Paul Graham Shares Number of Y Combinator Companies Sold or Valued at Over $40M | Hacker News Reader

Paul Graham Shares Number of Y Combinator Companies Sold or Valued at Over $40M

mattermark.com

143 pointsdmor87 comments

This numbers don't give away too much because of subjective nature of "valued at" as well not actually knowing the mean or the variance, let alone the nature of distribution.

However what is impressive is that if you were going to do startup with YC, your chance of becoming significant company is about 10%. Due to nature of power law in these things, I tend to think, may be about 20% of other startups managed to become less spectacular but still successful and other 40% may be just sustainable to make a reasonable living(lifestyle business) and rest had to fold.

"be just sustainable to make a reasonable living(lifestyle business)"

Pure speculation and just throwing this out there to see if others agree but somehow I don't feel that YC is spinning off lifestyle businesses. Not to mention that it's hard to believe that some of the founders (and supporting people) that get involved in and strive to be in YC will stay put at what ends up amounting to a lifestyle business.

I'm not sure I follow. OP's point is that a failed startup might still be profitable enough to be a lifestyle business.

That opens up some pretty cool options for the founders.

"I'm not sure I follow. OP's point is that a failed startup might still be profitable enough to be a lifestyle business."

But who will operate the lifestyle business? The founders and the team?

My point is that the type of person/team that gets involved in wanting and getting into YC (both the founders and the team) don't seem to me to be the type of people that would want to continue in "merely" a lifestyle business. Not that some wouldn't but my speculation is that the majority would not.

From your profile I see that you do LSAT prep.

My guess is that the majority of people who put in the time to get into law school are generally shooting to be lawyers.

If they end up graduating and not getting employment in law (or not being able to sustain that) they might very well end up having to do another job. (Actually they will or they will starve.).

But now we have a case of people who pursue the startup lifestyle presumably for the big win.[1] And now after all the work "all" it turns out to be is "lifestyle" I would think a high percentage of them would then try to get involved in another startup (whether their own or someone elses) not stay and have "only" a lifestyle business.

Otoh if enough time goes on before it becomes apparent that it is only going to be a lifestyle business it's possible that members of the team might have priority changes (marriage and kids) and decide that the lifestyle business is pretty attractive.

[1] Edit: Or to change the world.

You can sell the business. Or you can put work into automating it and hire someone to manage it.

The point was that it's not "Be Airbnb or have nothing to show for it". You can join YC, aim for the moon, but maybe end up with something pretty decent for your efforts even if you fail at the big one.

The startup mantra is Go big or Go home. You want alignment with your investors. To even tell them you're tolerable of a lifestyle business would not be alignment with the use of their funds.

Don't confuse temporary state of 'sustainable' with 'going to continue to maintain a lifestyle business'. Many startups are happy to be at a sustainable state while figuring out product / market fit, how to scale, etc - because it gives them a longer runway without having to raise funds.

It'd still be considered a 'failure' from the investors perspective, and again, if aligned correctly - the founders also. One of my favorite quotes from an investor is 'Don't return my money. Return me 10x or fail ($0 return) trying'

It think the "battle" is over owning the "startup" label. Can a lifestyle business be a startup? I personally don't care much, but if you're taking investment, you can't aim for a lifestyle business (as you explain well).

We're not intentionally spinning off lifestyle businesses, but that is a common outcome for seed investments. For investors it's a failure, but for the founders it could be good.

Also depends on how much investment they got surely.

Loopt apparently sold for $43m, but had $32m investment. I don't know if that's classed as 'success' or not.

That's easy, no.

The more interesting part of his question was the objective criteria for determining success, not opaque opinions about one example that fits into the gray area.

Imagine the success-failure spectrum.

On one end, there are the companies that fail; for every dollar put in, exactly $0 came out in the end.

The other ("success") end is more difficult to define because there is no end point; the company that is furtherest along that end of the spectrum might not hold that position forever if an even more successful company comes along.

