Deal Sourcing Best Practices for VC: The Habits That Compound

Every sourcing operation looks similar from outside: lists, emails, meetings. The difference between funds whose sourcing compounds and funds whose sourcing decays is not tooling or headcount; it is a small set of habits, run weekly, honestly measured. This guide is those habits, collected in one place, with the anti-patterns that undo them.

Practice 1: Diversify channels like a portfolio#

The first failure mode of sourcing is monoculture: one channel (usually inbound) carrying the fund until it quietly decays. The four sourcing channels (inbound, outbound, network, platform) have different costs, lead times, and failure modes, which is exactly why a portfolio of them is robust. The rule of thumb: no single channel above 50 percent of sourced meetings, reviewed quarterly. When a channel drifts dominant, find out why before it drifts further; dominance usually means the other channels are decaying, not that one is winning.

Practice 2: Prefer leading signals over trailing ones#

Trailing signals (announcements, press, database entries) are visible to everyone at the same moment, which makes them pricing exercises. Leading signals (engineering acceleration, hiring patterns, release telemetry, community activity) arrive weeks earlier and convert the same meeting into a cheaper, better-informed one. The strongest 2026 example is public GitHub activity: across the 350+ organization panel this site runs, sustained acceleration has preceded announcements by a median of three to six weeks (methodology). The practice: for every trailing source in your stack, name the leading source that feeds it, and instrument that one instead.

Practice 3: Verify before outreach, always#

A signal is a reason to look, never a reason to send a calendar link. The verification pass (real vs hack-week burst, stage context, warm path) takes 20 minutes per company with the right checklist, the GitHub due diligence pass being the template. The discipline pays twice: outreach volume drops (reputation preserved) and meeting quality rises (partners stop dreading sourcing meetings). The signal vs noise guide covers the heuristics for the judgment calls.

Practice 4: Attribute honestly, especially when it hurts#

Sourcing debates inside funds are settled by attribution: which channel actually produced the funded deals. Attribution is also where honesty dies, because everyone's favorite channel is the one they own. The practice: tag every first meeting with its true source at the moment of logging (not retroactively, when memory flatters), roll the numbers quarterly, and let a channel die if the data says so. The deal flow scoring framework packages this into a scorecard that survives partner politics.

Practice 5: Run a weekly cadence, not bursts#

Sourcing in bursts after slow partner meetings feels productive and measures terribly; signal systems decay between bursts, and referral paths go cold. The weekly loop (Monday refresh, Tuesday verify, Wednesday outreach, Friday attribution) is documented in the sourcing analyst playbook, and the solo-scale version lives in the weekly sourcing workflow. The cadence matters more than the volume: 15 verified names weekly beats 200 names twice a quarter, every quarter.

Practice 6: Write down the universe#

The cheapest practice and the most skipped: define the investable universe (stage, sector, geography) in writing, so "we missed it" becomes diagnosable. A missed company inside the universe is a signal failure (fixable: better sources); outside it, it is thesis scope (a partner decision, not a sourcing one). Funds that skip this step argue about both interchangeably, which is why those arguments never end. The emerging manager playbook shows the one-page version.

The anti-patterns#

1. CRM theater: meticulous pipeline hygiene on stages nobody enforces. If companies do not advance or exit on aging rules, the CRM is a diary, not an instrument. See pipeline stages.
2. Signal mimicry: adopting a signal because a competitor blogged about it, which by construction erases its edge. Proprietary advantage comes from signals others ignore; the proprietary deal flow guide covers the durable versions.
3. Zombie pipeline: companies aging in stage forever, inflating every metric. The 90-day rule (advance, exit coded, or archive) is the cure.
4. Volume worship: meetings per week as the headline metric, which optimizes for spray. The ratios (watchlist-to-meeting, meeting-to-term-sheet) are the truth; volume is just the denominator.
5. Network extraction: taking referrals without reciprocity. Referral paths rot silently; the sourcing network guide covers the reciprocity accounting.

How to start Monday#

Pick one leading signal source (this site's weekly panel is free and public), cap a 15-name list, run the verify-outreach-attribute loop for four weeks, and read the ratios at the end of the month. That is the entire minimum viable sourcing operation, and every practice above is an extension of it. The free VC data sources guide covers the zero-budget stack for the list-building half.

Key takeaways#

Sourcing excellence is boring: diversified channels, leading signals, verification before outreach, honest attribution, weekly cadence, a written universe. The funds that do all six compound; the funds that do five argue about which one to skip. Start the loop Monday, measure the ratios monthly, and let the habits do the compounding.