I've seen a few threads recently about the power law in venture capital and what this means for individual company returns. We looked at our data on 11,350 companies backed by 259 funds from 1986 to 2018. /1
37
112
500
192,788
So we quickly looked at our data to see if DPI in year 5 is a good predictor of final performance. We analysed 71 funds from vintage years 2005-2008. The correlation between year 5 DPI and current DPI was just 0.22. /1
DPIs - it takes 10-12 years for a company to go from being founded to IPO. Don't expect much DPI in the first 5 years of an early-stage fund's life. Also, liquidity in VC comes in waves. 2021 was one of these waves. So, no surprise later vintages are lagging on DPI. /5
19
48
311
205,891
We know that there is a power law for early-stage VC funds, but what does this look like for growth funds? We've just updated our power law analysis to Q2 2024 and have included growth funds for the first time. Here's what we found. /1
16
44
308
138,779
1/ Lots of talk in VC about the power law - ie small number of companies that are responsible for a large proportion of overall value. But is this actually true or just one of those urban myths?
13
46
275
Amazing week for @peakxvpartners First Groww and now Pine Labs. Two of the biggest returns we’ve seen in our 20 years investing in India VC.
Usually, this page talks about what I do: the products, the technology, the milestones, and the news. But today is monumental. And I choose to speak as myself: Pine Labs.
Because there is someone I need to thank. Truly, openly, and from the heart: Shailendra Singh, Managing Director at @peakxvpartners.
Shailendra has the rarest gift in the game: the ability to see a championship version of me long before I could even imagine it. He didn’t just believe in my potential, he made me believe in it too.
When the pressure was at its peak and self-doubt crept in, his certainty was the playbook. He didn’t just say I was capable; he backed that belief with unwavering trust, creating a scaffold that built me up for the big stage.
He has never been a controller, always the ultimate enabler. He handed me the keys and expected me to drive, trusting my judgement and integrity completely. That trust fostered true ownership and accountability, something I value deeply.
When organisational waters turned turbulent as they sometimes do, he was the steady hand on the helm, restoring clarity and confidence. And he understood the power of community, connecting me with the right anchors, people who offered safety, wisdom, and fellowship when I needed most to keep my head in the game. And who, quite remarkably, also became the wind beneath my sails, helping me move forward with strength and purpose.
For nearly two decades, his presence has been constant, commitment absolute. Not a single board meeting missed. His time, freely and generously given, remains the most valuable resource he has shared.
And now, as I stand ready to take on the biggest challenge yet, he isn’t handing me a farewell note. He’s simply asking, “Which new shores shall we set sail for?”
Shailendra. Our journey is far from over, and, as yet another mark of the truly great coach you are, thank you for never stopping to believe in what lies ahead.
2
5
230
41,199
Your regular reminder of how, in VC, a few funds, and a few companies, drive nearly all the returns.
We looked at how the 119 funds we backed between 2010 and 2020 performed over the last 12 months. Here’s what we found 👇 /1
11
22
197
62,310
What are realistic return expectations for Limited Partners investing in venture capital funds? @MeghanKReynolds posted recently about top quartile DPI performance (only 1 vintage year since 1998 with 3x DPI). Can this really be true? /1
7
27
160
98,154
Loved this report from @PeterJ_Walker and @cartainc with some fantastic data. However, I'm seeing a lot of bad takes on how this should be interpreted. Some thoughts on the specific points. /1
New data for VC emerging managers.
From a pool of 1,800+ funds, vintage years 2017-2022.
Fresh benchmarks on IRR, TVPI, etc, alongside deployment pace, graduation rates, and much more.
Link to the full report in the following post 🙏
3
25
159
129,530
Top 10 fund returners by multiple of cost are as follows:
10. Coinbase
9. Slack
8. Portal Software
7. Facebook
6. Avanex
5. Google
4. Ariba
3. Yahoo
2. Brocade
1. DoorDash
Multiples range from 140x to over 800x
7
18
123
83,648
7/ So in percentage terms, this means that 52% of global VC exit value since 2010 came from just 1.3% of exits.
3
20
123
9/ So from an LP perspective, it makes total sense to optimise investment strategy around the VC firms that have a proven ability to consistently back these top 25 companies year in year out. This is the single most important factor we look for when assessing managers.
