Why This Is Non-Negotiable
A bad investor is worse than no investor. They can block future rounds, poison your board, push you toward an exit that isn't right, or check out the moment things get hard. The good news: founders who've worked with these people will tell you the truth — you just have to ask. Most founders don't. Do the work, and you'll have information that changes your decision.
Build Your Reference List
Start with the investor's portfolio page. Most firms list their companies publicly. Your goal is to identify 8-12 founders across three categories: current portfolio companies, past exits (successful and unsuccessful), and companies that no longer exist. The failed ones are the most informative.
Current Portfolio
Companies the investor is actively involved with. These founders see the investor regularly and can speak to current behavior — responsiveness, board conduct, willingness to help operationally.
Exits & Acquihires
Founders who've been through an exit with this investor. Ask whether the investor was aligned on timing, helped or hindered the process, and acted in the founder's interest vs. purely their own return.
Failed Companies
The investor's real character shows when things go wrong. Did they support the founder? Help wind down gracefully? Or disappear, block decisions, or behave badly? Find these founders — they'll be honest.
Target ratio
Aim for at least 3 calls with founders from companies that didn't succeed or exited early. One glowing reference from a fund portfolio company they hand-picked tells you almost nothing.
Find Backchannel Contacts
Any investor will offer to connect you with founders they like. Don't just use that list. You need to go off-script and find founders they didn't introduce you to. This is where the real signal lives.
How to find uncurated contacts
- LinkedIn search: Search the fund name on LinkedIn. Filter for “Past company” to find former employees and founders. Use Apollo to find their emails directly.
- Crunchbase + PitchBook: Find portfolio companies that aren't listed on the fund's website — early bets, failed companies, and stealth deals often don't make the highlight reel.
- Mutual connections: Look at who in your network has worked with, been funded by, or co-invested alongside this investor. A cold introduction from a mutual is more likely to get a candid conversation.
- Twitter/X and forums: Search the investor's name on Twitter and in forums like Hacker News. Founders sometimes publicly share experiences — especially bad ones.
Prepare the Right Questions
Weak questions get weak answers. “Would you work with them again?” is too easy to deflect. You need questions that force real specificity and create space for people to share things that are uncomfortable to say. Here are the ones that actually work.
Questions that get real answers
On board behavior
“Can you walk me through a specific board meeting where there was real disagreement? What happened?”
On support
“What's one time you needed something from them urgently — an intro, a decision, a reference — and what happened?”
On hard times
“When things were going badly, how did they show up? Did they lean in or pull back?”
On alignment
“Did you ever feel like they were optimizing for their fund's returns rather than your company's long-term success?”
On control
“Were there decisions they pushed hard for that you later disagreed with? How did that play out?”
The critical close
“Is there anything you wish you had known before taking their money?”
Listen for what's not said
Pauses, hedging, topic changes, and overly diplomatic answers are data. “I mean, they're very commercially focused” is usually code for something specific. Ask the follow-up: “What do you mean by that?”
Run the Calls
A reference call is a skill. Most people treat it like a formality. The founders who do it well get information that actually changes their decision. Here's how to run them.
Keep it to 30 minutes
Respect their time. Ask your 4-5 most important questions, then leave space for them to add anything they think you should know. Don't over-schedule — 8 good 30-minute calls beats 15 rushed ones.
Create psychological safety
Open with: “I'm not going to share anything you say directly with the investor. I just want to make a good decision for my company.” This unlocks honesty. Most founders are relieved to be asked.
Take notes immediately
Write up the key points within 30 minutes of each call. Memory fades fast. Track patterns across multiple calls — if three different founders mention the same behavior, that's your answer.
Ask for warm introductions
At the end of each call, ask: “Is there anyone else in their portfolio you think I should talk to — especially someone who might have had a harder experience?” This is how you find the names no one volunteers.
Check the Public Record
Beyond conversations, do a proper background check on the firm and the specific partner you'd be working with. This takes two hours and can surface serious issues that founders won't mention because they don't know about them.
