Ask a VC who they like to co-invest with and you’ll get a warm, generous, and almost completely useless answer. The names that actually show up on their cap tables are a different list entirely, and the only way to find it is to stop asking and start reading.
Why “who do you co-invest with?” is the wrong question
The stated-relationship trap: friendly lists that never show up on a cap table
Most VCs will happily name five or six funds they “love working with.” Some of those relationships are real. Many are aspirational, outdated, or simply the funds that come to mind first because they had coffee last month. A stated relationship costs nothing to claim and nothing to verify, which is exactly why it’s a bad input for a fundraising decision. You need evidence that two firms have actually put capital into the same rounds, more than once, at a similar stage.
What co-investment actually is: repeated, revealed, round-specific behavior
Real co-investment is a pattern, not a sentiment. It shows up as the same two or three fund names appearing together on priced rounds, repeatedly, over a meaningful window of time. One overlapping deal could be coincidence, a mutual portfolio company that needed a bridge, or a favor between partners. Three or four overlaps across different companies and different years is a structural relationship: two firms that trust each other’s diligence, split work the same way, and show up together because it works for both of them.
The goal: a syndicate graph, not a longer target list
The output of this process isn’t a bigger spreadsheet of fund names. It’s a graph: nodes are firms, edges are confirmed co-investments, and edge weight is how often a pairing repeats. That graph tells you something a flat list never can, which firms move together. This is the natural extension of the sift-not-spray premise in How to Build a Target Investor List for Fundraising: a good list is already filtered for fit. A syndicate graph adds the layer that tells you which filtered names tend to close in groups, so you can pitch one and effectively reach three.
Step 1: Read firms’ portfolios for co-investment overlap
Pull the last 12 to 24 months of a firm’s deals, not its marquee logos
Every fund’s website leads with its best exits and best-known names. Those are usually the worst rounds to study for co-investment, because marquee deals attract opportunistic participants who show up once and never again. The useful data is in the firm’s ordinary, recent activity: the last 12 to 24 months of rounds, pulled from a source like Crunchbase or PitchBook, cross-checked against SEC Form D filings when the round is a private placement, since Form D often lists the participating entities.
The overlap signal: which other funds appear repeatedly in the same rounds
Once you have a firm’s recent deal list, the question isn’t “who else was in this round” for any single deal, it’s which other fund names recur across multiple, unrelated rounds. If the same seed fund shows up alongside your target firm in four different companies over the past year, that’s a signal worth weighting heavily. If a name shows up once, in the firm’s biggest deal, treat it as noise until you see it again elsewhere.
Separate one-off syndicates from genuine repeat partners
This is the step people skip, and it’s the one that keeps a syndicate graph honest. Build a simple table as you go, and only promote a pairing to “repeat partner” once it clears a threshold you set in advance, for example three or more shared rounds in the trailing 18 months.
| Co-investor pairing | Shared rounds observed | One-off or repeat |
|---|---|---|
| Fund A + Fund B | 4 rounds, 18 months | Repeat partner |
| Fund A + Fund C | 1 round, 1 marquee deal | One-off |
| Fund A + Fund D | 3 rounds, 12 months | Repeat partner |
This is the same reading discipline laid out in VC Fund Portfolio Analysis for Founders, which walks through seven signals you can pull from a firm’s book. Co-investor overlap is one of those seven, and it’s the one most founders read past because it doesn’t look like a thesis statement. Treat it as a first-class signal, not a footnote.
Step 2: Anchor the graph on a credible lead
Why co-investment maps are built outward from a lead, not a follower
A syndicate graph only becomes actionable once you pick a starting node. That node should be a lead, not a follower, because leads are the ones who set terms, do the heaviest diligence, and have the standing to invite others in. A follower’s co-investment history tells you who they tend to join, but it doesn’t tell you who they’d follow into your round specifically. A lead’s history tells you exactly that.
Identifying the firm most likely to price and anchor your round
The lead you anchor on needs to actually fit your stage, check size, and sector, not just have a recognizable name. How to Find a Lead Investor for Your Round sets out a seven-signal sifting system for identifying that firm, covering things like typical check size, stage concentration, and how recently they’ve priced a round in your category. Run your shortlist through that system before you build anything else, because every downstream step in this process depends on the lead being right.
Leads pull their regular co-investors, followers don’t pull anyone
Once you have a credible lead, that firm’s repeat co-investors from Step 1 become your highest-priority secondary targets. This is the mechanical heart of the method: you aren’t guessing who might join a round you don’t have yet, you’re reading who already joins this specific lead’s rounds, over and over, and assuming that pattern holds for yours too.
