Closed-Won Analysis: What Your Best Deals Have in Common

Every sales team has a folklore answer to “who is our ideal customer?” It usually lives in the VP Sales’ head, arrived at through three years of close calls, and it is almost always partly wrong. Closed-won analysis is how you replace folklore with a pattern that the CRM can enforce.

Done well, the exercise takes about a day per quarter and changes how the next quarter’s pipeline is built. Done poorly, it produces a slide deck that nobody references again. The difference is almost entirely in the discipline of the seven signals below and the decision to weight by revenue instead of deal count.

This guide walks through the definition, the seven signals, the quarterly cadence, and the three places the output has to land to actually matter. It pairs with the Cluster 2 post on RevOps vs Sales Ops and the pipeline hygiene guide in the hub.

What is closed-won analysis?

The exercise looks simple. Pull every closed-won deal from the last quarter, put them in a spreadsheet, and look for what is common. The complication is which columns to include, how to weight them, and where to draw the line between real pattern and noise. A team that does this with three columns will get a sharper ICP than one that does it with thirty.

It is not the same as win/loss analysis. Win/loss compares wins with losses, usually through buyer interviews, and answers “why did we win or lose this one?” Closed-won analysis looks only at wins and answers “who are we actually good at selling to?” The two are complements, not substitutes.

The seven signals worth extracting

Stop at seven. Any more columns and the analysis becomes a research project that ships too late to use. Any fewer and the ICP comes out too narrow.

1. Firmographics. Industry, sub-vertical, employee count, revenue band, region, and tech stack signals. The core question: what does the best-fit company look like before anyone talks to them?
2. Contract size (ACV). Segment by deal size band rather than averages. Whale deals, mid-market, and SMB win for different reasons and should not be collapsed into one pattern.
3. Sales cycle length. Time from first touch to close, in days. Short cycles usually signal a well-matched use case. Long cycles often signal a misfit that got over the line anyway.
4. Source channel. Where the deal originated: inbound, outbound, partner, event, referral, PLG self-serve. Different sources produce different closed-won shapes.
5. Stakeholder role. Title and seniority of the economic buyer and primary champion. A deal that closed through a director of RevOps looks different from one closed through a CFO.
6. Product usage pattern. In the first 30 days post-close, which features got adopted, by how many users, and how often. This is the bridge between sales and expansion.
7. Expansion behavior. Did the account upgrade, add seats, or add products in the first 90 days? The accounts that expand quickly are the real ideal customers.

How to actually run the analysis

A one-day operator exercise, not a consulting project. The workflow:

1. Pull wins from the last 90 days. Export from HubSpot, Salesforce, or Pipedrive. Minimum sample: 20 deals. Below that, treat the output as qualitative.
2. Enrich with billing and usage. Join Stripe or the billing system for actual collected revenue, not booked ARR. Join the product analytics tool or the CS platform for early usage.
3. Segment into three ACV bands. Whale, mid, SMB — or whatever the natural break in your data is. Analyze each band separately.
4. Score each signal by revenue contribution. If 62% of revenue came from accounts 200–1000 employees, note the band. If 38% of revenue came from partner-sourced deals, that is a target signal for next quarter.
5. Write the two-line ICP per band. “Mid-market wins: Series B SaaS companies, 300–800 employees, HubSpot as CRM, director-level buyer, 42-day cycle.” Two lines per band beats a ten-page deck every time.
6. Feed the results into three places. CRM lead scoring rules, BDR outbound target list, and marketing targeting criteria. If the output does not change behavior in those three systems, the analysis did nothing.

What the output should look like

A useful closed-won deliverable is shorter than most teams expect. For a typical B2B SaaS with 40 wins per quarter, the final artifact is one page per ACV band:

That table — with the specifics of your business — is the closed-won analysis. Everything else is commentary. The three distinct ICPs become the three different playbooks marketing, sales, and CS each run.

The three places the result has to land

A closed-won analysis that does not change behavior in one of these three systems was a waste of a day:

• CRM scoring rules. Lead scoring should weight the firmographics the closed-won data identifies. If 200–800 employees is the mid-market sweet spot, that band should add 30 points. Anything outside it, subtract points.
• BDR outbound target list. Next quarter’s BDR list should over-index on the patterns that closed-won data validated. If partner-sourced Whales produce 42% of revenue, the partner team gets quota, not just MDF.
• Marketing targeting criteria. Paid social audiences, paid search campaigns, and ABM lists all inherit the firmographic filters from the analysis. This is where most of the ROI shows up — targeting the wrong companies is the single largest marketing waste at growth stage.

The three mistakes that kill the exercise

1. Treating booked ARR as revenue. Some of those deals will churn in 90 days. Use collected revenue from Stripe or the billing system, or at minimum net of early churn.
2. Analyzing deal count instead of revenue-weighted wins. The SMB motion usually wins the deal-count race and loses the revenue race. Both patterns matter, but only if they are kept separate.
3. Skipping the feedback loop. Running the analysis and never updating the CRM scoring is the most common failure. If nothing in the downstream systems changes, the exercise is theater.

How Fairview runs closed-won analysis automatically

Fairview connects to HubSpot, Salesforce, Pipedrive, Stripe, QuickBooks, Xero, Shopify, Google Ads, Meta Ads, and HubSpot Marketing Hub via native OAuth. Once connected, the operating view segments closed-won deals by ACV band, enriches with billing and source data, and surfaces the firmographic and motion signals for each segment without a warehouse project.

When a pattern shifts quarter over quarter, Fairview writes a named next-best action: "Whale deals >$150K shifted from partner-sourced to direct-sourced this quarter. Cycle time extended from 87 to 112 days. Recommendation: rebuild the partner qualification track before Q3 hiring."

See pricing and tiers for the plan that fits your stack.

Key takeaways

• Closed-won analysis replaces ICP folklore with pattern evidence.
• Seven signals: firmographics, ACV, cycle, source, stakeholder, usage, expansion.
• Weight by revenue, segment by ACV band, minimum 20 wins per window.
• Output lands in CRM scoring, BDR target lists, and marketing criteria — or it was theater.
• Run it quarterly. Track what shifts between quarters as carefully as what stays the same.

Frequently asked questions