Best Practices for Accelerators: What Works and What Doesn't

[Josh Futterman]

Hi Hugh,

After getting a deep taste of mentoring with the first JFDI batch, I dived in and have been mentoring with Techstars and EFN in the United States for the last several years. I have enjoyed it so much that I am contemplating starting my own Harlem-centric accelerator here in NYC. I know that you, Meng, and many of the others involved with JFDI spent a lot of time thinking about what worked and what didn't over the last four years. Some of it may be specific to SE Asia or even Singapore, but I am sure that much of what you learned would be applicable worldwide. I am, therefore, very interested in hearing your take and the thoughts of others who were involved with JFDI. As discussed, let's use this new thread to have a conversation about best practices for every aspect of starting and running an accelerator, as well as the things you or others have tried that did not work out the way you expected.

To give some structure to the discussion, here are some key topics that might be interesting:

1) Goals and Mission

2) Messaging / Building a Profile for the Accelerator

3) Educational / Training Aspects vs. Business Needs (Structuring the Program)

4) Highly Structured vs. Looser / Sector Focused vs. General

5) Soliciting Angels and VCs for Input and Demo Days

6) Raising Money for the Accelerator

7) Budgeting (incldg non-obvious costs)

8) Mentors: Generalists vs. Subject Matter Experts / Recruiting / Structures for Mentoring / Time Commitment

9) Teams: Soliciting Applications / The Application and Acceptance Process / What To Look For and What to Avoid / Dealing with Problems

10) Capital Investment in Teams / Companies

11) Post-Accelerator Investment in and Involvement with Participants

I am certain that there are other topics that have not even crossed my mind, so feel free to add where I missed. Thanks ahead of time for your thoughts and for being such a stand-up guy.

[Hugh Mason]

Thanks for your post to this group, Josh.

For the avoidance of doubt, JFDI is still very much in business doing innovation with corporate partners (more on that to come in a separate thread) and it's only the open accelerator program that we have brought to a close. This thread is about the experience of running it.

On Saturday, 19 November 2016 22:25:04 UTC+8, Josh Futterman wrote:

let's use this new thread to have a conversation about best practices for every aspect of starting and running an accelerator, as well as the things you or others have tried that did not work out the way you expected.

I would greatly appreciate practical questions about what people want to know because I am compiling a book over the next few months and will welcome all comments on it. I am thinking of four parts to the book:

1. What an accelerator is and, in that context, for anyone thinking of setting one up how to clarify what you want out of it
2. On the basis that an accelerator program must grow out of the community, how to map the current reality of startup structures on the ground, aiming to understand the motivations, hopes, ambitions and frustrations of the stakeholders behind them.
   
3. How to evaluate the range of possible accelerator and related models available to choose what's right for your location at this stage of its development.
4.  A cookbook of recipes for the operational aspects involved, much like your list of points numbers 2-11. Actually I identified about 50 functions that an accelerator needs to fulfil here 

Cheers

Hugh

[drllau]

> 8) Mentors: Generalists vs. Subject Matter Experts / Recruiting / Structures for Mentoring / Time Commitment

If you look at an accelerator (or learning environment) it is fundamentally a service model and that by definition doesn't scale very well since you are constrained by people ... in this case quality mentors. From my enjoyable time there I roughly grouped mentors into coaches, critics and cheer-leaders ... The first mainly gave specialist advice, perhaps from an industry specific background (marketing, finance, etc), the second were the investors who did the tough love thingy to bully the participants into critical thinking of their business canvas/pitches and the last were the peers or previous cohorts who gave encouragement and sometimes handholding. So in effect, the cohorts were (indirectly via the equity retention regime) paying for skills, stories (funding experience from the other end) and support. If you had to calculate the objective market value of their time, then perhaps you can work out objectively how much cohorts cost to fill in their competency gaps and increase their survivability length. Fortunately most mentors would consider this a philanthropic endeavour, donating for a worthy cause but it creates distortions when there are too many accelerators and not enough mentors or their quality time commitments. For example, when I compared JFDI with DBS Fintech, I pointed out that 50% of their mentors were either operational or regulatory internal staff. Great if you want to do a mini-bank (cf Singapore Airlines v Scoot) but not favorable to truly disruptive startups. So in some sense the pick of mentors shapes the type and trajectory of the startups ... so when I look at the new FinTech cluster I see too many incremental me-toos but cheaper and not enough innovation (see my prior posts on cryptocurrencies). Singapore is hoping for a big win, their first $1B unicorn to draw attention but I'm not convinced about the long-term sustainability of innovation culture.

>11) Post-Accelerator Investment in and Involvement with Participants

Singapore unfortunately falls into the category of capital-lite investment (save for pet govt projects on biotech) so in a sense it is competing with labor intensive tech-oriented outfits in China / India who can scarily scale at speed. The situation is different in other countries, with US much more equity oriented + algorithmic/quant focused and Europe more debt-based and dare I say impact investment angle. The benefits of JFDI is that it is a regional accelerator who can tap into the experiences of multiple ASEAN countries which is a big advantage given the variety of regulations and cultural differences ... how the alumni can help each other or new cohorts is probably not well explored (cf PayPal mafia). There is potential for deeper engagement ... for example other countries have fractional CxO and an accelerator could be a speed-dating pond for startups to find board members or fill in temporary missing gaps, but this needs to be carefully thought out to avoid conflicts of interests. The social capital/connections aspects are probably not easily measured but may be the most value long-term benefit.

[drllau]

this guy reviews 35 papers on accelerators v incubators

https://www.linkedin.com/pulse/business-incubators-vs-accelerators-review-literature-mustafa-torun

On Saturday, 19 November 2016 09:25:04 UTC-5, Josh Futterman wrote:

[Josh Futterman]

Thanks for the link, Lawrence.