Doubling Down On The Overpay

One thing I've seen many VCs do wiith their initial investment in a company is invest more when the valuation gets expensive. They are ownership driven, not valuation driven. So if they originally wanted to invest $4mm at a $20mm post money valuation and buy 20% of the company, they talk themselves into investing $8mm at a $40mm post money valuation so they can still buy 20% of the company.

I have never liked this approach. When the price of an initial investment goes up, I prefer to invest less, or nothing at all. Investing nothing at all is a fairly obvious approach when the price gets beyond your comfort zone. Investing less is not as obvious.

My rationale for investing less has to do with the fact that most venture investments involve multiple rounds. If you believe there will be additional opportunities to invest in the company and you really want to be involved, then you can invest less and reserve more funds to invest later in hopes that the risk reward of the investment improves. Since you will be an investor in the company, you will be shown those opportunities before or at least alongside new investors in future rounds.

Investing more when the price is too high makes no sense to me. If you are overpaying by 2x, doubling down feels like overpaying by 4x.

I think the root of this "doubling down on the overpay" issue is that many VCs manage large funds of other people's money and they really don't care so much about how much they invest in each deal. They are looking to buy large stakes in companies and hope that one or more turns into a big winner.

I try to invest as if I have a fixed amount of capital and it is my own capital (some of it is). I like to think that every investment we make takes funds away from other investments we can make, even though this is not actually true. Our firm could raise more money if we wanted it and needed it. But I think raising larger funds will ultimately lower the returns we can deliver to our investors and we have resisted doing that.

So instead of being ownership focused, I prefer to be valuation focused. And the key figure I look at is average valuation of our entire investment. We take the total amount of capital we have invested in a company and divide it by our total ownership. We like that number to be as low as possible relative to the current value of the business. I believe that is the recipe for the best returns and that is what we seek to deliver to our investors.

#VC & Technology

Comments (Archived):

  1. reece

    Can you teach this lesson to the government?

    1. fredwilson

      i don’t think sobut i do think technology can bring down the cost of governmentdramatically if we force iteverytime i check into a flight via a kiosk i think of getting mydrivers license renewed and cringe

      1. Dave Pinsen

        In REAL ID states (such as New Jersey), you need to get your driver’s license renewed in person to prevent fraud. Remember that several of the 9/11 hijackers used fake driver’s licenses to board their flights. You can renew your registration online though.

        1. kidmercury

          yes, very important to have extra security measures to protect against another 9/11. the hijackers, who were trained at US military bases, never would have gotten through without fraudulent licenses. those who disagree are probably terrorists.

        2. ErikSchwartz

          I actually believe the hijacker’s driver licenses were real and in their true names.

          1. Dave Pinsen

            How does that contradict my initial comment that “several of the 9/11 hijackers used fake driver’s licenses to board their flights”?

          2. ErikSchwartz

            Because we have no idea what ID they used to clear security because that data is not logged (still to this day). Given how easy it is to get valid ID, why bother with fake that might get you tripped up? They were flying under their own names.

          3. Dave Pinsen

            OK. Perhaps I should have said “buy airline tickets” instead of “board”. The article I linked to notes that,The hijackers used the fake IDs to obtain credit cards, enroll in flight schools and buy airplane tickets, law enforcement officials said..

        3. JLM

          What is amazing is that most of the 9-11 murderers were in the country illegally and if cross checked when they received those drivers licenses and student pilot permits would have been apprehended.It could have been done instantly or over the course of their pilot instruction.We have the right bodies of data, the right systems but we cannot make them talk to each other and we fail to act.It is a simple lack of accountability. And it is deadly.

          1. Dave Pinsen

            Not just the right bodies of data. Atta almost got flagged by ticket agent Michael Touhey:Michael Tuohey of Scarborough said he was suspicious of Atta and Abdulaziz Alomari when they rushed through the Portland International Jetport to make their flight to Boston that day.Atta’s demeanor and the pair’s first-class, one-way tickets to Los Angeles made Tuohey think twice about them.“I said to myself, ’If this guy doesn’t look like an Arab terrorist, then nothing does.’ Then I gave myself a mental slap, because in this day and age, it’s not nice to say things like this,” Tuohey told the Maine Sunday Telegram. “You’ve checked in hundreds of Arabs and Hindus and Sikhs, and you’ve never done that. I felt kind of embarrassed.”

          2. JLM

            Wow, I had no idea. It really is amazing how many near misses there were on this score.