I think the natural tipping point between failure and success is the break-even point, perhaps adjusted for inflation. A company that does just a bit better than breaking even for investors isn't going to overwhelm anyone, but it's the point at which most people won't feel too bad about having invested in the company. Top investors who are used to yielding massive gains will perhaps not feel the same way about that, though.

Reddit, for example, sold for like 20-30 million so isn't included.

I know some other YC businesses that seem unlikely to ever be huge but are profitable and probably will be as long as they want to be. Own Local has that feeling to me, although perhaps it's actually going to be larger someday I don't really know.

Perhaps a 3D graph would be interesting.

   - Batch year
   - Valuation
   - Years to reach valuation*

* First dinner to m&a, discounting prior traction/validation.

That way, gross valuation / year can be more easily elucidated.

Nor Wufoo, sold for 35 million.

Are we sure that is what this is saying? I was reading it as 'sold' || 'valued > 40 million', not 'sold > 40 million' || 'valued > 40 million'

Depends what you mean by subjective. The valuations are not just our internal estimates of the values of these companies. They're the post-money valuations resulting from actual investments, and at these valuations, invariably by institutional investors.

So while of course not all of these companies are going to succeed (a $40m valuation is roughly a bet that a company has a 10% chance of ending up worth a billion), the people making these investments are professionals at valuing startups, and they're betting their own future income on the accuracy of their estimates. There's always uncertainty when you have an asset appraised, especially one whose value is so hard to predict, but you can't generally get a more accurate appraisal than one by a professional who is actually willing to buy the asset at their estimate of its value.

Don't many of these rounds have diluton consequences for existing shareholders if some performance target isn't met? Is that taken into account?

No, I've never heard of a funding round with performance targets.

Err I think I meant preferred shares? FOO buys in at $X valuation. If a liquidity event happens that doesn't meet price target $Y, FOO gets paid out before the previous holders altogether or at some higher percentage.

Yes, investors with preferred stock usually get their money back first. Sometimes they get a multiple, but that's considered overreaching nowadays and the more promising startups never have to agree to that.

I suppose that is implicitly a target valuation in a sense. But no one views it as a target, because it only matters if things go badly.

Tranched rounds with performance targets didn't used to be unheard of.

Grandparent poster has since clarified his question (he was talking about liquidation preferences and multiples). But yeah, back in the day, tranched rounds with company performance targets were not unusual. In fact, they're still practiced today -- typically by big corporate investors, as opposed to angels or VCs.

My personal opinion is that performance targets specifically, if not tranched payments in general, are bad for all concerned. Ostensibly they cover some downside for the investor, but startup investing is abot maximizing upside, not covering downside per se. At their worst, tranches and targets hold the company hostage to goals that are often arbitrary, and are almost always gamed accordingly. As a result, you have a company that is checking boxes and focusing on its arbitrary numbers, and you've got investors who don't realize they're creating perverse incentives. Also, it's basically impossible to run any variant of the Lean Startup methodology when under the gun of performance-based tranches.

I say this having worked for a startup that accepted big corporate money on a tranch-based system with performance goals. Total headache for all concerned.

Note to dmor: I wouldnt put that copyright on the image. The data is shared by pg, not you. The image itself without the data has little value. Therefore, that copyright seems greedy and portrays a bad image to me. All in all, not a big deal, but someone can take a detail like that a turn it into a PR nightmere.

This is a really good point, I just copy/pasted the format for some old slides. I'll update it, definitely not the intent

To put that in perspective, most of the first non-founder engineers couldn't buy (outright) a 3br house within walking distance of Y Combinator's office with their stock option proceeds.

I think what you're saying is that if a company were sold for $40m, employee #1 wouldn't get enough to buy such a house. That would depend on how much stock he or she got and how much he or she was diluted by later investments. The cases I've seen fall on both sides of the line, but there are plenty above it.

But also, we're not talking about 42 companies with valuations of $40m. $40m is the lower bound.

In other news, women share how much their diamond rings were valued at when they got married. Unfortunately, when they go to sell their cherished tokens after their eventual divorce, they soon discover that diamonds aren't worth nearly as much as they thought they were.

