Saturday, March 19, 2005
Fascinating example of an LT business model. The more I think about this, it seems obvious that retailing is the prime candidate for losing out to an LT strategy. After all, the economics of retailing consists of creating scarcity by controlling shelf space and location on the demand side, and building of economies of scale through this limited choice of goods on the supply side. Both of these are absolutely vulnerable to the economics of abundance made possible by an LT strategy.
The G interviews Trip Hawkins - recommended.
A nice piece in the Times today:
"...Mr. Martikovic, 30, a junior architect who relies on a cellphone for his normal calling, says he never uses the desk phone - but he pays $360 a year to keep it hooked up.
"I have to pay for a service I'm never using," he said.
He has no choice. His telephone company, SBC Communications, will not sell him high-speed Internet access unless he buys the phone service, too."
File under: Muni WiFi market drivers, telco strategy decay, FCC vs the People.
lso note that what kill the telcos time and time again is not
technology - the new fiber initiatives is not the answer. What does it is strategy decay, time and time again - their strategies are almost (hilariously) perversely misaligned with market needs, tech shifts, industry boundary blurs. By now, I'd guess it's a case of inertia - been down so long, don't know which way is up.
Art of the Day
VGMaps has the potential to stop me from enjoying the springtime weather which struck London with a vengeance this weekend.
Friday, March 18, 2005
A proposal to end the arms race between adware and blockers - make adware more effective. Kind of related to Cringely's point about spam (ie, counterintuitively, that volumes will decrease when effectiveness rises, not vice versa).
More Human than Human
"...Using such non-judicial variables as race, sex and age, an artificial neural network can predict with greater than 90% accuracy who will receive a death sentence.
Built and trained by researchers at Loyola University New Orleans, the artificial neural networkï¿½a multiprocessor computer that resembles the way biological systems process informationï¿½has cast further doubt on the fairness of capital punishment."
The Neverending Decline of Star Wars
Just when you thought it couldn't get any worse, Lucas describes Episode III as 'Titanic in Space'.
Copyright Law's Theory of the Consumer. Dense, but recommended.
Video Game Economics
Structure and Competition in the US Home Video Game Industry. If you need an intro...
Thursday, March 17, 2005
I recommend Jeff C's coverage.
Tags are Playlists are Network FX
Yesterday I made the point that tags and playlists are essentially the same thing. Today, I discover Phonetags, which let you use SMS to 'tag' songs on the radio you like...cool.
Open Access Models
JHU School of Public Health launches OCW (open courseware). The growth of open access models across markets and domains continues to amaze me.
Stringer Vs Morita
CNet thinks Samsung is the new Sony. Funny, I just read an analyst report that suggested this.
Look, Samsung is so not 'the new Sony'. Sony is the new Sony. Samsung has no content assets, totally different competences (ie Sony can still make things cool), and they have no idea how to marry content and technology. Well, neither does Sony - but it's learning. This is not a great comparison - if you look at business vs technology, you see why.
The Office on MySpace. Note the continued emergence of connectedness as a key consumption driver across media markets.
Lessons From Hollywood
Everything you ever wanted to know about the film industry - this is a great basic resource.
Wednesday, March 16, 2005
Tags, Peer Production, Peopleweb
Fred has a cool post on Pincus's idea of the Peopleweb - that the unit of webbiness will flip from 'pages' to people (or other assets, if you like).
This is fundamentally about the notion of peer production. The economics of surfing people/assets are superior to surfing 'pages' - specialization means greater efficiency (for the vast majority of domains).
What's gonna enable this? Pincus is opening up Tribe profiles - open-access is necessary, but not sufficient. Like I pointed when discussing Zopa, the peer-production/distributed ecs of scale model becomes revolutionary when two components are joined: open-access on the supply side, and equally important, some form of shared search/coordination on the demand side.
So we still need search and coordination mechanisms (ie, sharing). This is where tags are gonna come in - because they transform search from hierarchical to multidimensional, massive efficiency gains can be created by surfing people/assets multidimensionally rather than pages hierarchically. Tags are like playlists - a simple form of sharing that has absolutely massive economic consequences.
I'm sure we'll see other (cooler) mechanisms as well, but for now, tags are how, for my money, peer production models like the Peopleweb will explode. But first, we really need better name for all this stuff!!
Video Game Economics - Lessons from Hollywood
By now, we all know that the strategy most game publishers are using is to acquire IP (ie, brands like the Matrix etc) from the film industry, and leverage it to market games more efficiently. The problem, as Costik has been pointing out for years now, is that this is also a fairly efficient innovation killer - what we might call the developer's POV. This is why Seamus Blackley and some friends tried to set up CEG - a production fund, targeted at the lack of capital for innovative developers, a year or two back (but this didn't work out).