12
4
103
Following the interesting article from @packyM on the growth in value of the largest US VC-backed exits, we have done a deep dive into the Pitchbook data. We looked at exits >$10k from 2005 onwards and grouped into blocks of five years. Highlights below /1
3
14
105
52,259
The best-performing fund in the sample has a DPI of 12.3x. In year 5 its DPI was zero. Conversely, the best-performing fund after 5 years had a DPI of 1.28x. It's now at 2.55x. /2
3
7
85
18,167
DPIs - it takes 10-12 years for a company to go from being founded to IPO. Don't expect much DPI in the first 5 years of an early-stage fund's life. Also, liquidity in VC comes in waves. 2021 was one of these waves. So, no surprise later vintages are lagging on DPI. /5
3
17
69
136,683
We knew that fund returners were essential for successful early-stage VC funds. But they're also critical for growth funds as well. /6
4
9
69
24,824
The ongoing debate on small vs large funds is totally missing the point. LPs need to optimise on the best managers, not focus on fund size. Some data on why this is the case. /1
6
12
67
13,378
Of these 1,186 funds, just 181 (15.3%) have DPI of 2x or higher. 70 (5.9%) have returned 3x or more back to LPs and only 25 funds (2.1%) have distributed 5x to investors. /3
2
15
66
231,358
Summary - VC is hard. Average VC fund performance is disappointing. Most companies will fail. It takes a decade for the winners to really develop. This report confirms that market reality has been restored. /7
1
11
67
6,433
25 to 30 companies each year are responsible for over half of the total exit value generated by the VC industry. What does this mean for LPs investing in VC funds?👇
Tune into the Origins Podcast to find out! Thanks for having me @nchirls and @beezer232
#OriginsPodcast #OpenLP
📣The last Origins episode of the year is dropping tomorrow, and you won’t want to miss it!
VenCap’s @daveclark85 joins hosts @nchirls of @notationcapital & @sapphireprtnrs’ @beezer232 to lift the hood on venture performance.
As 3 individuals with 3 unique strategies - Nick, emerging manager; David, LP investing in established; & Beezer, LP investing in emerging & established - find out how their models differ & what David’s return data shows.
Subscribe below👇
🎧open.spotify.com/show/23H0tO…
#OpenLP #OriginsPodcast
2
9
66
34,370
UK VC-backed companies are actually punching above their weight when it comes to raising capital. In 2023 they raised over $20bn in funding. This ranks third behind the US and China. As a percentage of GDP we rank #1 amongst the 10 largest economies. /1
The lack of UK institutional participation in venture capital is very sad.
I’ve raised $500M+ over my career; 96% from US, 3% from Europe (exc UK) and just 1% from the UK.
“I have raised more money in Michigan than I have in London”.
@matthewclifford 👏
@hkanji @fdestin @cape @andreasklinger @Ljungman what can we do to change this?
1
9
63
18,823
The three rules for LPs investing in VC funds:
1. Invest with firms with a history of backing the best founders
2. Maintain a consistent investment pace
3. Don’t let your ego screw up 1 and 2.
#3 is harder than it sounds.
2
5
64
5,581
So what's the takeaway from this data? It again reinforces just how much of a power law business venture capital is. It's the top 1% of companies that ultimately drive the majority of VC returns. /9
4
9
64
19,722
And then finally, what about fund returners - companies that returned 100% or more of the committed capital of the fund that backed them. Just 121 companies (1.1%) hit this mark. /8
4
12
60
41,922
Admitedly, this is a bit rough and ready and it would be good to have more data in the sample. But early indications are that there is no positive correlation between early DPI and final performance. /3
1
2
57
5,229
I looked at the underlying Cambridge Associates data for this claim. It’s based on a tiny sample size for European VC funds - typically less than 10 for each vintage year. The data set is not representative enough to support the conclusion.
4
1
42
So what are the takeaways? We've seen the value of the top 1% exits essentially double every 5 years. If this trend continues then the top 1% exit for funds investing today would be $40bn+ (8-10 year holding period). This doesn't seem unreasonable. /6
4
9
57
8,421
Feels like the peak of inflated expectations. LPs shouldn’t be trying to market time sectors with VC fund commitments. Too blunt an investment tool. Just back the best managers who can consistently find the best founders.
