Fixing Venture Capital
Copied From Joelonsoftware.com
Many software companies these days are built using some form of venture capital. But the VC industry has been hurting lately. A lot of investments in dotcoms turned out to be spectacular flameouts. As a result, VCs are becoming ever more selective about where to put their money. To get funded these days, it’s not enough to be a pet shop on the web. Nope! You have to be a pet shop on the web with 802.11b wireless hotspots, or your business plan is going right in the dumpster.
The formerly secretive world of VC has become a bit more transparent, of late. VCs like Joi Ito, Andrew Anker, David Hornik, and Naval Ravikant have created weblogs which are a great source of insight into their thought process. That dotcom thing resulted in three great books by company founders that look deep inside the process of early stage financing (see footnote). But as I read this stuff, as a founder of a company, I can’t help but think that there’s something wrong with the VC model as it exists today. Almost every page of these books makes me say, “yep, that’s why Fog Creek doesn’t want venture capital.” There are certain fundamental assumptions about doing business in the VC world that make venture capital a bad fit with entrepreneurship. And since it’s the entrepreneurs who create the businesses that the VCs fund, this is a major problem. Here’s my perspective on that, from a company founder’s point of view.
When people ask me if they should seek venture capital for their software startups, I usually say no. At Fog Creek Software, we have never looked for venture capital. Here’s why.
The fundamental reason is that VCs do not have goals that are aligned with the goals of the company founders. This creates a built-in source of stress in the relationship. Specifically, founders would prefer reasonable success with high probability, while VCs are looking for fantastic hit-it-out-of-the-ballpark success with low probability. A VC fund will invest in a lot of startups. They expect about seven of them to fail, two of them to trudge along, and one of them to be The Next Netscape (“TNN”). It’s OK if seven fail, because the terms of the deal will be structured so that TNN makes them enough money to make up for all the losers.
Although the real spreadsheets are many megabytes long and quite detailed, this is the VC’s calculation:
|A||Probability of Success||10%|
|B||How rich I would get||$1,000,000,000|
|C||Expected Return (A x B)||$100,000,000|
But founders are much more conservative than that. They are not going to start ten companies in their lifetime, they’re going to start, maybe, two. A founder might prefer the following model:
|A||Probability of Success||80%|
|B||How rich I would get||$100,000,000|
|C||Expected Return (A x B)||$80,000,000|
Even though the second model has a lower expected return, it is vastly preferable to most founders, who can’t diversify away the risk, while VCs who invest in dozens of businesses would prefer the first model because it has a greater return. This is just Econ 101; it’s the same reason you buy car insurance and Hertz doesn’t.
The difference in goals means that VCs are always going to want their companies to do risky things. Oh, sure, they’ll deny it, but if they were really looking to do conservative risk-free things, they’d be investing in U.S. Treasuries, not optical networking companies. But as an entrepreneur, you’re going to be forced at gunpoint to bet on three cherries again and again and again. You know you’re going to lose, but the gunman doesn’t care, he’s got bets on all the slot machines and one of them is going to pay off big time.
There’s nothing controversial here. A VC would say, “that’s what VC is for: investing in risky ideas.” Fair enough. As long as the entrepreneur wants to take a 10% chance, VC may be the way to go. The trouble here is that the VC is now doing a perverse kind of selection. They are looking for the founders with business ideas where the founders themselves think the idea probably won’t work. The end result is that VC money ends up being used in bet-the-farm kind of ways. This kind of recklessness causes companies like WebVan to blow $800,000,000 in a rather desperate attempt to buy a profitable business model. The trouble is that they were going so fast that they didn’t have time to learn how to spend money in a way that has a positive return, which is, by definition, what you have to do to be profitable.
Here’s my philosophy of company growth. A growing company looks like this:
Oh, wait, I forgot to define the Y axis. Let’s assume this curve is my revenues:
There are some other things which grow at roughly the same speed. For example, the number of employees:
And the number of people who have heard of your product, which we’ll call “PR”:
There’s also the “quality of your code” curve, based on the theory that good software takes ten years .
