Every fund starts with a spreadsheet. It has a company name, a source, a stage, and a partner. It works for the first fifty companies. It breaks somewhere between one hundred and two hundred. The break is usually not sudden — it is a slow decline in the firm's ability to remember which companies it has seen, why it passed, and whether the reasons still hold.
Scaling deal flow is not about buying a bigger CRM. It is about building a small set of operating disciplines that keep the pipeline honest as the volume grows. This piece is the four disciplines and the four traps.
Discipline 1: one stage per company, one owner per stage
The single most common pipeline pathology is the company that sits in three stages at once because three partners are tracking it. The fix is discipline, not tooling: every company sits in exactly one stage, with exactly one partner as the owner, with exactly one next action and a next action date.
This is not bureaucracy. It is what makes the weekly pipeline review possible. When a partner asks 'what is happening with X', the answer is a single row, not a conversation.
Discipline 2: the pass memo
Every meaningful pass should generate a two-sentence pass memo: the reason, and the condition under which the firm would reconsider. This costs thirty seconds. It saves hours over the next three years, because the same company will come back — often at a different stage, sometimes with a different founder — and the firm will need to remember what it thought last time.
Firms that skip the pass memo end up with two years of institutional amnesia. Firms that keep it end up with a pattern-matching asset that gets more valuable every year.
Discipline 3: the source ledger
Every deal has a source. Sources are the highest-leverage input a fund has, and almost no funds track them well. A simple source ledger — every deal tagged with the person or channel that referred it, updated at ingestion — lets the firm answer questions like 'which of our top-of-funnel sources have produced the most investments' and 'which have produced the most write-offs'.
The answers usually surprise the partners. The best-referring source is rarely the loudest one.
Discipline 4: the weekly review
The pipeline is a garden. It has to be walked every week or it turns into a graveyard of half-dead leads. The weekly review is thirty minutes, all partners in the room, all live deals surfaced by stage. Every deal either moves forward or gets killed. Dead leads that have not been killed are worse than rejections — they steal attention.
A pipeline that has not been reviewed in two weeks is not a pipeline. It is a list.
The four traps
Trap 1: too many stages
Every fund is tempted to add stages — screening, first call, deep dive, thesis fit, portfolio fit, partner review, term sheet drafted, term sheet signed, closed. Six or seven stages is where funds land after they realize they had twelve. The right number is small enough to remember and specific enough to reveal blockers.
Trap 2: fields that nobody fills in
Every fund is tempted to add fields. Sector, sub-sector, geography, ARR band, growth rate, stage, source, thesis tag, partner sponsor. Fields that partners will not fill in are worse than useless — they degrade trust in the data that does exist. Keep the fields to the ones that get filled in every time.
Trap 3: the portfolio pipeline
Once companies are portfolio, they need their own pipeline discipline — support asks, hiring intros, follow-on decisions, valuation events. Firms that try to run this in the same pipeline as new deals conflate two very different objects. Split them. New deals live in one system. Portfolio companies live in another view on the same graph.
Trap 4: the CRM that is not the source of truth
Some firms have three sources of truth for the pipeline: the CRM, the partner's inbox, and a shared spreadsheet. This is fatal. When there is more than one place a company might live, no one trusts any of them. The fix is not more integrations. It is a single object per company that all views read from.
Scaling past two hundred companies
At two hundred companies in an active pipeline, memory stops working. The disciplines above stop being nice-to-haves and start being the difference between a firm that operates the pipeline and a firm that the pipeline operates.
The firms that scale past this number typically also invest in two second-order practices. First, structured sourcing: named channels, named campaigns, named events, each with a target and a review. Second, weekly deal-flow health metrics: number sourced, number to first call, number to term sheet, conversion rates between stages. Not to pressure partners — to make blockers visible.
Where Equavion fits
Equavion's pipeline is one graph across sourcing, diligence, closing and portfolio. One company object across every stage. One next action, one owner, one stage per company. Structured pass memos and source tagging as first-class fields, not custom columns. The company hub is the same graph — the deal that closes today becomes the portfolio company you support tomorrow, without a data migration.
Takeaways
- One stage per company, one owner per stage, one next action.
- Every meaningful pass gets a two-sentence memo. Institutional memory compounds.
- Track sources. The best-referring channel is rarely the loudest.
- Review the pipeline weekly. A pipeline that is not reviewed is a list.
- Keep the number of stages and fields small enough that everyone fills them in.
