platform before ever positioning Clay as a go-to-market product.
- Became the 'orchestration layer' connecting 150+ external data enrichment providers (Clearbit, ZoomInfo-adjacent sources, LinkedIn, Google Maps, GitHub, and more) into a single no-code, spreadsheet-like interface, using 'waterfall enrichment' to query multiple providers in sequence until a match is found.
- Investor CapitalG cites average customer satisfaction of 9/10 and customers reporting they'd pay 50-100% more for what Clay currently delivers — an unusually strong retention/expansion signal for a workflow tool.
- Marquee customers include OpenAI, Anthropic, Canva, Figma, Intercom, Rippling, and Vanta, each citing specific, quantified impact (e.g., Anthropic 3x'd its enrichment rate; Intercom grew outbound pipeline 140%).
HOW TO ARCHITECT IT
1. Build the underlying technical infrastructure (a flexible data model, a broad integration layer) years before you decide which specific vertical use case to sell it as — Clay's founders describe this explicitly as the reason its GTM product became so much more capable than purpose-built competitors that started narrower.
2. Solve the 'no single data provider is ever complete' problem structurally via waterfall/sequential querying across many providers, rather than betting on one data source being good enough — this is the single most cited reason customers prefer Clay over any individual data vendor.
3. Let AI agents (Claygent) handle the unstructured research a human would otherwise do manually, positioned explicitly as extending what a company's best reps already do by hand, not replacing judgment entirely.
DISTRIBUTION MODEL
Self-Serve Website, Content Distribution, API Distribution
dm
HOW THEY OPERATIONALIZED
- Grew initial GTM/RevOps adoption through self-serve sign-up and extensive documentation/content marketing (a dedicated GTM engineering blog) targeting the specific technical persona (RevOps, growth engineers) who build and maintain enrichment workflows.
- Distribution compounds through case studies from high-profile AI-native companies (OpenAI, Anthropic, Canva) that function as credibility signals within the tight-knit GTM/RevOps community, where peer references carry outsized weight.
HOW TO REPLICATE WHAT WORKED
What worked: positioning as the orchestration/workflow layer sitting above data providers rather than as another data provider itself — this avoids direct price competition with any single enrichment vendor and instead makes Clay complementary to (and often a customer of) the very vendors it might appear to compete with.
Trap if copied blindly: Clay's own users and reviewers cite a steep learning curve and unpredictable credit-based pricing that can 'balloon fast' for teams with variable prospecting volume — a founder replicating this model should budget heavily for onboarding/education, since a horizontal, highly flexible tool is inherently harder to learn than a narrow point solution.
| PATTERNS OF THIS MODEL
PATTERNS IN ORCHESTRATION LAYERS ACROSS FRAGMENTED DATA VENDORS:
1. BUILD THE FLEXIBLE DATA MODEL AND INTEGRATION LAYER BEFORE CHOOSING THE USE CASE. General infrastructure applied to a specific problem outperforms a purpose-built tool that started narrow.
2. SOLVE THE STRUCTURAL TRUTH THAT NO SINGLE DATA PROVIDER IS EVER COMPLETE. Sequential querying across many sources is the reason customers prefer an orchestrator to any individual vendor.
3. USE AGENTS TO EXTEND WHAT PRACTITIONERS ALREADY DO MANUALLY, not to replace judgement. Framing determines whether operators adopt or resist.
4. WHEN CUSTOMERS SAY THEY WOULD PAY SUBSTANTIALLY MORE, THE PRICING MODEL IS THE CONSTRAINT, NOT THE PRODUCT. Underpricing a workflow that generates revenue is the most common error in this category.
What companies with this model reveal
| OPPORTUNITY INTELLIGENCE
GOLDMINE 1 — BUILD THE HORIZONTAL INFRASTRUCTURE BEFORE CHOOSING THE VERTICAL.
Standard: Clay spent years building a flexible data model and integration platform before positioning as a go-to-market product. The founders credit this directly for why the GTM product outperforms purpose-built competitors that started narrower.
GOLDMINE 2 — SOLVE "NO SINGLE DATA SOURCE IS COMPLETE" STRUCTURALLY.
Standard: waterfall enrichment across 150+ providers, querying in sequence until a match is found, is the most-cited reason customers prefer Clay over any individual vendor. Do not bet on one source being good enough.
GOLDMINE 3 — LET AGENTS DO THE UNSTRUCTURED RESEARCH A HUMAN WOULD.
Standard: Claygent extends what the best reps do manually rather than replacing judgement.
THE PIT — YOUR PRODUCT IS AN ORCHESTRATION LAYER OVER SUPPLIERS WHO CAN RESTRICT OR REPRICE ACCESS.
150+ data providers is 150+ dependencies, each with its own terms, and several are also potential competitors. Aggregation businesses are only as stable as their least cooperative supplier.