Step 3: Reverse-engineer syndicates from comparable rounds
Find 3 to 5 companies like yours that recently raised at your stage
Your own cap table doesn’t exist yet, so you borrow evidence from companies that look like yours: same stage, similar sector, raised in the last year or two. Three to five comparable rounds is usually enough to see a pattern without drowning in noise. Crunchbase and AngelList are both reasonable places to pull recent round data if you don’t have a paid database seat.
Map the full participant list of each round, not just the lead
Founders instinctively focus on who led each comparable round, since that’s the name in the press release headline. For this exercise, you need the full participant list, including the smaller checks and the firms that never got a mention. Those secondary names are exactly where repeat pairings hide.
Spot the recurring pairs and trios across those cap tables
| Comparable round | Lead | Other participants |
|---|---|---|
| Company 1 | Fund A | Fund B, Fund D |
| Company 2 | Fund E | Fund B, Fund D, Fund F |
| Company 3 | Fund A | Fund D, Fund G |
In this illustrative example, Fund B and Fund D appear together twice, and Fund D shows up in every round regardless of who leads. That’s the kind of pattern you’re hunting for. This extends the six-step reverse-engineering system in How to Find Investors That Led a Competitor’s Round from just the lead to the entire syndicate, so the repeat pairings surface as named, confirmed targets instead of guesses.
Step 4: Confirm the pairing is thesis-driven, not coincidental
Two funds in one deal is not a co-investment relationship
A single shared round proves almost nothing. Funds end up in the same deal for reasons that have nothing to do with a standing relationship: a founder’s own network, a hot round everyone wanted into, or simple timing. Before you add a pairing to your priority list, you need a second kind of evidence beyond frequency, you need to know the two firms’ theses actually overlap.
Checking that both firms’ revealed theses actually point at your round
How to Find a VC’s Investment Thesis describes a revealed-thesis method: reading what a fund has actually funded, not what its website claims to fund. Apply that same method to both firms in a pairing. If Fund A and Fund D keep showing up together because they’re both consistently backing seed-stage vertical software at similar check sizes, that’s a structural fit. If their only overlap is one outlier consumer deal that doesn’t match either firm’s normal pattern, the pairing is coincidental and shouldn’t be weighted the same way.
Stage, check size, and sector alignment that makes a pair repeatable
| Fit factor | Fund A | Fund D | Aligned? |
|---|---|---|---|
| Typical stage | Seed | Seed | Yes |
| Typical check size | $500K to $1.5M | $750K to $2M | Yes, overlapping range |
| Sector focus | Vertical SaaS | Vertical SaaS, fintech | Yes |
When all three line up, the repeat pairing you found in Step 1 or Step 3 is a real syndicate relationship, not a statistical accident, and it deserves a place near the top of your sequenced list.
Your co-investment map in 30 minutes (free VC-sift workflow)
The three inputs you need: a lead, 3 comparable rounds, a sector filter
You don’t need a database seat to run a first pass at this. You need three inputs: a credible lead candidate from Step 2, three comparable rounds from Step 3, and a sector filter so you’re not drowning in irrelevant deals. With those three in hand, the overlap work in Steps 1 through 4 can be done in roughly half an hour using free sources like Crunchbase, AngelList, and SEC Form D filings.
How to build the graph without a PitchBook seat
Most of the signal in this method comes from reading, not from paying for access. A spreadsheet with three columns (firm, round, co-investors present) built from free sources will surface the same repeat pairings that a $30,000-a-year database subscription would, just with more manual pulling. VCSift’s own sifting workflow is built around exactly this kind of free, portfolio-first evidence gathering, which is the same approach behind VC Fund Portfolio Analysis for Founders and How to Build a Target Investor List for Fundraising. Use those two pages alongside this one and you have the full toolkit for turning raw deal data into a usable graph.
Turning the graph into an ordered outreach plan
Once the graph exists, the ordering question answers itself: start with your anchor lead, then move to that lead’s highest-frequency repeat co-investors, then to pairings confirmed by thesis alignment, and only then to the long tail of one-off overlaps. That ordering is the subject of the next two steps.
Step 5: Convert the graph into a sequenced target list
Rank by pairing frequency, not firm prestige
A recognizable fund name that has only shown up once alongside your anchor lead should rank below a less famous fund that has shown up four times. Frequency is a stronger predictor of whether a firm will actually follow into your round than brand recognition is, and it’s easy to let prestige override the evidence if you’re not disciplined about it.
