6 ICP Frameworks for B2B GTM
Last updated:Six named ICP frameworks for B2B GTM, with components, sequence, and applicability guidance. Compiled by The Starr Conspiracy.
6 Ideal Customer Profile Frameworks for B2B GTM
The B2B ICP Framework Catalog is a methodology inventory for B2B go-to-market teams that need to stop guessing about targeting and operate from a validated definition of who they sell to. Compiled by The Starr Conspiracy, this catalog covers six named ideal customer profile frameworks B2B teams use, organized by data maturity and GTM motion: Closed-Won Pattern Analysis, Fit-Intent-Capacity Scoring, Buying Committee Mapping, Signal Stacking, Jobs-to-be-Done ICP Segmentation, and Hypothesis-Driven ICP. Define it. Validate it. Activate it.
Why most ICP work fails
Most ICP work fails because you're treating ICP like slideware, not a system. The "ideal customer profile" gets built once a year from sales-leadership opinion, parked in a deck, and then everyone wonders why pipeline forecasts keep missing.
If your ICP is "mid-market SaaS," what exactly are you optimizing for?
Opinion-led targeting compounds three failures: confirmation bias from the loudest voice in the room, survivorship bias from cherry-picked logos, and sales-cycle distortion from the deals everyone remembers. The cost shows up as wasted SDR cycles, CAC spikes, and sales-marketing blame loops. This is how CMOs end up cutting programs that were working, because the targeting was wrong. This is how VPs of Sales miss the number two quarters running. This is how RevOps gets handed a forecast nobody believes.
If your ICP lives in a slide deck, it's already dead. Yes, this is the part where someone says "our ICP is everyone." No, it isn't.
When budgets tighten, targeting mistakes get punished immediately. ICP validation is a risk-control mechanism, not an annual ritual. A slide is a map screenshot. A framework is the GPS plus the rules for rerouting when the road changes. If it can't be run weekly, it's not a framework. It's a poster.
Most ICP content gives you one process. This catalog gives you six operating models and the rules to choose among them. Definition, validation, and activation are three distinct jobs, and the right framework depends on which job you're failing.
The six frameworks below address different versions of the problem, from a company with hundreds of closed-won deals to a Series A startup with twelve logos and no statistical signal. Used together, they form an operational targeting system with cadence, owners, refresh cycles, and routing rules. We use demand states instead of funnel stages throughout. What a buyer is doing matters more for targeting than where they sit in a pipeline report. For broader strategic context, see our B2B go-to-market strategy work.
The Starr Conspiracy has applied these frameworks across HR tech, fintech, and B2B SaaS engagements for 25 years. We've rebuilt ICPs for teams with 10 deals and teams with 10,000. The catalog below is the working inventory, not a theoretical taxonomy. We don't sell AI experiments. We build marketing systems that actually work.
The six frameworks
How to use this catalog: read the origin and components of each entry, check the "When to use" line, then jump to the decision layer to sequence them. If you only do one thing, start with Closed-Won Pattern Analysis when you have win data and Hypothesis-Driven ICP when you don't.
Closed-Won Pattern Analysis
Closed-Won Pattern Analysis is a retrospective segmentation method developed as the foundational ICP technique in B2B revenue operations and refined by The Starr Conspiracy across 25 years of GTM engagements.
- Pulls every closed-won deal from the last 12 to 24 months as the input set.
- Profiles firmographics, technographics (installed software and tools), deal velocity, ACV, and retention against that set.
- Identifies clusters where sales cycles compress and expansion revenue concentrates.
- Excludes one-off deals and outlier ACVs to prevent survivorship bias.
- Requires a minimum sample of about 30 closed-won deals to trust the math.
- Outputs a ranked list of fit attributes weighted by revenue contribution, not deal count.
Inputs: CRM closed-won, product analytics, enrichment data, finance ACV records.
Sequence: Run first. Everything else downstream depends on its output.
Not for: teams with fewer than 30 wins or recent category pivots that invalidate historical patterns.
When to use: Run this first when you have at least 12 months of closed-won data and a recognizable pattern of repeat sales motions.