THE SECOND PIT — CUSTOMERS REPORTEDLY WOULD PAY 50–100% MORE, WHICH MEANS YOU ARE UNDERPRICED, NOT SAFE.
MOVE WITH CAUTION — OUTBOUND VOLUME TOOLS FACE TIGHTENING DELIVERABILITY AND PRIVACY ENFORCEMENT.
Untapped Business Model / Gaps / Goldmines / Pits
Patterns & Insights
2
MARKET
mkt mt es
MARKET TYPE
Blue Ocean
WHY THEY WON
Data enrichment for GTM teams existed as a category (Clearbit, ZoomInfo) but the 'orchestration layer that combines many providers plus AI research agents into customizable no-code workflows' was not a defined category before Clay. Clay effectively created a new layer above the existing enrichment vendors rather than competing directly with any one of them. Transferable principle: when an established category has many good-but-incomplete point solutions, the genuinely new opportunity may be the orchestration layer that combines them, not a better version of any single point solution.
ENTRY STRATEGY
Greenfield Entry
EXECUTION
Clay entered via self-serve product-led adoption among technical GTM/RevOps practitioners rather than enterprise sales or channel partnerships, the natural entry mode for a horizontal workflow tool whose value is best demonstrated through hands-on use rather than a sales pitch.
FOOTHOLD STRATEGY
fs
Beachhead Strategy
The beachhead was RevOps and growth engineers at high-growth, technically sophisticated companies (many themselves AI-native startups) who were already frustrated by fragmented enrichment tooling and had the technical skill to build custom workflows in a flexible, spreadsheet-like interface. From that foothold, Clay expanded to broader sales and marketing teams within the same companies as those technical builders created reusable templates non-technical colleagues could run themselves.
GROWTH CAMPAIGN
CAMPAIGNS THAT WORKED
AI Hackathon weeks with partner companies (Google, and others): jointly building real GTM workflows in public, generating both product feedback and case-study content simultaneously.
Customer-published ROI metrics (Anthropic's 3x enrichment rate, Intercom's 140% pipeline growth, Figma's improved PLG conversion): used extensively across Clay's own marketing to make an abstract 'orchestration layer' concept concrete via specific, named-customer results.
Claygent AI agent launch: extended the product from data enrichment into automated unstructured web research, broadening the addressable use cases within existing customer accounts.
KEY LEARNING
If your category has many good-but-incomplete point solutions (data providers, in Clay's case), consider whether the real opportunity is the orchestration layer that combines them via sequential/waterfall logic rather than trying to become the single best point solution yourself — and invest in AI agents that extend what your best manual users already do, rather than positioning AI as a wholesale replacement for their judgment.
gc
Market Context
| MARKET INTELLIGENCE
THE STANDARD: When a category has many good-but-incomplete point solutions, the opportunity is the orchestration layer that combines them, not a better single solution.
RULE 1 — ORCHESTRATION MEANS CONSUMING YOUR SUPPLIERS, NOT COMPETING WITH THEM. Routing across many providers is a faster entry than out-collecting any one.
RULE 2 — WATERFALL LOGIC IS THE PRODUCT. Trying providers in sequence until a field resolves is a coverage and cost advantage no single vendor offers.
RULE 3 — DEFENSIBILITY IS THE WORKFLOWS CUSTOMERS BUILD. Data sources are replaceable; a team's operational logic is not.
RULE 4 — SUPPLIERS CAN RESTRICT ACCESS TO THE LAYER ABOVE THEM. Your cost base sits inside other companies' commercial decisions.
MARKET TYPE: Blue Ocean (GTM data orchestration).
| MARKET ENTRY PLAYBOOK
THE STANDARD: TECHNICAL PRACTITIONERS ADOPT TOOLS BY USING THEM, NOT BY BEING SOLD THEM — the entry is a product they can prove in an afternoon.
RULE 1 — SELL TO THE OPERATOR WHO BUILDS, NOT THE EXECUTIVE WHO BUYS.
Revenue operations staff can pilot without procurement and become internal advocates.
RULE 2 — AGGREGATING MANY DATA SOURCES IS THE WEDGE COMPETITORS WITH ONE DATASET CANNOT COPY.
Orchestration across providers is a neutral position no single data vendor will take.
RULE 3 — POWER-USER COMMUNITIES ARE THE CHANNEL AND THE ROADMAP.
Published workflows and templates market the product and reveal what to build next.
How to enter
| FOOTHOLD STRATEGY PLAYBOOK
THE STANDARD: Give technically capable users a flexible surface and they will build the use cases your roadmap never would.
RULE 1 — TARGET THE OPERATOR WHO CAN BUILD, NOT JUST BUY. Growth engineers and RevOps staff at technical companies will assemble workflows that a packaged product cannot anticipate.
