VC Deal Pipeline Stages Explained: From Signal to Term Sheet
The seven stages of a VC deal pipeline, what happens at each, the conversion benchmarks between stages, and how leading indicators compress the most valuable stage transitions.
Key Takeaway
Every fund's CRM has different stage names, but the underlying pipeline is universal. This explainer maps the seven stages from universe to close, defines what actually happens at each, gives the conversion benchmarks that separate disciplined funds from spraying ones, and shows where leading indicators (like engineering acceleration) compress the highest-leverage transitions: signal to watchlist, and watchlist to first meeting.
Open five different fund CRMs and you will find five stage taxonomies: SDI, New, Screening, IC, Partner Meeting, on and on. The vocabulary differs, but the machine is identical underneath. This explainer fixes the vocabulary so the rest of your sourcing operation can be measured, because a pipeline you cannot name stage by stage is a pipeline you cannot improve.
The seven stages#
- Universe. Every company that could fit the thesis: stage range, sector range, geography range. Practically, the set your channels can reach. Funds rarely write this down, which is the first mistake: an undefined universe makes every downstream metric meaningless.
- Signal. A reason to pay attention: a referral, a thesis match, a data trigger. Raw signals are cheap and noisy; the art is triage. Most funds generate 10-100x more signals than they can engage.
- Watchlist. Companies under active observation, not yet contacted. This is the stage where leading indicators do their work: a company whose engineering acceleration crossed your threshold last week sits here, uncontacted, accumulating context while you verify.
- Screened. Quick qualification against hard filters: stage, geography, team composition, traction floor. The purpose is deletion, not ranking. Most funds kill 90 percent of the watchlist here, correctly.
- Engaged. First real conversation with the founder. The metric that matters is watchlist-to-engaged conversion, because it measures signal quality: engage too low a share and your signals are noise; engage too high and you are spraying meetings.
- Diligence. Data room, references, technical review, code-level checks. Diligence is where funds die by process: too little and you fund errors, too much and founders walk.
- Term sheet and close. Including the post-signing crawl of exclusivity and definitive docs.
Some funds insert sub-stages (IC preview, partner meeting) between 6 and 7; the skeleton holds.
Conversion benchmarks worth knowing#
No public dataset publishes fund-by-fund funnel math, but the pattern from practitioner reporting is stable enough to plan against:
- Watchlist to screened: expect to cut 80-95 percent. If you screen out less, your signals are too weak; more, and you may be over-filtering early sight.
- Screened to engaged: the target band is roughly one in five. This is the signal-quality ratio.
- Engaged to diligence: one in three to one in five meetings earns diligence.
- Diligence to term sheet: one in three to one in ten, stage-dependent.
- Term sheet to close: 80-90 percent (the losses here are reputationally expensive).
Multiplied out, a fund doing one deal a quarter from a thousand raw signals needs every stage honest. The deal flow scoring framework packages these ratios into a weekly scorecard, and the deal flow management guide covers the pipeline hygiene (attribution, aging rules, stage discipline) that keeps the numbers real.
Where leading indicators compress the funnel#
Traditional pipelines are limited by one asymmetry: the interesting transitions (signal to watchlist, watchlist to engaged) depend on information that arrives late, at announcement. Leading indicators flip the asymmetry. A fund watching engineering acceleration sees the signal three to six weeks before the databases do (median, across the 350+ organization panel this site tracks), which means the watchlist stage accumulates context while competitors do not yet know the company exists.
The compounding effect: earlier signal means earlier engaged conversations, which means diligence starts before the round is competitive, which means better prices and better information rights. The funnel ratios stay the same; the calendar moves in your favor. The how VC firms find startups early post covers the full pre-announcement stack, and the weekly sourcing workflow schedules the cadence.
Stage discipline: the boring practice that separates funds#
Three habits keep a pipeline honest. First, time-in-stage limits: any company older than 90 days in a stage either advances or exits; zombie pipeline inflates every downstream metric. Second, single-owner deals: every deal has one name on it, or attribution rots. Third, exit coding: when a company leaves the pipeline, record why (stage, sector, traction, founder choice); the quarterly review of exit codes is where thesis evolution actually happens. The sourcing best practices guide expands these into a full checklist, and the emerging manager sourcing playbook applies them at solo-operator scale.
Tools: what actually matters#
Funds over-invest in CRM software and under-invest in stage discipline. The commercial platforms (Affinity, Attio, and peers) differ mostly in relationship intelligence and automation; none of them fix a funnel with dishonest stages. The best deal flow tools comparison covers the field honestly, including where free and open datasets outperform paid ones for early discovery. For pipeline analytics specifically, a spreadsheet plus honest attribution beats a misused enterprise CRM every quarter of the year.
Key takeaways#
Name your stages, measure the ratios, enforce time limits, and feed the top of the funnel with signals that lead the market instead of trailing it. The seven-stage skeleton is universal; the discipline is rare, and that is the whole game. Everything else, tools included, is furniture.