          3. Todd

            You are presuming that their fake ID’s were obtained from a DMV and not from a group who had the means to create ID’s themselves.Hell, even when I was in college I remember some people who had an ID created by someone who had somehow obtained the machine/materials to create 100% legitimate looking cards. I’m sure they wouldn’t have passed an in-depth investigation, but this was just some college kid.We don’t have a data set (or linked sets) strong enough for a piece of plastic to prevent this sort of thing, even if we require a blood test to get your drivers license renewed. We might as well be able to renew without a 3 hour wait.(btw, in California you can renew your license online depending on the time frame)

          4. JLM

            There are only approximately 80,000 student pilot certificates in circulation right now and all require a flight physical and a Federal issuance. They require a SSAN and a drivers license to obtain one. They are issued by the Feds after receiving a valid application. It is a pretty damn good screen. I have never heard of a fake student pilot certificate.

      2. reece

        huh…. never got an email alert from Disqus on this comment (hence slow reply), but i’m all for bringing down the (cost of) government.

    2. JLM

      Not to put too fine a point on things, perhaps the core dilemma of government — not just partisan politics — is whether they have the capability to actually learn.I frankly think it is a challenge of actual life experience; raw, native intelligence and Mother Wit.I see government constantly doing things which simply do not require doing — Gitmo and KSM stick in my mind. Why?I see us constantly allowing our enemies to select the terrain upon which we fight them.I see us constantly violating what we tell others are our “core values” — W increasing the size of government and spending while preaching conservative values.Not to be able to embrace small business as the driver of job creation in America today is like debating whether to have beer or milk on your Wheaties.

  2. Dan Ramsden

    Fred, I can see where this would be an issue on situations where sums raised in future rounds and valuations both go through the roof, so keeping up is expensive for a variety of reasons. But what about more down-to-earth deals, where follow-on rounds are more typical but still at a higher valuation? Does your position described here not trigger a signaling problem? And would you not see it as such if the situation were reversed?

    1. fredwilson

      are you talking about our participation in a follow on round? in thatsituation, we will almost always do our pro-ratai wrote this post from the perspective of the initial investment

      1. Dan Ramsden

        Oh, I see, sorry, my misread. Thanks.

      2. NicolasVDB

        why would you switch to a percentage-driven approach on follow up rounds? you may say “I don’t want to let the entrepreneur down” or “I don’t want to give the wrong signal”, but you could easily justify it to everyone by your logic above: that you’re a limited-size fund, focused on early stages.On the other hand, maybe the right answer is to increase your fund. With the power of social networks, successful companies become huge much faster than before (cf groupon and zynga). If you want to keep your percentage, that may call for a larger fund – the andreessen horowitz strategy

  3. Dan Lewis

    At what point does an increasing/”too-high” valuation get bad for the entrepreneurs? When investors who have some skin in the game walk, and instead you get dumb money, what level of investment is still worth it?

    1. fredwilson

      unfortunately you will only know the answer to this question after the fact

  4. JimHirshfield

    What about negative signal of not maintaining your same position? Seems like it hasn’t been an issue for you, but curious to hear you’re thoughts.

    1. fredwilson

      i should have been more clear in my post, i will fix thati was writing about the initial investmentwe almost always do our pro-rata in the follow on rounds

      1. LIAD

        when valuations go stratospheric like zynga’s how do you decide whether to invest pro-rata to maintain ownership or just pass and take the dilution?Is following on pro-rata in zynga/twitter mega-rounds harder to decide upon than initial investments in seed rounds?

        1. fredwilson

          we are still trying to figure that outwe’ve been conservative with these opportunities and that doesn’t seemto have been the wise approach in hindsightbut the game is not over

          1. Dave W Baldwin

            You’re right via the ‘game not over’. In the long run, using conservative parameters will serve you better.

          2. LIAD

            deleted for being a dumbass

          3. NicolasVDB

            @liad you lost me there. How would “not following on” result in a “decrease of the net value of an investment”?When DST comes knocking and puts a high valuation on a company, that may dilute the previous investors, but it certainly does not reduce the value of their holding in the company@fred when I did my first start up back in the 90s I was surprised that the same VC would say “we’re thinking of putting $1M at $2M pre or $2M at $3.5M pre’; I thought, how can they be unsure of my valuation by a factor of almost 2? The answer was “you have to look at it as a deal, not as a valuation exercise, and both deals make sense”.Are you saying that it’s not how you look at it? that you are firm on the valuation, so for example in this case you’d say “your valuation is $2.5M, and I don’t know yet if we’ll put $1M or $2M”?

