Vertically Focused Accelerators

[Hugh Mason]

I was asked this question in a private email and thought it better to answer it here in case anyone else wants to chip in with an opinion too:

Do you see a need for vertical focused accelerators?

As I understand the history, VC funds started out as generalists and have become more specialized over time (specialized by domain, by deal size, by geographical focus). So the received wisdom is that, over time, maybe accelerators will do the same.

Perhaps someone from the VC industry could comment but I can imagine several reasons why VC's have specialized, including:

a) specialization helps to pitch a distinctive investment thesis if you want to raise money from LPs.

b) specialization helps to make wise investments if it brings deep domain expertise.

c) geographical focus helps, especially with early stage businesses, if you want to offer more than money and to add value as a hands-on 'operator VC'

Now of course accelerators are not a form of VC (and they are not incubators or business angels either, as Susan Cohen cogently explains). Writers like Brad Feld have argued that a core part of their function is community building and in that sense geographical specialization seems very natural (hence we have Techstars New York, Techstars Seattle etc).

There are accelerators that specialize by domain and often that grows out of local economic geography. For example, The Brandery's focus is on marketing because it's right next door to P&G's global HQ. JFDI.Asia's focus is on Asia because we don't see any point in replicating what could be done better elsewhere. Rather we want to create tech that grows out of the special circumstances of Asia (mobile platform very definitely first, fragmented territory/culture/language/regulation, asian family values and life patterns etc) but of course that's still quite general. The domain definitely impacts the shape of an accelerator program because (for example), compared with SaaS, hardware takes longer to prototype and B-B takes longer to do customer discovery. And if you wanted to do hardware you'd have to be very confident that you had investors and supply chain management around you that could scale up in the happy circumstance that you hit gold, for example as Highway1 has done. Otherwise you risk replicating the fate of many a crowd-funded neat idea that crumples under the demands of supply chain management, scaling customer service and distribution and working capital etc. if it is successful.

We are thinking of setting one up for gaming, since [JV partner] brings in a lot of expertise in the gaming space and [JV Partner] has the distribution/marketing/a gaming platform. I don't see games in the JFDI portfolio - any particular reason why? Is it because they are a hit or miss business and currently there is no way to identify hits at an early stage?

Yes that's the main reason. I personally mentored 'creative industries' SMEs for about a decade and worked with scores of games companies. Like movies, there doesn't seem to be any particular logic to why an individual project succeeds or fails. It's possible to talk generalities (casual games and cheap schlock horror movies are lower risk because they are lower cost) but there seems very little science possible in that world. Even people who hit the big time can't really explain convincingly why they did - there are countless other games that look remarkably like Angry Birds to me.

Looking from the outside, the way to make money out of games might be to do what California's first millionaire, Samuel Brannan, did in the 1840s. He hyped the gold rush then sold the schmucks who came digging their picks and shovels. Now far be it from me to say that games developers are shmucks but I have worked with many, many games developers (and movie producers) who delude themselves that there is something magical about their creativity, even though they can't exactly define it. I personally would rather invest in companies doing games distribution, analytics or in-game advertising than in the gameplay gurus :)

[Meng WONG]

On Sun, Jun 22, 2014 at 9:25 PM, Hugh Mason <hu...@jfdi.asia> wrote:

I was asked this question in a private email and thought it better to answer it here in case anyone else wants to chip in with an opinion too:

Do you see a need for vertical focused accelerators?

Now of course accelerators are not a form of VC (and they are not incubators or business angels either, as Susan Cohen cogently explains).

Accelerators are a new kind of animal, but if we must see them as a chimera, then they are perhaps a cross between VC, co-working space, hackerspace, guildhouse, engineering school, and business school.

Universities scale horizontally with med schools, law schools, architecture schools, and public policy.

Accelerators do the same: now we have fintech, edtech, and hardware accelerators.

If we are to grow JFDI, it makes sense to scale that way, with faculties of this and that, expressed as different programs.