RULE 2 — AGGREGATING FRAGMENTED DATA SOURCES IS THE VALUE. Removing the need to evaluate and stitch many enrichment vendors is the wedge, not any single source.
RULE 3 — THE TEMPLATES YOUR POWER USERS BUILD ARE HOW NON-TECHNICAL COLLEAGUES ADOPT. Expansion inside an account happens through their work, not your onboarding.
RULE 4 — A SPREADSHEET-LIKE SURFACE IS BOTH THE ADOPTION MECHANISM AND THE COMPLEXITY CEILING. Beyond a point, users need software, not a canvas.
How to get the first strong position
MARKET PATTERNS & PLAYBOOK
3
MONEY
money rev pri
REVENUE MODEL
Subscription, Usage-Based
PRICING MODEL
Usage-Based Pricing
WHY THEY WON
Tiered subscription (e.g., an Explorer plan around $349/month) combined with a credit-based consumption model for enrichment queries and AI agent runs, meaning cost scales with actual usage volume/complexity of workflows rather than a flat seat price — reflecting the platform's role as an infrastructure layer processing variable-volume data queries.
Credit-based consumption pricing ties cost directly to enrichment volume and workflow complexity, targeting RevOps/growth engineering buyers who understand and can forecast their own query volume, though reviewers note this creates less predictable costs for teams with variable prospecting volume compared to flat per-seat SaaS pricing.
TARGET AUDIENCE
CUSTOMER BUYING BEHAVIOUR
tg cb
RevOps and growth engineers at high-growth tech companies (buying flexible workflow-building infrastructure); sales development and outbound teams (buying enriched, prioritized prospect lists); GTM leadership at AI-native companies (buying a unified data orchestration layer to replace multiple disconnected tools).
Self-serve and technical trial-first for individual practitioners building workflows, expanding to committee-level adoption (RevOps + sales leadership) once a workflow proves valuable and needs to scale company-wide — adoption is bottoms-up and usage-driven rather than top-down procurement.
| PRICING INTELLIGENCE
What makes this model effective & make customers pay
Charge for enriched records, not for seats, when the underlying data has real marginal cost and the customer's usage is bursty.
RULE 1 — CREDIT-BASED PRICING MATCHES A COST BASE MADE OF THIRD-PARTY DATA CALLS.
You are reselling many providers' data. A credit meter passes that cost through honestly and scales with genuine use.
RULE 2 — AGGREGATING MANY DATA VENDORS INTO ONE SUBSCRIPTION IS THE VALUE PROPOSITION.
The buyer replaces several contracts and the integration work between them. Anchor to the stack, not to a competitor.
RULE 3 — WATERFALL ENRICHMENT MAKES SPEND EFFICIENT, WHICH BUILDS TRUST AND REDUCES REVENUE PER LOOKUP.
Deliberately helping customers spend less is a defensible long-term position and a short-term cost.
RULE 4 — CREDIT METERS MUST ALERT BEFORE THEY SURPRISE.
Bursty campaign usage produces unexpected invoices. One shocking bill undoes months of goodwill.
A go-to-market team is buying research capacity they would otherwise hire offshore. Price against the headcount not added, and credits are compared to salaries rather than to data vendors.
PRICE & REVENUE
| Revenue Risk - The biggest threat to revenue stability
Credit-based consumption for enrichment and AI runs is the right unit for variable workloads and produces unpredictable bills that trigger evaluation at renewal.
Sitting on top of third-party data providers means your gross margin and your legality are both set by suppliers.
Enrichment data is subject to GDPR/CCPA exposure and platform anti-scraping enforcement — the asset can degrade without any commercial event.
Rapid growth in an AI-adjacent category attracts well-funded fast followers and, eventually, native capability from the CRM.
Reported strong growth and a high valuation in 2025; no audited figures published, and third-party estimates vary.
Where the model can break
4
MOTION
GROWTH EXPANSION MODEL
COMPETITIVE STRATEGY
motion ge cs
Platform Expansion
HOW THEY EXPAND
Clay expanded from a data-enrichment workflow tool into a broader GTM platform spanning account/contact research, AI research agents (Claygent), campaign orchestration, and large-scale audience management (removing the original 50,000-record workflow limits) — sequenced to progressively cover more of the GTM stack rather than deepen enrichment alone.
Differentiation
HOW THEY COMPETE
Clay differentiates against single-source data vendors like ZoomInfo by positioning itself as workflow orchestration rather than a data intelligence platform — ZoomInfo makes its own data available everywhere, while Clay pulls from many providers including ZoomInfo-adjacent sources, a sequencing that lets Clay remain complementary to, rather than a direct substitute for, most individual data vendors.