1
4
52
21,306
Takeaway for LPs?
1/ If you want to build a successful VC portfolio, you need to have meaningful access to the very best companies.
2/ There are a small number of managers who can consistently back these companies at the early stage.
Miss those and you miss the returns. /7
6
7
48
6,332
We know that the last 12 months have been a challenging time for startups. But how has this affected the loss ratios for the venture funds that backed them. A short thread /1
4
11
45
7,721
3/ Since the start of 2010, there have been over 20,000 VC-backed companies that have completed an exit (IPO, M&A or buyout). These companies have generated a total exit value of just under $3.0 trillion.
2
5
43
Do emerging managers outperform? Yes, back to the age-old question for LPs. Pitchbook has published a new report that looks at the respective performance of emerging vs established managers. pitchbook.com/news/reports/q… /1
2
11
43
15,001
Some really interesting data from our Senior Associate @JontyRussell on survivorship bias of Pitchbook performance data. This is why you need to go back to the source data and understand what it really says and how reliable it is before using it to build an investment strategy.
1
5
40
6,616
I continually come back to the stat that 90% of the 3x early stage funds we’ve backed have had at least one fund returner. Wiz will return multiples of the fund for Index and Sequoia.
2
2
39
22,374
Really enjoyed this conversation with @HarryStebbings. It was definitely a full and frank exchange of opinions with two quite contrasting views of how to invest in venture capital successfully.
No one has seen more in venture than @daveclark85.
🧠 32 years investing in funds
💰 300 fund commitments
🚀 45 companies returned $1BN+
My 7 key takeaways 👇
2
1
36
13,733
Across these 10 funds, just eight companies drove the vast majority of gains:
· Wiz
· OpenAI
· RocketLab
· Figma
· Robinhood
· Revolut
· Xiaohongshu
· Circle
Without these eight companies, performance would be essentially flat. /6
2
2
36
6,859
What does this mean for investors looking to build a VC programme? It means you have to find a strategy that consistently gives you access to the best-performing managers. Looking forward to joining @Beezer232 and @nchirls to discuss how we aim to do exactly this. /8
2
35
6,084
It’s no secret that VC-backed companies are finding it increasingly challenging to raise new capital.
But just how difficult could things get for startups over the next year or two? /1
1
5
35
13,098
This analysis is based on 233 early-stage VC funds and 71 later-stage/growth VC funds backed by VenCap from 1987 to 2015. /7
8
2
33
5,680
So what about ten-baggers - companies returning 10x cost or higher. Only 614 companies achieved this milestone. That's 5.4% of all the companies in our sample. /7
1
2
34
9,790
Over 6,000 of these companies (53.2%) are at less than 1x cost. Of these, more than 2,500 (22.6%) are complete write-offs. /3
4
2
32
12,542
Loved this conversation between @loganbartlett and @m2jr. At one stage they discuss just how many VC-backed companies really matter each year and suggested it was 20-30. We have some data on this /1
Episode 112 with Mike Maples (@m2jr), founding partner @floodgatefund
- Frameworks from his recent book Pattern Breakers: Why The Best Startups Succeed
- Why start-ups need to "force a choice, not a comparison"
- Reflections from his best and worst investment decisions and how 85% of his exits $$ have come from pivots
- Predictions on the future of VC
Spotify: open.spotify.com/episode/1iB…
YouTube: piped.video/watch?v=lV313WfY…
2
3
32
8,761
There were just 1,342 companies (11.8%) that have returned at least 5x cost. That's around one in every nine companies. /6
2
3
32
13,290
Graduation Rates - our internal data tells us that 60% of early-stage VC investments will lose money (25% are total writeoffs). 2015-2021 was an aberration due to the amount of capital looking for a home. We are now seeing a reversion to the mean. This is normal. /6
1
6
32
4,790
I didn't look at this in the original analysis, but have just run the numbers. 632 funds (53.3%) have a DPI of less than 1x. I knew VC was tough, but I'm actually shocked by this statistic.
