Your ICP is your ABM foundation. But most ICPs are too broad, too old, or based on internal assumptions instead of actual data. A tight ICP means sales and marketing chase the same 100-150 accounts. A loose ICP means they chase different universes of 500+ accounts and waste each other's time. This guide maps how to refine your ICP using data, not intuition.
The first mistake: defining ICP once and treating it as static. The CEO says "we target enterprise software companies between $50M and $500M revenue with 50+ person sales teams in NOAMER" and that becomes the ICP. But actually, after 18 months of selling, you've learned that $75-150M companies with 30-person sales teams close faster. You've learned that APJ region converts better than EMEA. But the ICP never updates.
The second mistake: basing ICP on wishful thinking instead of closed deals. You want to sell to enterprises (it's prestigious), so your ICP is "$1B+ revenue, 2,000+ employees." But historically, 90% of your customers are $100-300M, 500-person companies. Your ICP is fiction.
The third mistake: creating an ICP that's too broad to be useful. "Mid-market software companies" is not an ICP, it's a market segment. An ICP should be tight enough that your sales team can have a specific conversation with that company. If your ICP matches 2,000 accounts and you can only close 30 per year, your ICP isn't tight enough.
Step 1: Analyze Your Closed Deals
Pull the last 24 months of closed-won deals (or as many as you have). For each customer, document:
Then segment by profitability: - Tier A: High ACV, fast sales cycle, high retention, high gross margin - Tier B: Moderate ACV, moderate sales cycle, good retention - Tier C: Low ACV, long sales cycle, mediocre retention
Example: You've closed 40 deals. 12 are Tier A (avg $150K ACV, 8-week cycle, 95% retention). 20 are Tier B (avg $40K ACV, 12-week cycle, 75% retention). 8 are Tier C (avg $8K ACV, 20-week cycle, 40% retention).
Now identify the firmographic profile of each tier:
Tier A profile (most profitable): - Revenue: $75-200M (average $120M) - Employees: 400-1,500 (average 800) - Industry: SaaS + Financial Services - Region: NOAMER primarily - Growth: 15%+ YoY - Customer profile: Series C+ funded, 5+ years old, public or planning IPO
Tier B profile: - Revenue: $30-150M - Employees: 100-800 - Industry: Software + Services - Region: NOAMER + EMEA - Growth: 5-20% YoY
Tier C profile: - Revenue: $10-75M - Employees: 50-400 - Industry: Mixed - Region: All regions - Growth: <10% YoY
Your refined ICP = Tier A profile (because that's where money is).
Step 2: Analyze Your Deals Closed, Lost, and Stalled
Now pull deals that went to close (lost) and deals stuck in pipeline (stalled 90+ days). For each, document the same fields plus: - Why did it close lost? (price, product gap, competitor won, wrong buyer, timing) - Where in pipeline did it stall? (discovery, demo, proposal, negotiation)
Cross-reference with Tier A profile. Pattern matching: - Did we lose deals to competitors who were cheaper? (ICP might be wrong - if price is blocker, maybe ICP is too low-ACV) - Did we stall in discovery? (wrong persona reached, wrong use case) - Did they have the right org structure to buy? (if we couldn't find a budget owner, maybe our ICP role/title was off)
This tells you where your ICP definition is missing constraints. Maybe your ICP needs "must have a dedicated RevOps person" because companies without RevOps never prioritized your tool.
Step 3: Build Your Refined ICP Statement
Write a tight ICP statement, not a loose one.
Bad ICP: "Mid-market software companies with 100+ employees and $30M+ revenue." Better ICP: "SaaS companies, Series B or later, $75-300M revenue, 400-2,000 employees, NOAMER-based, spending $50K+ annually on sales enablement, with dedicated revenue ops."
The refined ICP includes: - Vertical specification (SaaS, not just "software") - Maturity stage (Series B+, not "any funding") - Revenue range (specific, not open-ended) - Headcount range - Geography - Tech stack or buying behavior (e.g., "already using Salesforce") - A negative (what you explicitly don't want: "not hardware companies")
Specificity drives focus.
Step 4: Validate Your Refined ICP Against Current Pipeline
Pull your current pipeline. How many accounts are in-market fit ICP? - Ideal: 60-70% of your pipeline matches your refined ICP - Warning: <40% match, which means you're chasing too many wrong accounts - Red flag: >80% match, which means your ICP might be TOO narrow or your pipeline generation is too filtered
If only 30% of your pipeline matches refined ICP, you have a problem. Either: - Your refined ICP is too narrow (you're excluding good customers) - Your lead generation is poor (you're filling pipeline with wrong accounts) - Your sales team is chasing shiny objects (inbound that doesn't fit ICP)
Pick one and fix it.
Once your ICP is refined, divide your target market into tiers:
Tier 1 (Core ICP): Exactly matches your refined ICP. ~100-200 accounts globally. - Action: Direct sales, ABM, executive engagement - Expect: Highest conversion rate, longest sales cycle, highest ACV, best retention
Tier 2 (Lookalike ICP): Matches some (but not all) ICP criteria. Maybe $50-75M (just below ICP), or they're the right size but wrong vertical. ~500-1,000 accounts. - Action: Account-based marketing, standard sales motion - Expect: Moderate conversion rate, moderate ACV, good retention
Tier 3 (Expansion/Upmarket): Below ICP but potential to grow into Tier 1. Startups, smaller companies. Or customers already buying similar solutions. ~2,000-5,000 accounts. - Action: Scaled nurture, product-led motion, land-and-expand - Expect: Low conversion rate, low initial ACV, strong expansion potential
This three-tier universe prevents your team from chasing everything. Tier 1 = focus. Tier 2 = opportunistic. Tier 3 = nurture.