          4. LIAD

            could the diluation hit by not following on in a major up-round result in an investors net $ value investment in the company going down (I owned 10% at 100m, now I own 3% at 200m?)…confused.

          5. NicolasVDB

            as long as the pre money valuation of a round is higher than the post money valuation of the previous round, the net value of the early investments keep increasing, whatever the dilution may be

          6. LIAD

            sure. i get that. as long as the share price goes up your making the Q is whether or not you invest at higher rounds to maintain ownership. at some of these mega rounds, maintaining ownership can be expensive

          7. mike gilfillan

            To add to Liad’s question: and when is a company simply raising too much money?What is the incentive for the board (ie: investors) to dilute themselves and accept a new round at higher than necessary valuation?

      2. bussgang

        Fred – if you do your pro rata in follow-on rounds, isn’t that risking the equivalent of the overpay? We struggle with this as well. You still have great conviction and you want to be supportive, but as a bread and butter Series A investor you don’t want to allow your “blended average post” (our lingo for the total capital in divided by the ownership – across all rounds of financing) to run away from you. A nice position in the Series A of sub-$10m blended average post can float up to $50-100m+ in hot deals if you continue along at pro rata. With constrained fund sizes, this is something one wants to be very careful with.

        1. fredwilson

          Great point Jeff. We do take our pro rata in venturish rounds at venturishvaluationsBut once they get beyond that we are more careful for the reasons you layout

  5. Harry DeMott

    You are talking about IRR’s and not gross dollar returns. A typical manager with an incentive structure usually cares solely about gross dollar returns – which maximizes the GP share of profits. A manager that cares about thier investing returns looks to IRR – and then figures out how much in capital they can run without meaningfully diluting their IRR’s to investors. Great managers can scale and retain their IRR’s – but there is always a limit. A great company at the wrong price is not going to be a great investment – so as good as Facebook is – perhaps at $35B that is just not going to return significant IRR’s to investors – but it might just deliver large gross profits because of the ability to put large sums of capital to work.

    1. John Frankel

      Great article, Fred, and great comment Harry. It IS all about IRR if incentive structures are aligned as they are with small VC funds as the management fee will never make you rich, just help cover overhead.Every market has price insensitive buyers, which allows for a dispersion of retuns.

  6. ErikSchwartz

    This helps explain why in their entirety VC funds (as a class) have not made any money for the last 10 years.

    1. Dave Pinsen

      The scarcity of IPO exits, due to the secular bear market in stocks over the same time period, helps explain it too.

  7. ShanaC

    Beyond investing in the valation, how do you all invest against the local inflationary effects (IE if enough people are paying 4x on a regular basis for an extended period of time, maybe 4x is the new 2x because of TVoM)

  8. JC

    Does that mean that your initial investments are taking into account future rounds? i.e. are you more inclined to decrease initial valuations, keeping in mind your final average valuation?

  9. Chris Rechtsteiner

    This post is like a David Byrne / Talking Heads riff … stop making sense!

  10. Venugopal Sathya

    Hi Fred,Do you think that you would still like to follow-on/ invest pro-rata in Twitter in the current fund raising that they are supposedly doing at such stratospheric valuations as $4B if the media reports are true …This is just an example however does that really makes sense in such scenarios is my query ..bcos you would have initially invested at lower valuations for a better % and in these/ such cases just to maintain your % you may have to shell out significant amount of money ..Whats your take on such cases..Also when you talk about crazy valuations these days are you talking about only for your initial investments or are you also alluding to follow-on investments for your existing portfolio cos too ?

  11. Mo Koyfman

    The money-shot here is a great point and very hard to argue: “If you are overpaying by 2x, doubling down feels like overpaying by 4x.”Throwing lots of money at deals when you’re overpaying is a bad strategy. Full stop.My fundamental point here, and it’s a simple one but I think nonetheless instructive, is that it all comes down to how much you think the company is worth.Oftentimes what looks like an overpay ends up looking very cheap down the road: Facebook, Twitter, Zynga, Groupon all had second or third round deals done at prices that seemed absurdly high to the broader market at the time but look ridiculously cheap in hindsight.This is even more true in instances where you are already involved in the company and can appreciate it’s current and future value potential *and risk profile* better than anyone from the outside.In those cases, so long as investors have tremendous conviction in those businesses and the belief that they can be multi-billion dollar assets with an acceptable risk profile for the price of their investment, can you give a company some additional future credit for value creation not yet realized? Or said another way: is it possible that the valuation actually *is* higher than you think because you have much more confidence/certainty in the future outcomes?From a financial/mathematical perspective: isn’t it all about your risk assessment or the beta of that underlying investment? (yes, i went back to Finance 101 to think through this :)If you think the risk is lower than others, you’d pay more…or vice versa.We’re operating in a venture world here where beta or risk is a very very speculative measure. There’s no easy chart to look up that tells you what it is. So doesn’t it simply come down to judgement in that case?And finally on the ownership point, if you have that level of conviction in the deal isn’t ownership important like in any other deal? So why not invest more in the company?I know i’m carving out a very specific type of investment here that is not illustrative of much of what we’ve seen in the market and what you’re referring to in this post. But this is the scenario that I was referring to where I think you can “overpay” for a deal. Because if you actually have the math right, you’re really not overpaying at all.(PS – forgot to mention – the link between all the businesses that I believe deserve those higher valuations are that they are category leaders with real network effects. If you are a category leader with clear and apparent network effects, your value can be significantly higher than it appears given the competitive advantage you have in the market and the therefore inherently lower risk profile of the business.)