I’ve drawn these curves moving up at roughly an equal rate. That’s not a coincidence. In a small company, you regulate each of these curves so they stay roughly in sync. Why? Because if any two of those curves get out of whack, you have a big problem on your hand—one that can kill your company. For example:
- Revenues grow faster than you can hire employees. Result: customer service is inadequate. Let’s tune in to Alex Edelstein over at Cloudmark: “[Cloudmark Sales are] pretty swamped, so they’re not getting back properly to everyone…. What’s happening here now at Cloudmark is a little like the early days at Netscape when we just had too few people to properly respond to the customer interest.”
- Revenues grow slower than you hire employees. Result: you burn cash at a ridiculous rate and go out of business. That’s an easy one.
- PR grows faster than the quality of your code. Result: everybody checks out your code, and it’s not good yet. These people will be permanently convinced that your code is simple and inadequate, even if you improve it drastically later. I call this the Marimba phenomenon . Or, you get PR before there’s a product people can buy, then when the product really comes out the news outlets don’t want to do the story again. We’ll call this the Segway phenomenon.
- Employees grows faster than code: Result: too many cooks working on code in the early days causes bad architecture. Software development works best when a single person creates the overall architecture and only later parcels out modules to different developers. And if you add developers too fast, development screeches to a halt, a phenomenon well understood since 1975 .
And so on, and so on… A small company growing at a natural pace has a reasonable chance of keeping these things in balance. But VCs don’t like the flat part of the curve at the beginning, because they need an exit strategy in which the hockey-stick part of the curve occurs before their fund needs to cash out, about six years according to VC Joi Ito . This is in direct conflict with the fact that good software can’t really accomplish this kind of growth. Hockey stick, there will be, but it will take longer than most VCs are willing to wait. Remember my chart of Lotus Notes? Good heavens, I am repeating myself.
VCs try to speed things up by spending more money. They spend it on PR, and then you get problem 3 (“PR grows faster than code”). They spend it on employees, and then you get problem 4 (“too many cooks”) and problem 2 (“high burn rate”). They hire HR people, marketing people, business development people. They spend money on advertising. And the problem is, they spend all this money before anyone has had a chance to learn what the best way to spend money is. So the business development guy wanders around aimlessly and accomplishes zilch. You advertise in magazines that VCs read, not magazines that your customers read. And so on.
OK, that’s the first part of the VC crisis.
The second part is the fact that VCs hear too many business plans, and they need to reject 999 out of 1000. There appear to be an infinite number of business plans looking for funding. A VC’s biggest problem is filtering the incoming heap to find what they consider to be that needle in the haystack that’s worth funding. So they get pretty good at saying “no,” but they’re not so good at saying no to the bad plans and yes to the good plans.
When you have to say “no” 999 times for every time you say “yes,” your method becomes whack-a-mole. Find the flaw, say no. Find the flaw, say no. The faster you find flaws, the more business plans you can ding. Over at VentureBlog you can amuse yourself for an hour with some of the trivial reasons VCs will ding you. PowerPoint too complicated? Ding! Won’t tell us your magic sauce? Ding! You didn’t research the VC before you came in? Ding! It’s not their fault; they are just trying to say no 999 times in as efficient a way as possible. All of this reminds me too much of the old-school manager who hires programmers based on what school they went to or whether they look good in a suit.
Naval Ravikant, a VC at August Capital, reveals the classic VC myopia of feeling like they just don’t have time to get to know entrepreneurs that aren’t ready to pitch yet. “Most VCs are too busy to ‘dance,’” he wrote. They are too busy vetting serious proposals to shmooze with interesting companies that might not need cash right now.
This is, roughly, the equivalent of the old joke about the guy searching for his car keys under a streetlamp. “Did you lose them here?” asks the cop. “No, I lost them over there, but the light’s better here.”