3
9
30
16,212
But what about the other end of the power law curve? Early-stage funds saw 5.5% of companies return >10x and 1.0% return 100%+ of the fund that backed them. For growth, 5.3% were > 10x and 1.6% returned the fund. This was surprising. /3
1
3
30
8,999
IRRs - these are meaningless in the first five years of a fund's life (and potentially even longer). The negative IRRs for 2021 and 2022 vintages are totally normal. This is the J-Curve. It disappeared for a while in the ZIRP era but it's now back. /2
1
3
29
3,618
8/ Bearing in mind the high number of VC-backed companies that simply go out of business, it's fair to assume that the actual concentration of value is even more extreme.
1
3
27
We are still seeing 60% of companies backed by mature early-stage VC funds return less than 1x. 27% of companies return zero. For growth funds, 43% return less than 1x and 16% are total write-offs. So as you would expect, a lower loss ratio for growth funds. /2
1
3
28
7,208
Great to get into the weeds with @Jason on how we think about venture and how we look to build a portfolio that leans into the power law.
New TWiST episode!
@Jason sits down with @daveclark85 of VenCap
- analysis of data behind the power law of VC
- portfolio strategy and finding that 1% of fund returners
- preparing with founders for when things may go south
3
25
24,496
5/ We then looked at what proportion of the exit value was generated by the top 25 exits each year. This ranged from a low of 35% in 2010 to a high of 71% in 2012.
1
2
25
For us, the key ratio continues to be (exit size x ownership) / fund size. Over 90% of the 3x early-stage funds we have backed have had at least one company that returned the entire fund. Even for growth funds, one single investment has to move the needle. /5
1
4
24
5,649
More signal doesn’t necessarily lead to an increase in funds. The increase in funds is driven mainly by strong liquidity to LPs. Even if the IPO window opens in 2025 weeks may not see a material rise in DPI until 12-18 months afterwards.
1
7
326
So ultimately, I would expect final DPIs for the 2000-2015 vintages to end up midway between current DPI and current TVPI. This would mean ~10% of funds exceeding 3x net performance and median performance being well below 2x. /7
2
3
24
14,776
Another 2,157 companies (19%) were marginally successful in that they returned somewhere between 1x and 2x cost. /4
1
2
23
11,189
IRRs cont. - the fact that median IRRS for 2021/22 vintages are lower than earlier vintages at the same time is again irrelevant. It just means the market has normalized and the days of easy money and instant writeups are over. This is a positive. /3
1
3
22
2,731
2/ Using data from Pitchbook on the number and value of VC exits globally we've been able to quantify the impact of this power law. Hat tip to @JontyRussell for crunching the data.
1
1
22
6/ In aggregate since the start of 2010, the top 25 exits each year (275 companies in total) accounted for over $1.5 trillion of exit value.
1
5
22
Also interesting to see there were 3.9x more sub-$100m funds raised than the rest combined. So when people say small funds comprise 60% of the top 10 performing funds, this means they are actually underperforming. They should represent 80%.
1
20
1,009
1,813 companies (16%) were moderately successful and returned more than 2x cost, but less than 5x. /5
1
1
21
10,428
More evidence that the VC market is rationalising. If you’ve backed great companies then you’ll do well. If you haven’t then you don’t deserve to survive. LPs need to look past the narrative as every VC has a compelling story.
Early takes from Upfront:
- Mass extinction of emerging managers is coming (LPs underwriting survivability)
- Many perceived to be great firms are not so quietly being recognized as on the decline
- Folks are tired of being “soft” and want to focus on performance>politeness
3
17
3,859
So why did growth funds have a higher proportion of fund returners than early-stage funds? At the minute, nothing jumps out of the data. Growth funds are generally more concentrated and the best companies keep surprising on the upside. But more analysis needed. /4
4
3
21
6,309
No surprise to see some of the usual suspects heading up the Figma cap table. @IndexVentures @GreylockVC @kleinerperkins and @sequoia
2
20
3,998
When looking at funds that delivered a 3x net TVPI, we found that 91% of early-stage funds had at least one fund returner. When it came to 3x growth funds, 81% of them had at least one fund returner. /5
1
3
20
5,897
So while the size of a top 1% fund has not even doubled over the last 20 years ($1.2bn to $2.1bn), the value of a top 1% exit is up over 7x. But not all of these large funds will perform equally well. As always, manager selection is critical. /4
1
20
1,590
These are all funds that we committed to at VenCap and were raised by some of the most successful VC firms in the industry. The data includes both realised and unrealised investments. /2
4
1
17
12,444
I’ve had the pleasure of knowing and co-investing with @beezer232 & chatting with @nchirls offline - so this on-the-record VC performance convo was particularly fun. Thanks so much for having me on!