Of course, in the long run, acquiring IP is a dominated strategy. That's because as the demand for IP increases, prices will get bid up, and as bidding wars erupt, the games industry will erode it's own margins. It's essentially supply-side price competition. In order to stave this off, publishers are consolidating in order to exert more buyer power over the film industry.
Now, to be fair, the process will take a long time - bidding wars are only starting to emerge. The average cost of IP, is, I believe, on the order of 20% of gross, with megalicenses like the Matrix costing upwards of 30% of gross. This number is slowly climbing.
All of this begs the question: why can't the games industry produce and market it's own brands as efficiently as the film industry? Understanding this is the heart of understanding the next wave of video game economics.
There are two possible reasons:
1) The film industry can attract better creative talent than the games industry.
2) The film industry realizes more marketing economies than the games industry.
Let's leave number 1 aside, and assume that talent is equally distributed (for now). I think a much more insightful answer comes from thinking about number 2.
The average marketing cost of a film is much greater than the average marketing cost of a game. For a film, it's $39 million (MPAA number - this sounds pretty high to me). The average development cost of a game is about $6 million; liberally assuming 100% marketing costs, the comparable is $6 million.
This is an order of magnitude difference. It should be intuitive to most of my readers. The interesting question, again, is why this is the movie industry can spend so much more on marketing than the games industry, especially given the fact that game industry heads trumpet the fact that the total market size for games is bigger than the market size for film box-office revenues.
The answer, and the point, is that the film industry has a more efficient supply and demand side business model than the games industry does: multiple revenue streams on the demand side, and risk-sharing agreements on the supply side.
The games industry only really has one significant revenue stream. While games sales are bigger than domestic box-office receipts, box office receipts are now the smallest component of film industry revenues. The other big components are DVD sales and library/syndication revenues.
At the same time, Hollywood has minimized the supply-side risk of movie-making by utilizing cofinancing structures, which are essentially syndication arrangements for financing across multiple publishers/studios/sources of capital.
All of this means that, on average, making a movie is relatively less risky than making a game. Films are still a risky, hit-driven industry - but because revenues are more diversified and capital is syndicated, making films is less risky than making games. Put another way, the film industry realizes greater marketing and brand economies than the games industry because it's more efficient: for the same unit of output, it earns multiple revenue streams while minimizing production and distribution risk.
What does this mean for the games industry? Well, leaving aside the isolated instances of IP flowing from games to films (Mario, Resident Evil), the business model needs significant deconstruction and reconstruction if the industry's growth is to continue - if we want innovation to heat up and innovative developers to see funding, the model needs to enable publishers to spend more
(not less) on marketing.
One answer is to follow the film industry's lead: diversify revenue streams and utilize risk-sharing arrangements, in order to maximize the amount of capital flowing into and out of the industry. Think about it this way: why does the games industry still only have one significant revenue stream?
In stark contrast, the film industry is constantly experimenting with new revenue streams - here's a great example. The games industry isn't an infant anymore - if it wants to survive (rather than simply becoming a marketing medium for Hollywood, which is what's on the cards now), it's got to think a little more strategically.
Now, a Hollywood model this won't completely address the developers' complaint - risk-averse publishers stifling innovation. We only have to look to Hollywood's annual summer schlockfest for evidence of this. But what it will do, I think, is give developers a lot
more freedom, because they will at least be able to develop their own brands, characters, and other IP. So it's a step in the right direction.
Of course, the most efficient (and innovation-friendly) model is a publisher-less one - something the Net's been promising for a decade but has yet to deliver. Examined in the next installment...
Monday, March 14, 2005
Worked too hard over the weekend and now I'm down with a nice cold.
RSS ads are really irritating.
Tomorrow, I'll respond to the GDC threads, discuss new b-models for the games industry, and why licensing IP from the movie industry is a dominated strategy.
This one's for the math geek...
Ramanujan's Lost Notebook, 29 years after its discovery
Srinivas Ramanujan is the pinnacle of human genius and one of the greatest mathematicians of all time, in my opinion.
A couple of BoP books:
Banker to the Poor: Micro-Lending and the Battle Against World Poverty
How to Change the World: Social Entrepreneurs and the Power of New Ideas
The website that tells you how much next door paid for their house
Innovation by integration. Remembered this example when looking at this similar site in the US.