Group targets into syndicate clusters, not a flat 300-name sheet
A 300-name spreadsheet treats every firm as an independent, isolated bet. A syndicate graph treats groups of firms as a single cluster that tends to move together. Organizing your target list by cluster, rather than alphabetically or by check size, means you can see at a glance which five or six names belong to the same syndicate and should be approached as a unit.
Decide who to approach first so the rest follow
This is where the seven-layer system in How to Build a Target Investor List for Fundraising becomes the finishing step for everything you’ve built so far. The co-investment graph gives you the clusters; that system gives you the sequencing logic to turn clusters into an ordered, workable outreach plan instead of a spray.
Step 6: Let the lead pull its co-investors in
Why you pitch the anchor, not the whole syndicate at once
It’s tempting to blast every firm in a cluster the moment you’ve mapped it. Resist that. Co-investors follow leads, not the other way around, so pitching the whole syndicate before the lead has committed usually means several firms pass simply because there’s no anchor yet to react to.
Asking a committed lead for its regular co-investors: the warm path
Once a lead has verbally committed or issued a term sheet, ask directly whether they typically bring in the firms your research already identified as their repeat co-investors. This isn’t a cold ask, it’s confirming a pattern the lead already knows exists, and most leads are glad to make the introduction because it’s less work for them too.
Using recurring pairings as the reason for a forwardable intro
How to Get a Warm Introduction to a VC describes building a forwardable asset, something a connector can pass along with minimal friction. A committed lead who regularly co-invests with a specific fund is close to the ideal version of that connector: the intro email writes itself, because the relationship between the two firms already exists and doesn’t need to be explained.
Step 7: Track the syndicate as one state, not N conversations
Why a co-investment cluster rises or falls together
If your lead stalls, the co-investors you mapped around them usually stall too, because their involvement was conditional on that lead anchoring the round in the first place. Tracking each firm in a cluster as a separate, independent conversation hides this dependency and makes it easy to keep chasing dead names.
Logging which pairing you’re pursuing and who’s anchoring it
How to Track Investor Outreach During Fundraising argues for tracking state rather than activity, a conversation’s actual status rather than how many emails you’ve sent. Adapted to syndicates, that means logging a cluster as a single state tied to its anchor: “pursuing, Fund A committed, Fund B and Fund D pending Fund A’s close,” rather than three disconnected rows that don’t reflect how the group actually behaves.
Reallocating when a lead stalls and its syndicate goes cold
When an anchor lead goes cold, don’t keep individually chasing their usual co-investors, since without the anchor, the pattern that made them likely to join no longer applies. Reallocate your time to the next cluster in your sequenced list instead. This is the practical payoff of tracking by state: you see the stall immediately and move on, rather than discovering three weeks later that an entire cluster was dead the moment the lead stopped responding.
Frequently asked questions
What does it mean when VCs “co-invest together”? It means two or more funds have repeatedly participated in the same priced rounds, at a similar stage, across more than one company, which suggests a standing working relationship rather than a one-time coincidence.
How do I find which VCs co-invest with a specific firm? Pull that firm’s deal history from the last 12 to 24 months using a source like Crunchbase or PitchBook, then look for other fund names that appear across multiple, unrelated rounds rather than just once.
Can I just ask a VC who they like to co-invest with? You can ask, but treat the answer as a hypothesis, not evidence. Stated relationships are often aspirational or outdated; verify them against actual portfolio overlap before relying on them.
Do co-investors commit before or after the lead? Almost always after. Leads set terms and anchor the round, and most of their regular co-investors wait for that commitment before deciding to follow.
How many comparable rounds should I analyze to spot a syndicate pattern? Three to five recent, stage-matched rounds is usually enough to separate genuine repeat pairings from one-off overlaps, without the research taking more time than it’s worth.
Can I map VC co-investment relationships without an expensive database? Yes. Free sources like Crunchbase, AngelList, and SEC Form D filings are usually sufficient to build a reliable first-pass graph.
Should I pitch the whole syndicate at once or start with the lead? Start with the lead. Pitching co-investors before a lead has committed usually produces passes, since those firms are typically reacting to a lead’s commitment rather than deciding independently.
A syndicate graph won’t tell you who will say yes. It will tell you who tends to say yes together, which is the only version of “who do you co-invest with” that’s actually worth building a fundraising plan around.