Fit-Intent-Capacity Scoring
Fit-Intent-Capacity Scoring is a prospective account-scoring model built on the fit-plus-intent split popularized by intent-data providers like Demandbase and Cognism, and operationalized by The Starr Conspiracy as a three-axis system that adds purchasing capacity.
- Scores fit using attributes surfaced by Closed-Won Pattern Analysis.
- Scores intent using third-party signals, site behavior, and category research activity.
- Scores capacity using budget proxies: funding stage, headcount growth, tech-stack maturity.
- Weights the three axes by deal data, not gut feel.
- Produces tiered account lists (A/B/C) that route to different plays and channels.
- Refreshes monthly so scoring tracks demand states, not last quarter's reality.
Inputs: CRM, intent platform, enrichment, finance signals. In practice, most intent stacks separate fit and intent but rarely model capacity, which is why pipelines fill with accounts that can't buy.
Sequence: Run after Closed-Won Pattern Analysis, before channel play design.
Not for: pre-revenue startups without a validated fit profile.
When to use: Deploy when you have a stable closed-won profile and need to convert it into a prospective targeting system that sales and marketing can both operate from.
Buying Committee Mapping
Buying Committee Mapping is a stakeholder-modeling framework adapted from enterprise B2B sales methodology and extended by The Starr Conspiracy to bind ICP work to the humans who actually sign.
- Names every role typically involved in a deal: economic buyer, champion, technical evaluator, blockers, end users.
- Maps each role to its dominant demand state and primary objection.
- Documents the message and proof point that moves each role forward.
- Identifies where deals stall by role, not by stage. Deal-cycle tooling makes role-level stall data extractable from CRM (DealHub).
- Pairs roles to channels and content formats that actually reach them.
- Updates as category maturity shifts committee composition.
Inputs: CRM contact roles, win/loss interviews, deal-desk notes, support tickets.
Sequence: Layer after Fit-Intent-Capacity Scoring once you have tiered accounts to multi-thread (engage multiple stakeholders in parallel).
Not for: SMB transactional motions with single-decision-maker deals.
When to use: Add this layer when deal sizes warrant multi-threading and your win/loss data shows deals dying at specific roles rather than specific stages.
Signal Stacking
Signal Stacking is a dynamic ICP framework developed by The Starr Conspiracy for B2B categories where static firmographics no longer predict revenue and demand states shift faster than annual planning cycles.
- Layers firmographic, technographic, behavioral, and event-based signals into a single composite score.
- Weights signals by observed correlation with closed-won outcomes, not vendor claims.
- Treats signal decay explicitly. A signal six months old is not the same signal.
- Triggers plays automatically when signal stacks cross defined thresholds. Example: a security vendor weights "new CISO hire" plus "pricing page visits" higher than "category keyword research."
- Distinguishes "research" signals from "ready" signals to avoid premature outreach.
- Retires underperforming signals on a defined cadence.
Inputs: intent platform, enrichment, web analytics, news and trigger feeds, CRM activity.
Sequence: Run after Fit-Intent-Capacity Scoring, when single-axis scoring stops producing differentiated tiers.
Not for: early-stage categories without enough behavioral data to detect patterns.
When to use: Reach for Signal Stacking when your category is crowded, buying windows are short, and fit-only scoring is producing too many false positives.
Jobs-to-be-Done ICP Segmentation
Jobs-to-be-Done ICP Segmentation is a problem-based segmentation method derived from Clayton Christensen's Jobs-to-be-Done theory and applied by The Starr Conspiracy to B2B categories where buyers cluster by the job they're hiring software to do, not by industry code.
- Defines segments by the functional job the buyer is trying to complete.
- Captures the emotional and social dimensions of the job, not just the functional one.
- Maps competing solutions, including spreadsheets, status quo, and in-house builds.
- Surfaces the trigger events that move a buyer from passive to active demand states. Voice-of-customer platforms make these triggers measurable at scale (Qualtrics).
- Defines the outcome the buyer measures success against.
- Connects messaging directly to the job, not to product features.
Inputs: customer interviews, voice-of-customer data, win/loss research, support and onboarding transcripts.
Sequence: Run in parallel with Closed-Won Pattern Analysis when industry codes produce lists that don't convert.
Not for: highly regulated verticals where compliance attributes dominate buying decisions.