Indeed, they are often horrified to learn that the valuation of the diamond at point of sale was horrendously overpriced through the use of psychological arbitrage and cultural indoctrination, that the cost of their production is surprisingly low thanks to copious amounts of slave labor and theft, and that the wholesale price that the diamonds will clear at is so low as to be an insult to their very dignity.

Oh, were we talking about startups?

I must have digressed into accurate land.

For anyone who's being a bit slow today:

    women = investors
    diamonds = startups
    marriage = first rounds
    divorce = likely failure of startup
    valuation at point of sale = optimistic valuation by investors at inception
    cost of production = ramen, servers and underpaid employees without equity
    wholesale price = true value of a startup on average

I'm glad pg also shared the percentage breakdown. If just the number of startups based on funding year had been shared, YC's progress/success would not be nearly as telling as with the percentages also shared.

To put this in perspective, any company that has raised 10M or more is likely to be valued at something near 40M. The numbers seem less startling to me after factoring that in (my first reaction was that there must be a lot more acquisitions than what we see in the news).

Sold or valued at.

I also found the equivalency funny.

pg in 2012 Sept

"The one thing we can track precisely is how well the startups in each batch do at fundraising after Demo Day. But we know that's the wrong metric. There's no correlation between the percentage of startups that raise money and the metric that does matter financially, whether that batch of startups contains a big winner or not."

(http://www.paulgraham.com/swan.html)

Almost none of these valuations are the result of investments after Demo Day. These are later rounds, and later rounds converge on an ultimate valuation.

The current graphs don't say much. How about average/median/top valued company by year funded 2 years after funding?

Say in 2005 company X got funded, then its value from 2007 should be used.

Of course 2013 and 2012 cannot be included then, but it's obvious that older companies had a much longer period to build up value and thus the current graphs don't say all that much.

YC's investment thesis is that the whole game of investing in startups boils down to investing in big stars. If you got into Google with an early stake, you won. Medians don't matter: what matters is how many big stars you caught since they'll pay for the rest many times over. So I think PG is measuring the same thing he's trying to make.

Though I wish we could claim credit for discovering that, it's really the thesis of every startup investor.

So many confuse mean with median. It's why lotteries work. The mean return might be positive. But the median is always $0.

Alright, top valued then. Or the median/average from the top 10%. Shouldn't that give okay results?

Must be nice. ;)

But on a serious note, I wonder how much that 2009-2012 bump has to do with inflation or a lack of other opportunities in the market. Maybe I am out of touch, but it seems like there are less and less places to put your money. Not to suggest that startups are a bubble. The rise is probably due to YC becoming a better filter, picking better horses in the race.

42 companies * $40M is $1.7B. Wow.

Almost half of one snapchat (with only 20 employees).

It's 40MM or more. There are some significant outliers, like Dropbox and AirBnB.

42 * $40M * 6% (average stake YC gets in company) = $100M

According to http://techcrunch.com/2013/10/25/y-combinator-13-7b-valuatio...: - YC company valuations total $13.7 Billion - 511 companies total

$11,000 + $3000 * # of founders = total investment in YC companies (let's just go with 17k for 2 founders on average) = $17K

$17000 * 511 = $8.7M

$100M - $8.7M = $91.3M

Clearly a pretty good investment =P

That's (very very roughly) 30% IRR which sounds kind of nuts to me.

account /2 for average probable dilution.

What about dilution?

YC Equity stakes are likely unable to be diluted. Not all shares can just get 'diluted' aka expanded.

Nope, YC's equity gets diluted like everyone else's. In fact, while many investors will buy more shares (at the new price) to keep their % ownership, YC does not, so their ownership % when the company exits is probably 3% or less, down from the original 6%.

We get diluted. Every investor gets diluted. They may have the right to buy more stock in later rounds to maintain their percentage, but they do have to buy it, at the price of that round.

Interesting

I wonder whether stake is correlated (perhaps inversely) with valuation.

