A Guide to Seed Fundraising
themacro.com
themacro.com
There was a great comment on HN a while back (I wish I remembered who said it) that said, "If you're asking about traction or revenue, you aren't making a seed investment".
Andy Bechtolsheim was a seed investor in Google -- he gave them a check based solely on their idea and who they were. Paul Graham/YC made a seed investment in reddit and Justin.tv and whole host of other companies -- none of them existed as more than idea when YC invested.
But VCs are risk averse people (ironic given they are in the risk business), so it makes sense that they would rather invest after a company can show a little traction, especially now that it's so easy to get that initial traction.
I just think we need a new term besides "seed" to differentiate it from the true seed investments.
It's gotten much cheaper to get traction. In some consumer markets, it is basically cost-free; all it takes is a motivated technical entrepreneur with a couple months of savings.
a) You raise money after having invested a year of work on your own dime.
b) You raise money based on your connections/pedigree ("reputation").
Not boasting, but simply stating fact, that on my own, I've written:
- A system for managing employee leave and sickness with google apps integration in 2 months
- A credit control application for one of my clients in 3 weeks, automatic emails, dunning letters, cash-flow forecasting
- A purchase order system that would automatically raise invoices in their accountancy system in 3-4 weeks
- An entire search system a la booking.com in 4 weeks, with named location search, geo location, tags, categories, blah, blah, blah. Basically the heart of any listing style startup.
And while I realize I'm a good programmer, I'm not a phenomenal programmer, I'm fairly lazy, I find it hard to focus on more than about 5 hours of actual programming unless I'm doing something really interesting.
And these are some off the top of my head examples. I've written loads of small startup sized apps that would be of the complexity that YC fund in much less than 12 weeks, let alone 12 months.
You're looking for an MVP with some traction, not a polished product. Reddit, for example, launched without the ability for users to create their own subreddits, and many would describe that as one of the key reasons of Reddit's growth. I believe they actually launched without commenting, but I can't find a source on that.
While today's startups have more expected of them, it's now easier than ever to write a web app, even SPAs that even 3 or 4 years ago would be too hard for individuals to write on their own.
reddit launched with only voting and submitting of links (no self posts). Commenting, subreddit creation, self posts and all that was added years after launch.
(Source: I was there :) )
VWO, a bootstrapped A/B testing product, started getting paying customers with a very basic interface and reports[1], and today it is a ~$10M business.
The thing about this is if you read about ycombinator it's all about getting a product out of the door within 3 months. And they don't expect you to have a product when you start. In a team of two they seem to say one person should be focusing on the business side, so that is one person getting a product out of the door AND get some growth in 3 months with some help from the other. They expect it themselves.
There's no way you could clone Dropbox's first version in anything less than 6 months, and that assumes you're a great programmer.
It took the Dropbox team something like 18 months to publicly release, which was all paid for by funding. It required significant funding to build and to operate (freemium).
reddit was also created with funding and could not have existed without funding. It made no money.
I know the story of a lone developer creating a little app that takes off sounds good, but the reality is most significant projects take a good amount of time and money.
BTW. Andy Bechtolsheim invested in Google three years after they'd started working on it (as part of their Phd), they'd already launched a public version (google.stanford.edu) with real users by the time he invested. Justin Kan had been working on Kiko for a year before YC invested. Alexis & Steve had been working on MyMobileMenu for a year before they pitched it to YC and PG convinced them to pivot.
It's easy to romanticizes away the slog of work that goes in before a startup raises money, but many successful founders had years of work before they raised seed funding.
And some of them did and were brave enough to write it up:
I agree. I have been thinking about what to call an investment at the concept stage. The obvious one is “seed bed” (you need a seed bed before you can plant a seed). The other is something combining founder and investor as any investor at the concept stage is really more a founder than an investor - festor doesn’t have quite the ring I would like though :)
Money from seed rounds in the past would be usually spent in infrastructure and distribution. With cloud providers and app stores that part of equation got changed, so some ideas (not all) can be validated on the cheap. (I believe even one of your examples, Google, existed as a proof of concept product called Backrub, working on Stanford network, before the incorporation event took place).
Another point is that investors judge the pitch by the quality of other pitches they heard this day/week/month/year. If you browse AngelList, there are companies raising seed rounds with a functional team, a built-out product, traction and a plan to scale. If someone comes right after them and tries to raise a similar round with nothing but a Powerpoint deck, they will be (even subconsciously) compared against the others.