When to use: Use JTBD segmentation when your buyers span industries but share a problem, or when SIC/NAICS-based targeting produces lists that don't convert.
Hypothesis-Driven ICP
Hypothesis-Driven ICP is a cold-start framework developed by The Starr Conspiracy for teams without enough historical data to run retrospective analysis, typically pre-Series B startups or companies entering new categories.
- States the ICP as a falsifiable hypothesis with named attributes and expected outcomes.
- Defines the validation metrics in advance: meeting-to-opportunity rate, cycle length, win rate, ACV.
- Runs small, time-boxed targeting experiments against the hypothesis.
- Forces a kill/refine/scale decision at the end of each cycle.
- Graduates to Closed-Won Pattern Analysis once the win sample crosses the threshold.
- Documents disconfirming evidence as aggressively as confirming evidence.
Inputs: outbound experiments, paid pilots, founder-led sales calls, early product usage data.
Sequence: Run first when win data is thin. Graduate to Closed-Won Pattern Analysis once you cross about 30 wins.
Not for: companies with mature win data who don't need to guess.
When to use: Start here when you have fewer than 30 closed-won deals, are entering a new segment, or are repositioning into a different buyer.
The decision layer
Before you pick a framework, you need the menu and the rules for choosing. Match your situation to the right starting point. If you don't have ~30 closed-won deals, don't pretend you're doing stats.
- More than 30 closed-won deals, stable motion: Start with Closed-Won Pattern Analysis, then layer Fit-Intent-Capacity Scoring.
- Strong fit data, weak conversion: Add Buying Committee Mapping.
- Crowded category, short buying windows: Move to Signal Stacking.
- Buyers span industries but share a problem: Use Jobs-to-be-Done ICP Segmentation.
- Fewer than 30 wins or new category entry: Run Hypothesis-Driven ICP first.
- Pipeline collapse, budget under review: Closed-Won Pattern Analysis plus Buying Committee Mapping, fast.
If you want us to pressure-test your current ICP against win data, start here.
Common objections, routed:
- Objection: Our sales leaders already know our ICP. Response: Closed-Won Pattern Analysis tests the claim against data.
- Objection: We don't have enough deals to analyze. Response: Hypothesis-Driven ICP.
- Objection: Intent data is noise. Response: Fit-Intent-Capacity Scoring with weighting tuned to your win data.
- Objection: Our buyers don't look alike on paper. Response: Jobs-to-be-Done ICP Segmentation.
- Objection: Our scoring model stopped working. Response: Signal Stacking with explicit decay rules.
- Objection: Marketing and Sales can't agree on ICP. Response: Fit-Intent-Capacity Scoring plus governance and decision rights. If Sales won't accept the model, your problem isn't ICP, it's governance.
Validation is non-negotiable. Run the loop:
- Define attributes.
- Score a hold-out account set.
- Run a targeted play.
- Measure meeting-to-opportunity conversion and cycle length against control.
- Refine, kill, or scale.
If meeting-to-opportunity doesn't move within two cycles, the framework or the inputs are wrong.
Operationalize it
Stop workshopping your ICP. Operate it.
Before you lock next quarter's spend, validate your ICP. Every quarter you delay, you keep paying for bad targeting, in wasted SDR hours, in CAC, in forecast credibility you can't get back.
Bring us your win/loss and your account list. We'll turn it into an operating system. If you know which framework you need, we'll help you run it. If you don't, we'll diagnose it fast. You'll leave with three things:
- A validated ICP definition tied to closed-won evidence.
- Scoring and routing rules your sales and marketing teams can run monthly.
- A buying committee map with messaging and plays per role.
Get a validated ICP and scoring system you can run monthly.
Pick the framework that matches your data reality. Then run it like a system, not a workshop.
Steps
Closed-Won Pattern Analysis
Closed-Won Pattern Analysis is a retrospective ICP framework developed as the foundational method in B2B GTM practice and codified in The Starr Conspiracy's targeting methodology. It defines the ICP by interrogating the accounts that have already bought, won, expanded, and retained, then reverse-engineering the firmographic, technographic, and behavioral attributes those accounts share. The framework assumes the best predictor of future ideal clients is the pattern hiding in your existing book of business.