Public record checklist
| What to Check | Where to Look | Red Flag |
|---|---|---|
| SEC filings & enforcement | SEC EDGAR, FINRA BrokerCheck | Any enforcement action |
| Lawsuits involving the firm | PACER, state court records | Disputes with portfolio founders |
| Partner reputation | Twitter/X, HN, Reddit, Glassdoor | Pattern of founder complaints |
| Fund performance | Pitchbook, public LP disclosures | Multiple funds below market |
| LP base & fund health | SEC Form ADV (registered advisers) | Shrinking AUM, LP attrition |
Check the fund's LP situation
A fund that's struggling to raise its next fund may not be able to follow on in your next round. Ask the investor directly: “Are you actively fundraising for a new fund, and what's your reserve policy?” Their answer tells you a lot.
Synthesize and Decide
After 8+ reference calls and a public record check, you should have a clear picture. But information overload is real — here's how to cut through it and make a good decision.
Scoring framework
| Dimension | Weight | What to Assess |
|---|---|---|
| Board behavior | 30% | Do they add value? Do they micromanage? Are they constructive in conflict? |
| Alignment in hard times | 25% | Do they disappear when things go wrong, or lean in? |
| Actual helpfulness | 20% | Intros, hiring help, customer access, strategic guidance — does it materialize? |
| Fund health | 15% | Can they follow on? Are their LPs happy? Is there a new fund coming? |
| Public record & reputation | 10% | Any legal issues, public controversies, or pattern of founder complaints |
What to do if you find red flags
- One yellow flag: Address it directly with the investor. “A founder mentioned X — can you share your perspective on that situation?” How they respond is itself a data point.
- Multiple yellow flags: Negotiate protective provisions in the term sheet — information rights, consent rights on major decisions, and clear removal mechanisms.
- A clear red flag: Walk away. No amount of money is worth a bad board member. If three founders describe the same harmful pattern, believe them. Your company and mental health are not worth the check.
The Most Important Thing
The investor who is nicest during the fundraise is not necessarily the best partner when things get hard. You need to know what they're like at board meetings when you miss a quarter by 40%, when a key executive quits, or when a strategic pivot is on the table. Reference checks are your only way to see that — so do them seriously, go off-script, and trust what multiple founders tell you independently.
Tools & Resources
These are the tools I recommend for finding portfolio founders, tracking your reference process, and researching investor backgrounds.
Apollo — Find Portfolio Founders
Use Apollo to find emails for founders at portfolio companies — especially ones that aren't listed publicly. Search by company name, filter by title, and get verified contact info without relying on warm intros.
- + 275M+ contacts, company-level search
- + Find founders at specific portfolio companies
- + Verified emails — no bounces
Pipedrive — Track Your Process
Use a lightweight CRM to track which founders you've contacted, spoken with, and what they said. Pipedrive is simple and fast to set up — you can build a reference tracking pipeline in under an hour.
- + Pipeline view for tracking outreach status
- + Notes and call logs per contact
- + Free trial, easy to configure
Capsule CRM
Another lightweight option for tracking references. Great for founders who want something simpler than Salesforce but more structured than a spreadsheet.
See Capsule review →SEC EDGAR
Search any registered investment adviser by name. View their Form ADV, which includes AUM, LP count, and any disciplinary actions. Free and comprehensive.
efts.sec.gov/LATEST/search-index5 Common Mistakes Founders Make
Only calling the investor's curated references
An investor wouldn't introduce you to a founder who'll say bad things. If the only people you talk to are the ones they connected you with, you're not doing a reference check — you're reading their marketing materials.
Doing it after you've emotionally committed
Start reference checks the moment you receive a term sheet or serious interest — not after you've shaken hands on a deal. Once you've mentally accepted the money, you rationalize red flags instead of acting on them.
Asking yes/no questions
“Were they helpful?” → “Yeah, pretty helpful.” That tells you nothing. Ask for specific examples. “Can you give me a specific example of a time they helped you through something difficult?” forces real answers.
Not checking the specific partner
Firms are not monolithic. A fund might have a great reputation, but if you're getting the junior partner who barely shows up and defers everything, that's a totally different experience. Reference check the specific person who will be on your board.
Letting term sheet urgency short-circuit the process
Investors sometimes create artificial deadline pressure. “We need a decision by Friday.” A good investor will give you time to do proper diligence. If they won't, that's itself a red flag. Never let a 72-hour deadline compress a 2-week process.