Doesn't the 6% get diluted once additional money is invested?

yes, YC doesn't do pro rata. so basically they're down to probably at least 2-3%, but probably more.

Are you saying that YC only invests 11K per company plus 3K per founder?

Currently it doesn't seem to be that low ( "we invest a small amount of money ($14-20k + an $80k note") from http://ycombinator.com/ , but yeah, it's not a huge amount of cash. It is very early stage startups though, remember.

That's what YC says:

How much do you invest?

Usually $11,000 + $3000 per founder. So $17,000 for two founders, $20,000 for three or more. We've also arranged for each startup to get $80k in convertible notes automatically. The goal is usually to give you enough money to build an impressive prototype or version 1, which you can then use to get further funding.

http://ycombinator.com/faq.html

$40m is the threshold, but this is a domain with a pretty steep power law distribution, so their total value is a good deal more than that.

I would be curious to find out why he chose $40M as the benchmark.

There's a sort of threshold around there. A startup can raise at $20m after Demo Day by looking really promising, but to raise at $40m you usually have to be launched and growing consistently.

So presumably, companies that are on a ~$3M revenue run rate are a safer bet. Assuming a multiple of 15.

Is there a similar breakdown of companies YC lost money on?

I'm sure they could, but it's not really relevant. It's more about the 2-3 Dropbox, AirBnB, Heroku like companies.

Think of it this way with Heroku:

YC gets 6-7% initially and probably ends up with close to 3% after Series A and B. That's $6.2M on the $212M exit price.

A single exit like Heroku returns them enough to run YC for at least a couple years and returned >10x for other investors.

So while you could show that, it doesn't really matter to YC or to investors, because they have those 2-3 huge wins.

Interesting: to a close approximation, the odds of a positive payoff from YC (vs opportunity cost) is about the same as buying a California scratch ticket--about one in ten.

Does anyone know how many are actually sold vs valuations?

You should be able to get a rough idea from http://yclist.com/

7 of the 42 were acquisitions.

Thanks

Adjusting for time (takes a while to build the company), it seems like about 15-20% of YC's startups will eventually hit that number.

So how does it compare to companies borne out of Y Combinator? That would be interesting to see.

I agree. That would be very interesting - how much better is YC relative to the overall data set ?

Techstars publishes a fair amount of data too: http://www.techstars.com/companies/stats/

Then there's general studies of VC-backed companies. Not much available on pre-VC companies though.

Wow, that and the follow-up tweet are impressive. What happened in 2013 though?

Younger companies probably haven't had time to be bought yet! Or to be valued as highly.

Ah, of course. Thanks for clearing my brain fog.

2013 companies haven't had enough time yet

I dont think that is true, YC seems to be progressively accepting later and later companies. the 2005 batch was a bunch of kids in college. 2013 is seasoned engineers working on it for a year already with massive traction.

I think what's expanding is the standard deviation/variance. The mode is not moving all that much. Mean/median, a bit. There is definitely a trend toward some later companies (it used to be ~none), but there are still a lot of early stage.

Also, there is little correlation between how far along a company is when we fund them and how well they end up doing.

I am sure this is true, but very specifically getting to just the level of above 40m, it can't hurt, can it? I am not saying if you fund them later they will become dropbox, just have a higher likelihood of hitting 40-50m

Dropbox is an ironic example, because Drew applied right before the deadline with a video, a string-and-wires prototype, and no cofounder. He'd been working on it 3 months and had just barely quit his day job.

This is the first batch where I knew people in YC. They started their company (with an entirely different idea) on Feb. 1. The founders were all between 24 and 26 at founding, all 4, 2010 or 2011 graduates. Reddit was started after Alex and Steve graduated from UVA. DropBox (S07) was Drew's 4th startup.

Typical LISP guy: delimiting the pairs with whitespace - good enough :)

Somebody, fix "OFY" in graph header.

Anyone know which two are the 2013 companies?

Names?

Reply on news.ycombinator.com