    1. fredwilson

      you make some great points Mo and it is always a struggle to determine if a business is one that is worth overpaying fori also think the Facebook, Twitter, Zynga, Groupon examples are a big part of why the early stage venture market has gotten out of hand. it’s like every company is going to be worth billions in a few years.and we all know that is not going to happen.

      1. Mo Koyfman

        If it’s worth it, you’re not overpaying.The hard part is figuring that out.And I think this debate applies to any round where you’re buying a meaningful (10% or more) incremental position in the company, first round or later round.Follow-ons are a different discussion.

        1. David Semeria

          Some stories just keep getting better.But as you pointed out, the real knack is seeing the overshoot ante post.

      2. JLM

        There is a reason why the basis of finance is the TIME value of money. Because the passage of time both rewards and punishes the trend. The most powerful forces ever unleashed on the earth are gravity and compound interest.

    2. ShanaC

      Do you think that network effect is disruptable – and hence one big players can be forced into marginality?I keep thinking DIGG when I read this – it had huge network effects when it started, and now it seems to be getting copied and taken over in particular ways (hackernews) and in more general ways (reddit) And while I realize with the reddit example they both sort of co-evolved, there seems to be some sort of open ended discussion to be had about “leadership roles” and disruptability.I’m not sure if any of these players can make the needle move in five years. As individuals we tend to mis-guess are future by quite a lot :)Though really, this is an excellent comment

      1. Mo Koyfman

        i think it’s possible, but large networks of scale are usually disrupted bya new paradigm.meaning: search (google) is disrupted by social (facebook).i think this is true in the digg case as well, as digg is being disrupted bynew and more useful ways of sharing and ranking data.

  12. William Mougayar

    Your approach makes a lot of sense and it takes discipline to adhere to it.Is this what happened in Twitter’s last round when you sat out that one.I wonder if u can provide more real examples

  13. MaxBranzburg

    Booming fund size is certainly one of the factors causing some VCs to invest negligently. I’ve written about that here:…But small VC funds should be just as focused on ownership as big VCs, shouldn’t they? They just face a lower valuation threshold, above which they prefer not to invest. That threshold keeps them honest and demands that they focus on capital efficient early stage deals. Shouldn’t smaller funds – more so than big funds – invest less early and reserve more for follow on rounds, even if that means doubling down at a higher valuation (as the company grows and ultimately nears an exit)?

  14. kidmercury

    IMHO new pricing mechanisms are needed; broken accounting protocols and faulty monetary policy have made valuation impossible, as always. i.e. company X may be worth 4 million USD, but it is more meaningful to say it is worth 45,000 barrels of oil, or some basket of commodities. valuation needs to go up against commodities so that purchasing power of shareholders increases.

    1. RichardF

      or maybe the yuan ;)…great to see you back Kid

      1. kidmercury

        or maybe a basket of currencies….a new currency….a virtual currency….a new world currency for a new world order!

        1. Donna Brewington White


    2. Andrew Greene

      Are you serious?

      1. kidmercury

        yes, i am sincerely communicating my ideas in that previous comment you replied to.

  15. msuster

    Great topic. It’s a difficult decision process, for sure. I’ve erred on the side of arguing for no involvement rather than less with one exception. We’ve looked at some expensive deals that as a partnership we liked. We were to pay a large price and from best I could tell would have little influence on the board or with management. In these instances I’ve “laid on the tracks” and said no. My argument – then we’re just stock pickers and might as well trade a more liquid asset.There have been some cases where we did what you said, invest less and hope to see another round where we’ll have better information.The toughest call is knowing when to “pay up.” Everybody now likes to use the examples of Facebook, Twitter & Zynga. The biggest problem is that post hoc you know which ones to pay up on but a priori it’s never completely obvious.I love the topic because it has been one of the biggest topics of debate at our partners’ meetings and one of the few things that entrepreneurs really get no visibility into. It’s also one where I’m not yet 100% sure where my compass is.