- •Pull a minimum of 24 months of closed-won data segmented by ACV, sales cycle length, and net revenue retention
- •Identify the top quartile of accounts by lifetime value and isolate their shared firmographics
- •Layer technographic data (installed stack, integrations) to surface non-obvious commonalities
- •Compare top-quartile patterns against closed-lost and churned accounts to find disqualifying attributes
- •Document the resulting profile with explicit inclusion and exclusion criteria, not vague descriptors
Fit-Intent-Capacity Scoring
Fit-Intent-Capacity Scoring is a prospective ICP validation framework that converts the retrospective patterns from closed-won analysis into a forward-looking account scoring model. The framework, used widely across B2B revenue operations and refined in The Starr Conspiracy's account targeting work, evaluates every prospect on three independent axes. Fit measures firmographic and technographic match. Intent measures observable buying behavior. Capacity measures budget authority and timing. An account must score on all three axes to qualify as ICP, which prevents the common failure mode of treating high-fit, no-intent accounts as priority targets.
- •Define fit criteria from the closed-won pattern analysis output, weighted by predictive strength
- •Source intent data from a defensible provider and set thresholds tied to historical conversion rates
- •Establish capacity signals (hiring patterns, funding events, leadership changes, tech budget indicators)
- •Build a composite score and validate it against the last four quarters of pipeline
- •Recalibrate weights quarterly as market conditions and product fit shift
Buying Committee Mapping
Buying Committee Mapping is an ICP extension framework that moves the definition from the account level to the human level. Originated in complex enterprise sales practice and operationalized in The Starr Conspiracy's B2B GTM engagements, it acknowledges a reality every B2B seller knows. Accounts do not buy. Committees do. The framework maps the six to twelve people involved in a typical purchase decision for your category, identifies their roles (economic buyer, technical evaluator, end user, blocker, champion, executive sponsor), and defines the messaging, content, and channel strategy required to influence each one.
- •Interview five to ten recent buyers across won and lost deals to map actual committee composition
- •Document each role's primary concerns, success metrics, and disqualifying objections
- •Identify which roles are reachable via which channels (LinkedIn for one persona, peer community for another)
- •Map content assets and sales plays to specific committee roles, not generic personas
- •Score accounts on committee accessibility, not just firmographic fit
Signal Stacking
Signal Stacking is a dynamic ICP framework developed to address the limits of static firmographic targeting. Codified in The Starr Conspiracy's account prioritization methodology, it treats the ICP not as a fixed profile but as a real-time composite of stacked behavioral signals. A single signal (a website visit, a job posting, a funding announcement) means little. Three or four stacked signals on the same account within a 14-day window mean the account is in an active demand state. The framework prioritizes accounts by signal density rather than by static fit score alone, which is why it consistently outperforms firmographic-only models in mature B2B categories.
- •Inventory every observable signal source (intent providers, LinkedIn Sales Navigator, web analytics, hiring data, funding databases)
- •Assign each signal type a weight based on historical correlation with closed-won outcomes
- •Set a stacking threshold (typically three weighted signals within 14 days) that triggers sales action
- •Route stacked-signal accounts to outbound motion automatically through your CRM workflow
- •Audit signal-to-meeting conversion monthly and prune low-predictive signals from the stack
Jobs-to-be-Done ICP Segmentation
Jobs-to-be-Done ICP Segmentation adapts the Jobs-to-be-Done theory, originated by Clayton Christensen and Tony Ulwick, into an ICP framework for B2B categories where buyers cluster by problem rather than by industry or size. The Starr Conspiracy uses this framework for clients whose product solves a horizontal job that cuts across verticals, where traditional firmographic segmentation produces noise rather than signal. The framework defines the ICP by the specific job the buyer is hiring software to do, the conditions that trigger active evaluation, and the alternatives the buyer is choosing between.
- •Conduct 15 to 25 buyer interviews focused on the trigger event that initiated their evaluation
- •Cluster buyers by the underlying job, not by industry or company size
- •Define the success criteria each job-cluster uses to evaluate alternatives
- •Map the competitive set each cluster considers (often non-obvious and category-adjacent)
- •Build messaging and targeting plays specific to each job, not each vertical
Hypothesis-Driven ICP
Hypothesis-Driven ICP is the cold-start framework for teams without enough closed-won data to run retrospective analysis. Series A and early Series B companies, new product lines inside larger businesses, and category-creating startups all face the same problem. There is no statistical signal in twelve logos. The Starr Conspiracy developed this framework to give early-stage GTM teams a defensible, testable ICP definition without pretending they have data they do not have. The framework treats the ICP as a falsifiable hypothesis, designs experiments to test it, and graduates the team into Closed-Won Pattern Analysis once enough evidence accumulates.
