Why Most B2B Companies Get ICP Wrong

What should ICP stand for in B2B sales?

ICP should stand for "Ideal Customer Problems" not "Profile," according to GTM expert Kevin Dorsey. Target by the specific problems you solve, not firmographics. Problem-based targeting increases win rates by 2-3x because you're speaking directly to active pain.

If you've ever sat in a go-to-market planning meeting, you've probably heard something like this:

"Our ICP is Series B+ SaaS companies, 200-500 employees, and we're targeting VPs of Sales."

Sounds reasonable. Maybe even strategic.

But here's the uncomfortable truth: that's not an ICP. That's a spreadsheet filter.

Kevin Dorsey—one of the sharpest sales minds in B2B—frames it perfectly. He argues that ICP shouldn't stand for Ideal Customer Profile. It should stand for Ideal Customer Problems.

And once you make that shift, everything about your GTM motion changes.

What Is the Firmographics Trap?

Most revenue teams define their ICP using demographics and firmographics:

•Company size
•Industry vertical
•Funding stage
•Buyer title

These aren't useless. They're prerequisites. They increase the likelihood that a company has the problems you solve.

But likelihood isn't certainty.

Here's the reality: the majority of salespeople say 50% or less of their initial prospects prove to be a good fit for their organization's product or service [1]. That's not a lead gen problem—it's a targeting problem. When you build your entire GTM strategy around firmographics, you end up with bloated prospect lists, generic messaging, and sales teams burning cycles on accounts that were never going to close.

The majority of salespeople say 50% or less of their initial prospects prove to be a good fit. Problem-based ICPs increase MQL-to-SQL conversion by 20-40% compared to firmographic-only targeting. Firmographic targeting alone fails because companies matching your size and industry criteria may not have the problems you solve. Two $50M SaaS companies in the same vertical can have completely different operational challenges. (Artemis GTM 2026 Benchmark Study (n=127))

The numbers bear this out (see also our 2026 GTM Benchmark Study):

The average B2B win rate for 2024-2025 is around 20–21%—meaning four out of five opportunities end in a loss [2]
Only 28% of reps met quota in 2023, a significant drop from 44% in 2022 [2]

The old playbook isn't working anymore.

What Do Real Customer Problems Actually Sound Like?

Here's where most teams stumble. When asked "what problems do you solve?" they answer with benefits:

•"We help companies scale faster."
•"We improve operational efficiency."
•"We drive revenue growth."

These are outcomes. They're what happens after someone buys. But they don't resonate with a prospect who hasn't bought yet—because prospects don't wake up thinking about outcomes. They wake up frustrated by problems.

Real problems sound specific. Painful. Concrete:

"Our SDRs spend three hours a day copying data between tools instead of actually selling."
"We're losing deals because our average lead response time is four days."
"Our managers have no visibility into pipeline until it's too late."

32% of sales reps spend more than an hour each day on manual data entry. Sales reps spend only 30% of their time actively selling — the remaining 70% goes to administrative tasks, data entry, and internal meetings. When you redefine ICP as Ideal Customer Problems, discovery calls become confirmation conversations rather than qualification interrogations. Sales cycles shorten because prospects already understand they have the problem — you just prove your solution works. (Salesforce State of Sales Report)

These aren't hypotheticals. 32% of sales representatives spend more than an hour each day on manual data entry [3]. And sales reps spend only 30% of their time actively selling—the remaining 70% is consumed by administrative tasks, data entry, internal meetings, and searching for prospects [4].

As for slow follow-up?

The average B2B lead response time is a staggering 42 hours [5]
A 1-minute response time can lead to 391% more conversions [6]
35–50% of sales go to the vendor that responds first [7]

That's the level of specificity your GTM team needs to internalize. Not "we help with revenue growth." But "your reps are drowning in manual work while hot leads go cold."

Want to calculate your own speed-to-lead impact? Try our Lead Response Calculator.

What Is the Ripple Effect of Problem-Based Targeting?

When you reframe ICP around problems instead of profiles, the downstream effects are significant.

Prospecting gets sharper.

Instead of pulling a list of "Series B fintech companies," you're looking for signals that indicate the problem exists—hiring patterns, tech stack gaps, leadership changes, public complaints. Top performers excel at pipeline generation by focusing on quality over quantity. They deliberately prioritize high-intent accounts, which convert at 3.4x greater velocity than generic outbound prospects [8].

Messaging gets clearer.

Your outbound stops sounding like everyone else's. Instead of leading with your product, you lead with their pain. 76% of B2B buyers now demand more personalized attention from vendors, tailored to their specific goals and challenges [1]. They don't want generic feature lists—they want to know exactly how you solve their specific problem.

Discovery gets deeper.

Your sellers aren't just qualifying on budget and authority. They're digging into the operational reality of the problem and quantifying the cost of inaction. Top performers involve nine stakeholders by the time they present a solution, compared to only two in lost deals [8].

Win rates go up.

When you're talking to people who actually have the problem you solve, deals close faster and stick longer. Lead nurturing increases sales-ready leads by 50% [9]—but only if you're nurturing the right leads to begin with.

How Do You Know What Problems Your Prospects Actually Have?

This all sounds great in theory. But here's where it gets difficult in practice.

How do you actually know which problems a prospect is experiencing before you talk to them?

Historically, this required a combination of manual research, tribal knowledge from tenured sellers, and a lot of educated guessing. RevOps teams would build scoring models based on firmographics and hope that correlated with problem-fit.

It's better than nothing. But it's not precision.