    1. M. Edward (Ed) Borasky

      Fred, Mark – there’s probably technology that could help you model the risk / return characteristics of startups relative to other investments “suited for accredited investors”. But I don’t see the computational finance guys working in that area – they either went into hiding after the mortgage debacle or are focusing their energies on wringing the last nanosecond of latency out of the existing automated trading infrastructure. ;-)The closest I’ve seen to the sort of thing that could be possible so far is Liquid Scenarios (http://www.liquidscenarios…. but their stuff is highly proprietary so I have no idea what’s “under the hood.”

    2. JLM

      I very much like the comment about becoming stock pickers as it demonstrates a keen understanding and appreciation for how different VC investing is as compared to buying common stock in a public company. Fair play to you.I would however note — a parallel observation not one in contention — that part of both “stock picking” and skillful VC investing is assessing the jockey not just the horse.I know from personal experience I have done a whole lot better picking jockeys than horses in both arenas.Picking jockeys is a binary exercise — 1 or 0 — and I have never, ever regretted passing on a deal when the jockey has gotten a 0. I have never had a 0 make it to the pay window.I have had a few “1”s make it even though they were riding a fairly lame nag.

      1. fredwilson

        you can swap the jockeyit’s painful and full of riskbut it does work sometimes

    3. paramendra

      The word “visibility” is key.

    4. Dave W Baldwin

      I would say that you have to like the jockey, yet you are probably smart to have some position of influence. All sides should welcome that, for the set up should encourage interaction. Yes there are tough decisions to be made, yet it is better to have the best experience at the table.

    5. ShanaC

      There are no ways a priori to center your compass? Like fits the thesis, fits a whole slew of other targets?

  16. Jules Maltz

    I agree with the thinking for initial investments. Follow-on investments are trickier. If you are going to reserve 30-40% of a fund for follow-on investments, you’ll do a lot better following the “best” investments in each fund. These will often (but not always) be the deals that have large valuation step ups in the next round. I’d rather follow-on the very best deals (even if the valuation is high) than investments that have not performed as well. I think having discipline on follow-ons for the investments that don’t perform is the hardest part of the job.It will be interesting to see if Super Angels start to get more aggressive in following on for their winners (even if the valuation is substantially higher than their cost).

    1. fredwilson

      that’s what we have done. most of our follow on capital has gone intoour winners

  17. bijan

    it’s funny because i paid up once. it felt great at the time and it feels pretty good now.but it’s interesting that I haven’t done it again.

    1. Jules Maltz

      Let me know when you find another one that feels good 🙂

    2. fredwilson

      you didn’t pay upyou invested in the mess:)))

    1. fredwilson

      i can’t comment on that transaction

  18. Jh

    Fred, usually agree with your thoughts, but can’t agree with this post. Just because a round is priced at a higher valuation, doesn’t make it a bad investment. Sure, the highest multiple (though not always IRR) will be the first round, but if you like a company, know it well and think the valuation makes sense, all the more reason to participate even if it blends down your returns and multiples. I suppose if you are capital constrained (which it doesn’t sound like you are), you might optimize but I assume those earlier rounds have higher failure rates than the later rounds. Some of your high profile portfolio company late stage rounds seem to have very low downside with possible get lucky upside — why not make that bet? In fact, one could argue the opposite point that VC’s should press their bets on their winners and make fewer bets…

    1. fredwilson

      hi Jeffi think a lot of people misinterpreted this postit wasn’t an argument against paying up for follow on roundsit was an argument against paying up for the initial roundswhich is what we are being asked to do these days and what we areresisting doing

  19. paramendra

    There is something to be said of open talk.

  20. Rtan001

    This type of investment strategy is exactly what’s happening with Twitter.

  21. Brooke

    Fred, do you think it is easier to manage your follow-on valuations in NYC compared to Silicon Valley given that competition for deals there can be so much greater, or is this irrelevant? Can you keep funds’ entrepreneurs loyal to you guys so that they don’t chase higher valuations from other investors who want in?

    1. fredwilson

      i think the answer to both questions is no

  22. Timo

    Great post. This is what we do at for our portfolio. I believe that the “fat VC model” is dead.

  23. Steven Kane

    Amen brother Fred