Your ICP should be reviewed quarterly. For each quarter:
Example quarterly review: - Last quarter: 8 new customers, 7 match refined ICP, 1 is Tier 2 (good sign - ICP is working) - Market trend: AI adoption accelerating in your space, which is widening your addressable market - Action: Consider expanding Tier 1 to include companies showing AI-adoption intent (new ICP dimension: "actively evaluating AI tools") - Capacity: Your sales team closed 8, but there are 40 Tier 1 accounts in pipeline. You have capacity. Good.
Most teams define ICP on firmographics only (revenue, employees, industry). But these dimensions matter too:
Your refined ICP should include at least 3 firmographic dimensions and 2 behavioral/structural dimensions.
Once refined, communicate your ICP to the entire team: - Sales: "These are the accounts we're hunting. Here's why they're better (faster sales cycle, higher ACV, better retention)." - Marketing: "These are the accounts we're targeting. Here's the messaging that resonates with each tier." - Customer Success: "These are the accounts we're protecting. Here's why they're high-value."
Post the ICP in Slack. Embed it in your CRM. Review it monthly in sales meetings.
Abmatic automates ICP refinement by analyzing all your closed deals, lost deals, and current pipeline and identifying the firmographic and behavioral profile of your most profitable customers. Instead of manually pulling data, Abmatic:
This removes the manual guesswork from ICP definition and keeps it current.
Watch for these signals that your ICP needs refinement:
Red Flag 1: Your Win Rate Doesn't Match Your TAM You've defined your ICP as $100M+ companies. But your highest win rate is with $75-150M companies. Your ICP definition is missing context.
Red Flag 2: Sales is Chasing Outside ICP Your ICP says "software companies," but your best customers are "software companies in the insurance vertical." Sales knows this but keeps chasing generic software. Your ICP isn't specific enough.
Red Flag 3: Sales Cycle Variance is Huge Some Tier 1 accounts close in 8 weeks. Others take 24 weeks. This suggests your ICP is bundling very different buying motions (maybe some are product-focused, some are security-focused).
Red Flag 4: Retention Differs Wildly by Segment You have 95% retention for Tier 1 customers in Vertical A, but 60% retention for Tier 1 customers in Vertical B. Vertical B should be demoted or restructured.
Red Flag 5: Sales and Marketing Disagree on ICP Sales says the ICP is "companies with $10M+ ARR." Marketing says "companies with 100+ employees." You're not aligned. This disagreement will cascade to all other motions.
If you see these red flags, schedule an ICP refinement sprint. Pull the team (sales, marketing, customer success) and re-analyze your closed deals using the process outlined above. Your ICP should be a living document, refined quarterly, not a static decree.
Once refined: 1. Document it: Write a one-page ICP statement. Post it everywhere (Slack, wiki, sales desk). 2. Explain it: Host a 30-min all-hands explaining what changed and why. 3. Cascade it: Tie compensation, quotas, and metrics to the new ICP. 4. Measure it: Track what % of your pipeline matches refined ICP. Target: 70%+.
If less than 50% of your pipeline matches refined ICP after one month, either your ICP is wrong or your lead generation process is broken. Diagnose which.
A tight, data-driven ICP transforms your sales and marketing alignment. Instead of chasing different accounts, both teams focus on the same 100-150 Tier 1 companies where you have the best chance of closing deals fast and profitably. The best ICP is one that's specific enough to guide decisions ("should we pursue this account?"), measurable enough to track ("are we hitting our ICP targets?"), and dynamic enough to evolve as your business learns.
Book a demo with Abmatic to see how account analysis and ICP refinement can help you tighten your target market and focus your team's energy.
Abmatic is a mid-market and enterprise ABM platform that covers all 14 core account-based marketing capabilities in one product, including deanonymization, web personalization, outbound sequencing, multi-channel advertising, AI workflows, and built-in analytics. Pricing starts at $36K/year.
Abmatic covers every capability that 6sense and Demandbase offer, plus adds AI-native workflows, outbound sequencing, and web personalization in a single platform. Most enterprise teams find they can consolidate 3-4 point tools when they move to Abmatic.
Yes. Abmatic is purpose-built for mid-market and enterprise B2B companies. It is not designed for early-stage startups or SMBs. Enterprise pricing is available on request; mid-market plans start at $36K/year.
Singapore has established itself as the undisputed technology and financial services hub of Southeast Asia, hosting thriving ecosystems of enterprise technology companies, fintech innovators, and business software vendors serving regional markets across Southeast Asia and beyond. Singapore-based...
Pipeline coverage is the ratio of open pipeline (deals currently in your sales process) to your revenue target for a given period, typically expressed as a multiple (e.g., 3:1 coverage means you have $3 million in pipeline for every $1 million in revenue target).