But the great companies are often not the ones that spend all their time begging for investments. They may already be profitable. They may be too busy to look for VC, something which is a full time job for many entrepreneurs. Many excellent entrepreneurs feel that their time is better spent pitching products to customers rather than pitching stock to investors. It’s bizarre that so many VCs are willing to ignore these companies simply because they aren’t playing the traditional get-funded game. Get out there and pursue them!
Here’s another funny thing that’s happening. VCs are reacting to the crash by demanding ever stricter conditions for investments. It’s now considered standard that the VC gets all their money back before anyone else sees a dime, no matter what percent of the company they actually own. VCs feel like this protects their interests. What they’re forgetting is that it reduces the quality of startups that are willing to make deals. Here’s one of VC Joi Ito’s suggestion for VCs : “Sign a ‘no shop’ and get a letter of intent (LOI) signed quickly so an auction doesn’t start jacking up the price.” A no shop is sometimes called an exploding term sheet. It means that the company must either accept the deal on the spot or it won’t get funded at all. The theory is, we don’t want you going around to other VCs trying to get a better deal. It’s common among the second-tier VCs, but the best VCs are usually willing to stand on their own merits.
It seems to me that a company that accepts an exploding offer is demonstrating a remarkable lack of basic business aptitude. Every building contractor in New York knows you request bids from five or ten plumbers before you award the contract. If a plumber said, “I’ll do it for $x, but if you shop around, deal’s off,” the contractor would laugh his head off and throw the plumber out on the street. Nothing sends a stronger message that an offer is uncompetitive than refusing to expose it to competition. And that’s for a $6000 kitchen installation. Getting $10 million in funding for a business is the biggest and most important deal in the life of a company. You’re going to be stuck with this VC forever, they’re going to want to control your board of directors, they’re going to push the founders out and bring in some polished CEO as quickly as they can, someone who will take the picture of the cat off your homepage and replace it with the usual MBA jargon.
And now they want you to agree to all this in a matter of fifteen minutes without talking to anyone else? Yeah, right.
VCs who make exploding offers are pretty much automatically eliminating all the people with good business sense from their potential universe of companies. Again, it does make it easier to say no 999 times, but you’re pretty much guaranteed to say no to all the companies with a modicum of negotiating skills. This is not the correlation you’re looking for. In fact, just about everything the VCs do to make their deals “tougher,” like demanding more control, more shares, more preferential shares, lower valuations, death spiral convertible stock, etc., is pretty much guaranteed to be at the expense of the founders in a very zero-sum kind of way. And this means that smart founders, especially the ones with businesses that can survive a lack of funding, are going to walk away. VCs must realize that if the business flops, no matter how much control you have, the investor is going to lose everything. Look at the story of arsDigita. A nasty fight over control gives Phil Greenspun enough money to buy an airplane, and the VCs still lost every penny when the company went down the tubes. So all these tough deals are not really protecting the VCs, they’re just restricting the VCs’ world of possible investments to dumb companies and desperate companies. Sam Bhaumik, VC, says “VCs are being aggressive, but most requests are legitimate.” The capital belongs to public pension funds and university endowments, he notes, using the standard widows-and-orphans sob story. Boo hoo . Come on , public pension funds and university endowments are the savviest investors out there; don’t tell me they need coddling and protecting. They’re investing in risky venture funds for a reason: they want to get paid for taking risk. If they wanted protection, they’d invest in US Treasuries.
There are probably hundreds of software companies started every day. Of that universe, there is a small number that are actively looking for early stage investors. Of that small number, an even smaller portion is willing to go along with the current harsh deals that VCs are offering. Now slice away the founders who are afraid of being arsDigita’d. The population shrinks even more as VCs reject companies that don’t match their—quite reasonable—criteria for spotting a successful company. You wind up with a tiny number of investment opportunities which, quite frankly, is vanishingly unlikely to contain The Next Netscape.