Catch the full Origins Podcast episode here: open.spotify.com/episode/3G2…
▶️ The last Origins episode of the year is out now! Don’t miss @daveclark85 diving deep on 35+ yrs of established VC fund data to illustrate how venture is a game of power laws.
🎧Spotify: open.spotify.com/episode/3G2…
🎧Apple: podcasts.apple.com/am/podcas…
Big 🙏 to co-hosts @beezer232 of @sapphireprtnrs & @nchirls of @notationcapital for another year of hosting incredible guests for insightful conversation! 👏👏
Thanks to all for tuning in & sharing!
#OpenLP #OriginsPodcast
1
1
19
2,431
But do some background on the LP first and tailor your email to make it relevant to them. Tell them why they should spend any time on this vs the 49 other cold emails they get each week.
2
1
17
1,635
4/ Exit value is calculated as the post-money IPO value plus announced transaction value for M&As and buyouts. Obviously some M&A deals don't announce valuations so we can't include these in the total exit value.
2
1
17
While I don't think 2022 IRRs provide good evidence for @Rick_Zullo's argument, I do agree with him that we will be seeing a greater concentration of returns and a wider dispersion between the best VC funds and the rest. Some thoughts... /1
In 2014, the delta between top quartile and top 5% for VC performance was an additional 16% IRR, a 66% improvement
For the 2022 vintage, going from top quartile to top 5% represents an additional 48% IRR, a 1056% improvement
Being the best has NEVER mattered more
1
1
16
3,452
Really enjoyed this conversation with @DWeisburd getting into the weeds of investing in VC funds.
In this episode of the How I Invest Podcast, I interview @daveclark85 , CIO of @Vencap, to discuss the venture capital landscape. We discuss assumptions about small vs. large venture funds, unpack survivorship bias in performance data, and explore the power law dynamics in early-stage and growth funds. David Clark shares Vencap's strategy for identifying top-performing managers, insights on fund size limits, and the implications of political and economic shifts on venture capital. A must-listen for investors seeking a data-driven perspective on navigating the venture ecosystem.
We’d like to thank @reedsmithllp for sponsoring this episode!
#VentureCapital #VC #Startups #OpenLP
Link to Podcast in Comments Below 👇
1
3
16
2,799
So what's the takeaway here? LPs need to optimise for quality, not for size. The best managers can consistently deliver outperformance at scale, with greater predictability and lower volatility. /7
2
1
15
4,697
The challenge for the UK VC industry is not attracting capital, it's scaling and exiting companies. From 2005 onwards UK companies accounted for 5.2% of capital raised globally. But just 3.2% of exit value. And 2.8% of exit value from $bn companies. /2
2
12
751
TVPIs - similar to IRRs above. The J-Curve has returned and the market has normalized. This means TVPIs will take longer to develop. Don't be surprised to see 2017 TVPIs fall over the next few quarters as more writedowns/writeoffs happen. /4
1
3
14
2,891
LPs coming into VC from a background in PE often assume what works in PE will work in VC. But PE has a normal distribution of returns, whereas VC has a power law.
1
1
14
If the UK wants to attract more institutional capital, then it needs to deliver more world-class outcomes. I would agree that many UK pension funds are missing out on the strong returns VC can generate. But they need to invest in the best funds globally to capture these. /4
2
14
785
Overall, these funds are doing well – 3.1x TVPI and 20% IRR. Just three are below 1x overall. And given some are just five years old, we think there’s upside to these figures over the next few years. But let’s dive into what happened over the last 12 months. /2
1
2
14
3,334
We’re making very similar points but @fredwilson definitely puts it better
1
1
14
8,765
Is this just sheer luck or do a firm’s advantages compound with every successful deal they are involved in?
