Mistakes of falling in love
Deciding first and verifying afterwards is the most expensive angel mistake, known as the lover’s pit. It rarely feels like a decision. It feels like conviction arriving early, after which the diligence becomes a search for reasons that support it.
Skipping the work because the cheque is small is the same mistake. The small cheque carries the same total-loss risk as a large one.
So is outsourcing the work to a co-investor. Relying on someone else’s due diligence (selskapsgjennomgang) means adopting the conclusions of a person with a different position size, a different price and different information.
The published evidence is old but consistent. In the 2007 US study of 539 angels, median diligence was about 20 hours per investment. Investments where the angel spent more than 40 hours returned 7.1x, and those under 20 hours 1.1x, against a sample average of 2.6x.
The 2009 UK study associated 20 hours or more with significantly fewer total failures. Both are correlations and more than fifteen years old, but they remain the strongest published support for thoroughness. Budget forty hours for any cheque you would mind losing.
Mistakes of concentration
Most positions return less than the money put into them, and one company pays for the portfolio. A portfolio too small to contain that one company is the larger risk. Each of the three forms it takes looks like caution while you are making it, and each costs you the company that pays for the rest.
- Too few positions, so the arithmetic never gets a chance to work.
- No reserves, so the good company’s next round is funded by someone else and the position is diluted at the worst moment.
- The whole allocation spent in the first enthusiastic year, which concentrates the portfolio in one vintage and one mood.
The no-reserves case is the common one: in NorBAN’s published figures, only about a third of angels indicate capacity to follow on in a later round. If you are among the other two thirds, decide now whether to hold a reserve or accept the dilution.
Following on into the companies that need money most is the same failure from the other side. Follow-on investments in the 2007 study exited at 1.4x against 3.6x for investments never followed on, and the UK study found 1.2x. Keep reserves for the companies that are working.
Mistakes of price and terms
Paying a price the next round cannot clear costs the return, and rarely the investment itself. A valuation set at the top of the range needs everything to go right merely to stay flat, and when the market re-prices the correction is severe.
Klarna’s valuation fell from USD 45.6 billion in June 2021 to USD 6.7 billion in July 2022. Oda raised USD 151 million in December 2022 at about USD 353 million, down 61 per cent from roughly USD 900 million in April 2021, before raising again in September 2024 on terms that valued the earlier equity at nothing.
Those are late-stage examples of an early-stage discipline. Price for the stage’s risk, and test what the next round would have to pay for your position to be worth more.
The second mistake is signing without the protections because the round was competitive. Founder vesting, pre-emption on new shares, tag-along rights and reporting obligations keep a minority position worth something across years of future dilution (utvanning). A term dropped to win an allocation is one you will want in year three, when the company has changed shareholders twice.
Mistakes of involvement
Angels get involvement wrong in both directions, and the founder sees both. Too much involvement costs the company its founder’s time, and at this stage founder time is the resource the company has least of.
- Pushing the company toward the next raise ahead of the next customer.
- Treating growth as the only measure.
- Crowding a founder with advice outside your competence.
- Taking a board seat and using it to run the company. A board that cannot distinguish governance from management costs management time the company does not have.
Too little involvement is disappearing. In the 2007 study, angels in contact with a company once or twice a month recorded 3.7x, against 1.3x for those in contact once or twice a year, and investments connected to the investor’s own industry expertise returned roughly twice as much as those outside it.
Mistakes of judgment under social pressure
The last class is the hardest to see in yourself. It looks like good judgment while it is happening.
- Investing because respected names are already in.
- Moving into a theme because everyone is.
- Writing a cheque because a friend asked, and treating the friendship as diligence.
Rigorous evidence on angel-specific herding is thin. The nearest is a 2024 study of a European equity-crowdfunding platform, which found that pledges are strongly influenced by the size of and time since the most recent pledge; large investments read as public quality signals, quiet periods as negative ones.
That is an adjacent market, and the transfer to angel syndicates is an inference, but most angels will recognise the mechanism. The corrective is procedural: reach your own conclusion before you learn who else is in the round.
The corrective habits, in one table
Use the table as a checklist before each cheque; none of the correctives needs anything you do not already have, so there is no reason to skip one.
| Mistake | What it costs, and the corrective |
|---|---|
| Deciding before verifying | The diligence becomes advocacy and the risks surface after the money. Fix the sequence: screen, then diligence proportional to the cheque, then commit. |
| Skipping diligence on a small cheque | The same total-loss risk, taken without information. Set a minimum work standard per investment, independent of size. |
| Relying on a co-investor’s work | You inherit conclusions from a different position and price. Verify independently, whatever the syndicate looks like. |
| Too few positions | The portfolio cannot contain the one company that pays for it. Set a target position count and pace across vintages. |
| No reserves, or reserves spent on losers | Dilution in the winners and good money after bad in the rest. Set the reserve policy in advance and deploy on evidence of progress. |
| Paying a price the next round cannot clear | A flat or down round, and a return the company never earns back. Price against the stage’s risk and test what the next round would have to pay. |
| Dropping terms to win an allocation | No vesting, no pre-emption, no information when it matters. Treat protective terms as the price of entry, not as a negotiating chip. |
| Crowding the founder or overreaching on the board | Management time consumed by governance; the investor becomes the problem. Agree a defined role at investment, with governance separated from operations. |
| Disappearing after the transfer | The contribution that justified the position never arrives. Keep regular contact, and invest where your expertise is genuinely relevant. |
| Following the syndicate | Someone else’s judgment, applied to your money. Reach your own conclusion before you learn who else is committed. |
Which mistakes you are most exposed to changes with experience. Start with the ones that fit where you are now.
- A first-time angel makes the first three: conviction arrives early, the cheque feels too small for real work, and a respected co-investor looks like diligence. Write the first screen down before meeting the founder, set the hours standard now, and fix the target position count before the first cheque.
- An experienced angel is more exposed to price and involvement: the competitive round where a term gets dropped, and the board seat that turns into management. Write the term list before the round gets competitive, and agree your role with the founder in writing at investment.
- An angel with a sector focus is most exposed to conviction and to crowding the founder. You know the market, so the case feels verified before the hours are spent, and you have views on how to run the company. The doubled returns inside your own expertise justify the focus. They do not justify skipping the forty hours.
- An angel in a syndicate is most exposed to herding. Write your view down, with a date, before the lead’s memo or the list of names reaches you.