GROWTH ENGINE
GTM
ge n gtm
Content Flywheel, Community-Led Growth
The loop: technical practitioners build and share GTM workflow templates publicly (via the GTM engineering blog and community), which both attracts new technical users seeking similar workflows and demonstrates concrete ROI to less technical prospects evaluating the tool. It would break down if the RevOps/growth-engineering community's enthusiasm cooled, since the content and case-study flywheel depends heavily on customers self-selecting to publicly share their specific results.
Product-led, self-serve adoption among technical RevOps/growth practitioners, reinforced by heavy content marketing (a dedicated GTM engineering blog, published customer ROI case studies) and joint hackathon events with partner companies that generate both product improvements and case-study material simultaneously.
SUSTAINING MOATS
Switching Costs, High Customer Lock-In, Brand Power, Technology Advantage (complex enterprise scenarios)
moat
Clay's moat is the breadth of its waterfall-enrichment integration layer (150+ data providers) combined with the accumulated workflow templates and credit-usage optimization each customer builds over time — migrating away means rebuilding potentially dozens of interdependent enrichment and AI-agent workflows from scratch, a switching cost that grows the longer and more deeply a RevOps team has built on the platform.
| MOAT INTELLIGENCE
THE STANDARD: Orchestrating many data sources is a stronger position than owning one, because your product improves every time any provider does.
RULE 1 — WATERFALL ENRICHMENT MAKES YOU BETTER THAN EVERY SINGLE SOURCE. Querying providers in sequence until a match is found produces coverage no individual vendor can offer, and the customer stops caring which one answered.
RULE 2 — THE PROVIDERS BECOME COMMODITIES AND YOU BECOME THE INTERFACE. Aggregation inverts the power relationship — data vendors compete for inclusion while you own the customer.
RULE 3 — THE COMMUNITY OF PRACTITIONERS IS THE DISTRIBUTION ENGINE. In technical go-to-market tooling, published workflows and templates are what convert curiosity into adoption faster than any sales motion.
THE SIGNAL: the risk is that agents call data providers directly, removing the orchestration layer. The defence is the accumulated workflow logic customers have built — which is configuration debt they will not rebuild.
Why this company remains defensible
ARR & TAKEAWAY
ARR Journey - what to do at each stage
PRE-$1M ARR — SELL THE ABILITY TO COMBINE DATA SOURCES, NOT A DATABASE
Every sales team already buys data. The unserved problem is orchestrating dozens of enrichment sources and AI research into one workflow.
Waterfall enrichment — trying each provider until one answers — is a mechanic competitors sell separately and you assemble.
$1–5M ARR — BUILD A COMMUNITY OF EXPERT USERS
The product is powerful and hard. A certified expert community that teaches, implements and evangelises is what converts complexity into adoption.
WATCH: credits consumed per account per month — the honest usage and pricing unit.
$5–10M ARR — PRICE ON CREDITS, NOT SEATS
Consumption pricing aligns with your own data costs and lets accounts expand without a negotiation. In AI-era GTM tooling this is the correct default.
$10–50M ARR — AGENCIES AND EXPERTS ARE THE IMPLEMENTATION LAYER
A services ecosystem around your product raises adoption without adding your headcount.
$50–100M ARR — CAPITAL FOLLOWS CONSUMPTION GROWTH FAST
Clay has raised repeatedly at rapidly escalating valuations, reported into the billions, with ARR figures cited in the low hundreds of millions. These are third-party and company-stated figures, not audited.
Expect competitors and data providers to attack the aggregation layer directly.
$100M+ ARR — THE RISK IS THAT AGENTS ABSORB ORCHESTRATION
If general agents can research and enrich autonomously, the orchestration canvas commoditises. The defensible remainder is the data relationships, the workflow and the community.
Rule: aggregating other people's data is a fast business to build and a hard one to hold. Convert the head start into distribution and a community before the layer commoditises.
COPY PLAYBOOK : What Worked → What Failed → What to Replicate → What to Avoid
THE STANDARD: Positioning as the orchestration layer above your suppliers rather than as another supplier avoids price competition entirely and makes you a customer of the companies you appear to compete with.
SEQUENCE:
1. Sit above the data or service providers rather than becoming one.
2. Make the workflow, not the data, the product.
3. Budget heavily for education, because flexible horizontal tools are hard to learn.
WORKED: Orchestration positioning making Clay complementary to — and often a buyer from — vendors it might otherwise have fought on price.
CAUTION:
1. FLEXIBILITY MEANS A STEEP LEARNING CURVE AND UNPREDICTABLE CREDIT-BASED BILLING that users report ballooning fast with variable volume. Onboarding investment isn't optional; it's the cost of the position.
2. AN ORCHESTRATION LAYER'S VALUE FALLS IF ANY SINGLE SUPPLIER BECOMES DOMINANT enough to own the workflow itself.
bottom of page