We believe it’s the latter, although there are still ways great firms can falter (usually by growing too big or screwing up succession).
Index Ventures doesn’t get enough credit. Largest shareholder in Wiz, second largest in Scale AI, largest in Figma, all are generational >$15b exits within 4 months. Top 3 VC firm globally, and arguably #1 IMO due to the consistency in fund size and returns they put out
2
13
8,609
Our Senior Associate @JontyRussell has just updated the data to June 2023 and there's no surprise to see that over the last 12 months, loss rates for the post-2011 vintages are on the rise /5
2
4
13
8,603
2020-2024 - median was $113m and 90th percentile $1.6bn (greater than the top 1% from 2005-09). The 99th percentile again almost doubled to $10.2bn with 22 companies making the cut. /5
1
1
13
2,538
After a tough couple of years post 2021, these 119 funds were up 11.5% in aggregate over the last 12 months. But this aggregate figure hides a wide dispersion in individual fund performance. The power law is starting to re-assert itself. /3
1
2
13
3,315
A correction in AI feels inevitable. But nobody knows when it might come or how much value will be created before it happens.
For LPs, that’s the challenge: wait for the reset or stay exposed as value compounds.
Either way, trying to time the market is a fool's errand.
🧵👇
1
2
13
1,087
Venture capital has always been about finding the top 1% of startups. We believe that over the next couple of years, the gap between the top 1% and the rest of the market will become increasingly wide. /13
1
12
1,555
Pitchbook has DPI data on 1,186 VC funds raised from 2000 to 2015 (so at least 8 years old). What does this tell us about what LPs should expect from their venture investments? /2
1
12
7,225
This scarcity of exits is shown very clearly when you compare the DPI of UK funds (British Business Bank data) with the DPI of US funds (Cambridge Associates data). /3
2
11
1,047
Today, we published our thoughts on the UK government's #mansionhouse reforms. These aim to unlock additional capital for UK startups and improve returns for UK pension schemes. While we applaud the aims of these reforms, we worry they are focused on the wrong areas. /1
1
9
1,338
It should be a massive red flag for LPs if any of their GPs are still holding companies at 2020/2021 last-round values. But the pressure to show paper gains to help fundraising is real. /8
1
1
11
1,413
Pitchbook has TVPI data on 1,274 funds raised from 2000 to 2015. 416 funds (32.7%) are 2x or better, 208 funds (16.3%) are 3x or better and 66 funds (5.2%) are 5x or better. /5
1
1
11
7,323
Interesting to note the quartile distribution of first time funds. Evenly distributed. I’m struggling to see the evidence that they are outperforming. bfi.uchicago.edu/wp-content/…
Fund 1's, even accounting for survivorship bias, often have produced the highest top-end cash on cash return multiples in history (Initialized and Lowercase are two extreme examples).
At this point, managers are most closely aligned with the business model / thesis that inclines most closely to where they truly have assymetric edges.
Over time, externalities often push managers out of their respective strike zones (poor advice from other LPs, GPs, legal counsel, etc.) far too early. Similar to a company, drift (fund terms, fund size, product focus) can be deadly to returns and viability (A LP friend of mine mentioned only about 1 in 5 Fund I managers make it to Fund IV).
2
10
2,696
A challenge for conference organisers. Can you create an area that allows introverts to talk to other introverts about a topic they both find interesting. This would elevate the conference experience for so many people.
4
9
1,198
There are a number of fund returners that are 10-20x cost. But it's total amount invested by the fund, not just first cheque. The VCs did a good job of layering in capital over time and took a concentrated bet.
Median fund returner is just under 40x and mean is 80x.
2
9
3,059
Now, many of these VC funds will still have a number of unrealised companies in their portfolios and so final DPI is likely to be somewhat higher than the current level. But how much higher. Let's look at TVPIs. /4
1
10
7,581
50 of the 119 funds (42%) recorded negative performance over the last 12 months. The median fund return was just 2.9%. So how do we get to an aggregate return of 11.5%? That’s where the VC power law kicks in. /4
1
2
10
3,008
Do emerging managers outperform established managers? The VC orthdoxy is that they do. A report from Pitchbook suggests otherwise /1 pitchbook.com/news/reports/q…
1
9
1,324