New Venture Capital Models – The Rise of Business Accelerator Seed Funds (Part 3) - 03/15/2009

This is the 3rd and final post on Business Accelerator Seed Fund programs. I am assuming you have already read parts one and two, outlining the various types of emerging early stage programs and some of the trends driving their appearance. In this final post, I will look at overall success, what the future may hold, and if I would recommend that early stage companies apply. In retrospect, I could have easily done 10 posts on this topic and it has taken quite a while to get this final installment done while tracking the continued rapid developments in this space.

The first key question I want to answer is “Are these programs working?” Many of the programs are new, so there is limited operating history and many are not transparent. To evaluate success, I needed to select an older program that would have some investment history. The three I looked at were YCombinator, TechStars, and Digital Assets Deployment (DAD). There programs were all started in 2005-2006 and are generally transparent with their portfolios. In the end I chose YCombinator which is the oldest and has the largest available investment history.

There is a risk that in choosing one program, especially the earliest and best known, the results will be skewed. However, my position is that if one or more of the programs work, then the “model” works, and just like in any business category, in practice some will execute better than others. For comparison, historically, the VC model “works” but in reality dramatic returns are seen primarily by the so-called golden circle of top firms and not all firms make money for their investors.

I also decided to limit my analysis to YCombinator investments in the years 2005-2007, as these have the most available history and thus potential for exits. Looking at approximately 60 companies over those 3 years, and making some educated guesses along the way, the summary data looks like this:

YCombinators Investments 2005-2007

Making several more less-educated guesses, I did my best to estimate yearly portfolio performance so far, and the future potential given the number of companies remaining in the each portfolio that have received further funding:

Estimated YCombinator Returns

Based on this data, and depending on actual non-investment program expenses, it seems very likely YCombinator has been net profitable so far, though probably not dramatically so. Looking forward there is still tremendous potential in the portfolio.

YCombinator had a successful 2005 crop, with an estimated 7X return, though the returns were primarily driven by a single company, Reddit, which was sold for a rumored 12M. With 5 companies gone and just one still going (Loopt having received an estimated 17M in further funding), there is still potential for one more exit and even higher returns.

The 2006 year has been more difficult with 9 of 20 companies gone already and only 1 exit to date (the sale amount is not known though I estimated 5M for the purposes of my model). Four of 20 companies received further funding and 6 remain alive but with minimal or no further funding. Based on these findings, the 2006 year has probably not reached break-even yet, though the portfolio still has some potential.

2007 has been strong so far with 5 exits, and making several guesses regarding exit valuation, there is a good chance that YCombinator has seen a two times return already on this year. With a record 11 of 31 companies receiving follow-on funding and still going, the 2007 group has a very large return potential.

I wanted to combine this estimated historical performance with what is likely to play out in the future. As I outlined in an earlier post on angel investment returns, the average holding period for an early stage investment is 3.6 years, and for investments that return 5X or more, that average holding period is 4.6 years. No program has yet held investments that long. If we estimate the future performance of these programs based solely on statistical data from early stage angel returns in the past:

  1. Based on the number of companies funded and the expertise involved, there are likely still several home runs (10-30X return or more) that are still to come
  2. The highest return opportunities in the portfolio are probably not yet realized
  3. The fact that there have been some compelling exits to date means that the program is working as planned and is likely to show continued success

Therefore, based on my analysis, the model works and should play out very well for the investors. This does not mean every program will work, but the model itself looks like it will succeed and therefore it is likely several of the programs will be very successful.

With some of these programs succeeding, what does the future hold? There is no question we will see more of these types of programs emerge. The following graph captures the growth in seed capital programs from 2005-2008:

You can see we are on a steep and accelerating growth curve. So far in 2009, I am aware of at least five more programs that have launched, including most recently Capital Factory (Austin, TX) and GreenHouse (Ireland) both of which have surfaced since I prepared my first list.

One thing that will be very interesting to watch is how these programs develop in the area of funding/backing. By looking at the existing programs, and making some educated guesses based on the people involved and what I could dig up about each program, I put together this surprising pie chart:

This shows that over 60% of the existing programs have institutional support or backing, either from existing VCs or corporations. Only 17% appear to be funded exclusively by independent private investors and just over 20% are funded directly by the individuals running the program. This process involved making several educated guesses, but the bottom line is that by and large traditional early stage (angel) investors are not driving this trend, it is a handful of forward thinking individuals and experienced traditional VCs.

Looking forward, I would surmise that more programs will emerge having individual backing, as new geographic areas seek to put local programs in place and as the model becomes more widely accepted by angel investors. I know of several active discussions in this area, though so far, individual angels I have spoken to have not been particularly receptive to this new model – preferring to pick and choose their investments like they always have. Unfortunately, that model is increasingly not competitive with the fast-moving pace of Internet companies where business plans, waiting for monthly meetings, and extensive due diligence do not support the fast-time to market many ideas require. However, that is changing as evidenced by Paul Graham recently announcing his own training sessions for new angel investors.