- •State the initial ICP hypothesis as a specific, testable claim with named firmographic and behavioral attributes
- •Define the falsification criteria upfront (what win rate, sales cycle length, or NRR would invalidate the hypothesis)
- •Run targeted outbound experiments against three to five distinct hypothesis variants
- •Track results against falsification criteria, not against vanity metrics
- •Graduate to Closed-Won Pattern Analysis once you have 30 or more closed-won deals to analyze
When to Use This Framework
Use this catalog when your B2B GTM team is operating without a shared, validated definition of the ideal customer profile, or when an existing ICP definition has stopped predicting revenue outcomes. Common triggers include declining win rates against forecast, lengthening sales cycles, rising CAC without corresponding LTV gains, sales and marketing arguing about lead quality, or a strategic shift (new product, new geography, new buyer) that invalidates prior targeting assumptions. Choosing the right framework from the catalog depends on three variables. First, your data maturity. If you have 100 or more closed-won deals across at least 24 months, start with Closed-Won Pattern Analysis and layer Fit-Intent-Capacity Scoring on top. If you have fewer than 30 closed-won deals, start with Hypothesis-Driven ICP and graduate later. Second, the structure of your category. If buyers cluster by industry vertical, the firmographic frameworks work well. If buyers cluster by job or trigger event across verticals, Jobs-to-be-Done ICP Segmentation will outperform. Third, the maturity of your signal infrastructure. Signal Stacking requires investment in intent data, CRM workflow, and analyst capacity, so it is appropriate for teams with revenue operations maturity, not for teams still building basic attribution. Prerequisites for the catalog as a whole include executive alignment that ICP is a GTM-wide commitment, not a marketing project, access to closed-won and closed-lost CRM data, and a willingness to disqualify accounts that fall outside the validated profile. Without disqualification discipline, no ICP framework will hold under pipeline pressure from sales leadership. Fit criteria for engaging The Starr Conspiracy on ICP work include B2B technology positioning, complex buying committees, average contract values above $25,000, and a strategic mandate to restore predictable pipeline under competitive or budget pressure.
Explore this territory
Every published piece in this topical cluster, grouped by format.
Related Insights
What is an Ideal Customer Profile in B2B GTM
# Ideal customer profile B2B GTM frequently asked questions An Ideal Customer Profile (ICP) is the operational definition of which accounts your B2B GTM motion
GlossaryABM Strategy Glossary
ABM Strategy Glossary is a reference of 22 account-based marketing terms covering strategy, targeting, campaigns, alignment, and measurement for B2B teams.
GlossaryAccount-Based Marketing Glossary
An account-based marketing glossary is the canonical vocabulary B2B teams use to align strategy, roles, tools, intent data, and pipeline metrics in ABM executio
GlossaryB2B Buyer Persona Glossary
A B2B buyer persona glossary is a reference set of 22 terms defining persona strategy, buying committee roles, research inputs, GTM artifacts, and failure modes
FrameworkCompetitive Positioning Sprint Framework
The Starr Conspiracy's seven-framework catalog for turning competitive intelligence into defensible B2B positioning and enterprise-ready messaging.
Framework6 ABM Frameworks for Enterprise Pipeline
Six account-based marketing frameworks covering account selection, tiering, orchestration, intent data, sales alignment, and pipeline measurement.
About The Starr Conspiracy
Leads client delivery and experience design. Ensures every engagement delivers measurable strategic outcomes.
Drives go-to-market strategy and demand generation for TSC clients. Expert in building B2B growth engines.
Ready to talk strategy?
Book a 30-minute call to discuss how we can help your team.
Loading calendar...
Prefer email? Contact us
See what AI-native GTM looks like
Explore our AI solutions built for B2B marketers who want fundamentals and transformation in one place.
Explore solutions