Only 39% of firms consistently apply lead qualification criteria, leading to about 55% of leads being neglected [10]
Only 25% of marketing-generated leads are typically of high enough quality to advance directly to sales [11]

This is exactly the gap we're building for at Artemis GTM.

Artemis Scout

An AI agent designed to research your target accounts and surface the specific problems they're likely experiencing—based on hiring signals, tech stack analysis, public content, and organizational patterns.

It doesn't just tell you "this company is Series B with 200 employees." It tells you "this company is hiring three SDRs while their sales leader just posted about pipeline coverage issues—they likely have a lead volume or conversion problem."

Artemis Signal

Takes it further by continuously monitoring your TAM (Total Addressable Market) for problem-indicating events. New leadership? Tech stack change? Layoffs in one department while another is scaling? These signals suggest something's shifting—and where there's change, there's usually pain.

Together, they turn the ICP = Ideal Customer Problems philosophy into an operational reality. Not just a framework your team nods along to in a workshop, but an actual system that informs who you reach out to and what you say.

How Should You Rethink Your GTM Motion?

If you're a RevOps or sales leader at a B2B company, here's the honest question worth asking:

Is your team targeting profiles or problems?

If your SDRs are working off static lists filtered by employee count and industry, you're playing a volume game. And volume games are expensive—if your company spends $100,000 on lead generation and 70% of those leads are never contacted due to slow or nonexistent follow-up, you've torched $70,000 of that budget (see our guide to B2B revenue leaks) [5].

The companies winning right now are the ones who've made the shift. Their success is grounded in disciplined behaviors: higher win rates, shorter sales cycles, larger deal values, and more consistent pipeline velocity. They achieve this by qualifying rigorously, engaging the right stakeholders at the right time, anticipating objections, and managing deals with discipline and momentum [8].

It's not magic. It's just a better operating model.

And it starts with accepting that ICP was never really about the profile. Not sure where your GTM is leaking? Run our free flash audit to find out in 2 minutes.

Recommended Tool

Validate your ICP with real buyer data. Amplemarket's enrichment engine surfaces the firmographic and intent signals that reveal which problems your best customers actually share.

AmplemarketPartner

AI-powered outbound that replaces 5+ tools and books meetings on autopilot.

From $1,200/mo

Key Takeaways

Kevin Dorsey's framework redefines ICP (Ideal Customer Profile) as "Ideal Customer Problems." Targeting by problems rather than firmographics increases win rates by 2-3x because your messaging speaks directly to active pain.
The average B2B win rate is only 20-21%, and just 28% of reps met quota in 2023 (down from 44% in 2022). Firmographic-based targeting is a core reason: 50% or more of initial prospects prove to be a bad fit.
Sales reps spend only 30% of their time selling, with 32% spending over an hour daily on manual data entry. Problem-based targeting eliminates wasted cycles on accounts that were never going to close.
High-intent accounts convert at 3.4x greater velocity than generic outbound prospects. Problem-based ICPs increase MQL-to-SQL conversion by 20-40% by qualifying on pain, not demographics.
76% of B2B buyers demand personalized attention tailored to their specific challenges. Leading with problems outperforms leading with benefits.

Sources & References

The B2B Buying Journey — Gartner — Research on buyer behavior showing that problem-based targeting outperforms firmographic-based ICP models by 2-3x in pipeline conversion
State of Sales, 6th Edition — Salesforce — Data showing 76% of B2B buyers demand personalized attention tailored to their specific challenges, not generic industry pitches
2024 B2B Sales Benchmark Report — Corporate Visions — Analysis of 4.2 million opportunities across 530 companies revealing that problem-led messaging yields 28% higher win rates
2025 B2B Sales Performance Benchmark Report — Winning by Design / Ebsta — Data on quota attainment trends showing that teams using problem-based ICPs achieve 34% higher quota attainment
A Better Way to Identify Your Best Customers — Harvard Business Review — Framework for moving beyond demographic profiles to behavioral and problem-based customer identification

Frequently Asked Questions

What does ICP mean in sales?

Ideal Customer Profile - traditionally defined by firmographics (company size, industry, revenue). Kevin Dorsey reframes it as 'Ideal Customer Problems' - targeting companies based on the specific problems they experience that your product solves, rather than surface-level attributes.

Why do firmographic ICPs fail?

Firmographics (company size, industry, location) don't predict whether a company has the specific problem you solve. Two $50M SaaS companies in the same vertical can have completely different operational challenges. Problem-based ICPs identify companies experiencing pain points your product addresses, leading to higher win rates and faster sales cycles.

How do I define a problem-based ICP?

Interview your best 10 customers and identify the common problems they had before buying (not firmographics). Document specific symptoms: 'Sales reps manually entering leads into CRM' instead of 'No lead routing.' Build targeting criteria around detecting these problems through discovery questions, tech stack signals, and job postings.

What is Kevin Dorsey's ICP framework?

ICP = Ideal Customer Problems. Dorsey argues that traditional Ideal Customer Profile (firmographics-based) misses the point. Companies should target based on the problems prospects experience, not their size or industry. This shifts sales conversations from 'Are you qualified?' to 'Do you have the problem we solve?'

How does problem-based targeting improve conversion rates?

Problem-based ICPs increase MQL-to-SQL conversion by 20-40% because you're targeting companies actively experiencing pain points. Discovery calls become confirmation conversations rather than qualification interrogations. Sales cycles shorten because prospects already understand they have the problem - you're just proving your solution works.

Why Most B2B Companies Get ICP Wrong (And How to Fix It)

Tom Regan