Should you as a founder of a new software company seek to join one of these programs? Some have commented on the equity stake these programs require, and feel it is too much. I, personally, do not think so. As I commented in an earlier post, the implied valuations of these investments are not unreasonable for a new product at the idea stage with an unproven team. If the company is successful, a 6% holder will not materially impact things. If the company is not successful, well it does not really matter. If the programs are working out for investors, they are certainly working out well for the founders too.

On balance, I believe the value these programs can provide to an inexperienced team is well worth the investment. Many have commented on the quality of the program and the overall value of the experience. Raising money is always about who you know and the advice you get and rarely about the business plan. A local VC I know even stated “I receive a lot of business plans and have yet to invest in one.” A start-up company has so many challenges, they should take every opportunity they can to get help, advice, and support, and I don’t think a 4-10% stake is unreasonable for 15-75k and a lot of help.

In the end, I think the real decision on whether to sign onto a seed capital accelerator program comes to the team. For a brand new team with few family ties and limited startup experience, the decision is a no-brainer and they should seek a program with lots of support but possibly a smaller capital infusion. These teams can go anywhere and work out of hostels for the duration, if needed. And if they will fail, then it is best to fail-fast and move on. For a team with more experience, gained through working in larger businesses or prior startups, who can tackle industry-specific problems and even put a lot of their own time and money into the business, a better fit could be a more flexible or regional program with larger capital resources and a more flexible support model.  If the opportunity is large enough, these programs can help you get on the radar screen of VCs and open doors along the way.

The biggest advantage to the seed capital accelerators is the efficiency it brings to the market – the entrepreneur spends less time chasing dollars, the capital goes to work more quickly, and ideas are tested and validated faster. If you apply and don’t get in, move on - the most important thing is to get on with building your business, no matter what path you take. I have seen too many startups lose weeks and months seeking funding and ending up in a position where they can no longer survive without the money, because they stopped building the business along the way.

10 Responses to “New Venture Capital Models – The Rise of Business Accelerator Seed Funds (Part 3)”

  1. Marc Dangeard Says:

    This is great data, thank you for sharing this information.
    I am convinced that this type of model works, the key now is to fine tune the process to find the right balance between the value to entrepreneurs and the return for investors.
    I started Entrepreneur Commons ( to bring yet another option for entrepreneurs, no question also that this type of model is a much needed change in the funding process today.
    The good news is that Angel investment was approximately $20B in 2007 from the numbers I have seen, so now that data is starting to documenting that the model works, it can attract substantial amounts of money for real change.

  2. Paul J. Says:

    Great analysis.

  3. Evert Bopp Says:

    Very good analysis and delighted to see the GreenHouse being mentioned!
    Would you mind if I used some of your data while talking to people about the GreenHouse?

    Evert Bopp.

  4. karishma Says:

    great write up. thank you! good to see the momentum picking up on business accelerator funds, as i believe this is fundamental to get ideas off the ground and further increase the momentum of enabling a broader set of entrepreneurs.

    the question i have relates to your first question on “are these successful.” how do you define success of these business accelerator funds and is their business model successful given the small size of the funds, decent sized team that need to spend more “quality” time with the entrepreneurs than funds that may invest in later stages.

  5. jstjean Says:

    Evert - yes please use my analysis data per my private message to you.

    Karishma - my definition of success is pretty simple: do the investors make money? In end, that is the primarily motivation of these programs. Non-profit or traditional incubators of the past typically have not had this incentive which I believe has limited their effectiveness, in that they have more of an obligation to help many companies and could not be as focused or selective. I agree it has been an open question as don’t know the true overhead, but from what I see, the model is working at least for some programs - which means the model itself is sound. With later stage VC, many firms do not make money at all and I suspect the same will be true for seed funds. The three oldest programs - YC, Techstars, and DAD - look like they are generating investment returns despite the higher overhead. YC just raised money from Sequoia, Techstars expanded to Boston, and DAD expanded to the US.

  6. NEW - New Venture Capital Models – The Rise of Business Accelerator Seed Funds | New Hampshire Startup Blog Says:

    […]  New Venture Capital Models – The Rise of Business Accelerator Seed Funds (Part 3) - 03/15/2009 […]

  7. First Ascent Ventures final say on Microseed accelerators | VCPlan - The Path to Success Says:

    […] the subject, they look at the success rate. Without going into the detail I recommend you read the article; they recommend that you seek out the accelerators if that fits your business […]

  8. Carl Persson Says:

    Thank you for an excellent series of articles on Business Accelerator Seed Funds. I’m organizing an investment fund for this purpose that could be either for profit or non-profit.

  9. Pablo Molteni Says:

    Really great articles!! Only one question though, how do seed funds or business accelerators fund themselves? Is it with their own capital and the trading of the companies they invest on?


  10. Move Over Business Incubators and Make Room for Business Accelerators | D. Steven White Says:

    […] to the conclusion that the business accelerator model works. You may access part 1, part 2 and part 3 via these […]

Leave a Reply

[ Home ] [ Contact Us ] [Blog] [ FAQ ]