Institutional seed rounds used to not exist (or barely). Thus angels investing to get the company to the A round was aptly called Seed.
But now, there are many Seed VC funds around and the size of the round has grown. (and so has the size of the A round) You may recall articles in the past 18 months proclaiming that the Seed round is the new A round.
In such an environment, what used to be called the Seed Round can perhaps be called the Angel round. Many startups with credibility (YC, repeat founders, researchers, etc.) may be able to skip the Angel round and go to the Seed round with favorable terms.
>A rule of thumb is that an engineer (the most common early employee for Silicon Valley startups) costs all-in about $15k per month. So, if you would like to be funded for 18 months of operations with an average of five engineers, then you will need about 15k518 = $1.35mm.
and
>Most seed rounds, at least in Silicon Valley, are now structured as either convertible debt or simple agreements for future equity (safes). 17 Some early rounds are still done with equity, but in Silicon Valley they are now the exception.
As you can see, this means the entire guide is written from the point of view of Silicon Valley. It still calls raising a seed round "brutal" and "long, arduous, complex, and ego deflating."
But it is, at least, possible. For the vast majority of the cases they are talking about, after the long, arduous, complex, and ego deflating process, outside silicon valley there is still no seed round that has been put together.
So, if you are reading this guide outside of silicon valley, please realize is that your first step toward having a crack at this arduous process, is to move to SV. Airbnb, based in San Francisco, still had these 7 famous rejections for its modest seed round: https://medium.com/@bchesky/7-rejections-7d894cbaa084
There's not a chance in hell it would have been funded on that model at all basically anywhere else in the world. Do what it takes: if your business needs a seed round, you owe it to your business to move to silicon valley.
Similar to how clear and concise the SAFE documents are, this is really nice to read.
Imagine a variant of D&D where every encounter begins with you and your party naked, and after assessing your enemy you have to negotiate with Elves for your armor and weapons - to be paid for after either winning the battle. (If you lose, you might have to give back your battered armor, and try a different group of Elves). In the best case, if you beat the final Boss (which is always the same, the Grand Vizier of Product Market Fit) then you get to keep like 10% of the treasure and the Elves get 90%.
(You could grind lower level monsters to buy your own armor, but that can take a very long time and it might wear you down until you're no longer fit to be a warrior anymore. The Elves are smart because they risk money (which is replaceable) but not life (which is not).)
When you're in the seed stage, angels are investing in you and the idea. Some will want to know about traction MRR (walk away)and others will want to do deep-dives of your concept (again, walk away).
So if you are in a situation like this force their hand, the ones that are in will still be in and the ones that were just dicking you around will bow out right away.
The company is still around, they're doing fine. They will never be a 'homerun' but I don't care, they're good people and they work hard, it's just that their pivot turned out to be less of a money maker than the project they envisioned starting out (it was either pivot or die, I give them great respect for staying alive and generating a ton of work during the last 8 years), and I'll always have their back, no matter what.
That's a pretty broad statement I would take issue with calling angels as a group, amateurs and hobbyists. Especially since some well know VC's also do angle investing.
We made this mistake and failed miserably. It was only after we had a beta, some customers, and some press that people started to pay attention. Now we're close to closing our seed round and it's because we had better focus around what we were doing, how we were going to do it, and why it was important.
At the end of the day, a VC is interested in answering one question, "What's the 10x return?" How will you make them lots of money. Unless the pieces are all there to answer that question, Founders will get a lot of no's. The goal should be to go through your pitch deck and find ways to eliminate the ability for VCs/Angels/Investors to say "no."
I wish this was a bit more specific -- how long exactly is "several weeks"? 3 weeks (total growth of 33%)? 7 weeks (total growth of 95%)? 13 weeks (total growth of x3.45)? 52 weeks (total growth of x142)?
I wonder if a good way to explain this would be via examples of (starting point, growth rate, duration) tuples and an "interesting" / "not interesting" assessment for each. (Can YC publish anonymized data on its portfolio companies?) This would allow readers to look for an example which roughly matches their performance -- which may sound silly to people in the valley who are surrounded by startups all day long, but there are a lot of us outside of the valley who rarely meet anyone working for a startup and never see any sort of concrete numbers like these.
I absolutely agree about keeping the guide short though -- this is more a matter of something I'd like to see you (or someone else at YC) write more about in the future.
Great job~