[drllau]

I can only go on economic history in other sectors. Generally over time, any organisation, whether individual firm or geographical cluster acquires capabilities/resources which become increasingly specialised. Singapore is a classic example becoming a logistics hub. In the long-term these expertise persist, despite the erosion of the original industry base ... see London as Maritime insurance despite the UK merchant fleet being scattered to Greek or offshore flags. Venture capital is a relatively new sector, as distinct from private equity (which is more pre-IPO). The differences are

a) often new categories due to a foundational knowledge base (TCP/IP, Social-networking, bitcoin blockchain etc)

b) unexplored value chain ... often shifting or hard to predict who can capture the choke-points (eg Intel+MS after IBM's PC architecture)

c) risk-return curve often extreme (thunder-lizards etc)

Now the general pattern over time is that mobile capital (people, money, ideas) is successfully captured by geographical attractors whether social hub (SV sandhill), low transaction costs (TechStars), or supporting institutions/infrastructure (eg subsea cable to India). I suspect VC will follow the same pattern but the hard part is identifying what are the unique components which allow repeatable scalable sectors (allowing for technology evolution). This can be a combination of culture (cf US West/East coast), capital (eg Formula 1 test circuits), or simply connections (right people at right time). These may be emergent properties and thus not suspectable to top-down mandates (cf Malaysia multimedia supercorridor, Japan's 5th Gen, etc). As emergent, they are thus context sensitive and by definition hard to replicate elsewhere (cf Isreali pressure cooker). But that means that mere capital accumulation (buying machine or hiring people) alone is insufficient to create the specialist resources as it may be unique knowledge (eg chipfab economics) or the supporting ecosystem. For example, I've heard stories that there may be ~100 creative small studios in the movie industry about half which reside in or around LA which has all sorts of odd-ball expertise whether puppetry, animatronics or period clothing. Whether a single accelerator can deliberately create such a resource is an open question as it depends on both scale and scope. 

Lawrence

http://www.linkedin.com/in/drllau

On Monday, 23 June 2014 01:25:10 UTC+12, Hugh Mason wrote:

I was asked this question in a private email and thought it better to answer it here in case anyone else wants to chip in with an opinion too:

Do you see a need for vertical focused accelerators?

[Meng WONG]

On Mon, Jun 23, 2014 at 12:33 AM, Meng WONG <meng...@jfdi.asia> wrote:

Universities scale horizontally with med schools, law schools, architecture schools, and public policy.

Accelerators do the same: now we have fintech, edtech, and hardware accelerators.

If we are to grow JFDI, it makes sense to scale that way, with faculties of this and that, expressed as different programs.

Given that we are tooling up to run several batches a year, how about we theme one batch to each of

1. hardware acceleration – Internet of Things

2. fintech – bitcoin and other innovation in payments, remittance, and alternative currencies

3. emerging technology – tech transfer from academic and corporate R&D

4. Big Data

"Theme", here, means simply that a majority of startups in the batch are on that theme – but good teams with good ideas don't need to wait!

In the past we have seen themes emerge around EdTech and Social Platforms.

[Roland Turner]

On 07/20/2014 04:32 PM, Meng WONG wrote:

> Given that we are tooling up to run several batches a year, how about
> we theme one batch to each of
>
> 1. hardware acceleration – Internet of Things
> 2. fintech – bitcoin and other innovation in payments, remittance, and
> alternative currencies
> 3. emerging technology – tech transfer from academic and corporate R&D
> 4. Big Data
>
> "Theme", here, means simply that a majority of startups in the batch
> are on that theme – but good teams with good ideas don't need to wait!
>
> In the past we have seen themes emerge around EdTech and Social Platforms.

This is certainly a smaller initial step than starting with running a
parallel program; it may even improve the existing program by reducing
complexity within each batch. Has there been a large enough set of
viable applicants in previous batches to permit this sort of intentional
thematic focus?

- Roland