Diagnostics · Answered by Golden Section from more than 400 B2B software companies observed
Usually nothing is wrong with closing. The problem sits earlier: leads that were never qualified, deals that advance stages on activity rather than evidence, demos and pricing shown before value, and a champion's enthusiasm mistaken for the company's commitment. Measure conversion stage by stage for the last four quarters and find where volume falls off. If lead-to-qualified collapses, you have a targeting problem. If qualified deals die late, you have a value, pricing or buying-committee problem. Our planning assumption is that genuinely qualified deals close 15–25% of the time; well below that, qualification is loose or late-stage selling is broken. Next step: pull every closed-lost deal from the last two quarters and record the stage and the real reason it died.
Fix the stage that leaks before adding leads. A deal moves forward only on verifiable evidence, and effort at each stage should never cost more than that stage is worth.
| Metric | Value | What it means | Source |
|---|---|---|---|
| Qualified-deal close rate | 15–25% | deals won ÷ deals that passed the qualification stageGS planning assumption under normal conditions; use your own history once you have it | Golden Section playbookEnterprise Sales Process play |
| Cost of effort at the wrong stage | $200K vs $11.1K per closed deal | stage cost ÷ close probability at that stageA $10K onsite demo at a 5%-likely stage versus the same demo at a 90%-likely stage; worked example in the play | Golden Section playbookEnterprise Sales Process play |
A funnel leaks at a specific stage, and the fix depends entirely on which one. Adding leads to a funnel that loses them after qualification just produces more lost deals at a higher cost. The sales funnel play defines stages by a concrete milestone, a time frame and an expected conversion rate, which is what lets you see the leak instead of arguing about it.
Three patterns explain most of what we see. Qualification is soft, so the team spends its week on deals that were never going to buy. Stages advance on activity, so a demo counts as progress even when the buyer has not confirmed the problem or its cost. And value arrives after price, so the buyer compares you to cheaper tools on features. The enterprise sales process fixes the second by requiring evidence at each stage, and the value proposition and customer ROI play fixes the third with a monetized cost of the problem. Weekly pipeline management then keeps the stage data honest, which is where most founders discover their pipeline was smaller than it looked.
A company generates 300 leads a quarter and closes six deals. The founder assumes the reps cannot close. The stage data says otherwise: 120 leads reach a demo, only 30 of those were ever qualified against budget, authority and a forcing event, and the qualified deals close at 20%. The other 90 demos produce nothing. The fix is to move qualification ahead of the demo, drop the demo from the first meeting, and hand the reps a one-page cost-of-problem sheet for discovery. Lead volume falls; closed deals rise, and the reps get back the hours they spent on 90 demos. All figures are invented for illustration.
If the win rate on qualified deals is in range and total volume is simply too low, the problem really is top-of-funnel and more leads will help. Very low-priced, self-serve products run a different funnel, where conversion is a product and onboarding question more than a sales one.
From the Golden Section mistakes list, each paired with the play that prevents it.
Unqualified leads fill the calendar and drag the close rate down while looking like pipeline.
An early demo keeps unqualified prospects in the process and spends the most expensive resource on the least likely deals.
Many conversations and few closes is exactly the pattern this question describes.
Late-stage enterprise losses usually come from treating one champion's enthusiasm as the organization's decision.
In the order we would run them. Each is on its own page, most with a free Excel template.
Defines stages, milestones and conversion benchmarks so the leak can be located.
Requires verifiable evidence at each stage and matches effort to close probability.
Puts a monetized cost of the problem in front of the buyer before price.
Identifies who actually buys and who can block, which late-stage losses usually reveal was guessed.
Keeps weekly stage data honest, so rotting and unqualified deals get moved back or dropped.
Sales and marketing plays Funnel design, qualification, value proposition and pipeline review are all here, in the order a leaking funnel needs them.
Measure it on qualified opportunities, not raw leads, or the number is meaningless. Our planning assumption is 15–25% for genuinely qualified deals under normal conditions, and your own trailing history should replace it as soon as you have enough deals to count.
It depends entirely on where the demo sits. If the demo happens after qualification, demo-to-close should look like your qualified win rate; if it happens in the first meeting, the rate will be low and the demos expensive. We would move the demo later rather than chase a benchmark.
Look at the stage where they die and who said no. Most late enterprise losses trace to a champion who could not carry the decision, a value case that was never monetized, or effort spent at low-probability stages. Map everyone above and beside your champion who can stop the deal, and require evidence before a deal advances.
Reviewed by Dougal Cameron, CEO & Co-Founder on 2026-09-23. Golden Section observations are labeled separately from external benchmarks and illustrative arithmetic.