Tuesday, April 21, 2009

A Costly Price War

I've talked quite a bit in the past about the turning point that the year 2004 represented in the video game business. To recap, in that year the internet revolution finally rocked the game business to its core, with electronic distribution and new business models reaching mainstream acceptance. Between the release of World of Warcraft, the launch of Half Life 2 through Steam, the Shanda, TenCent and JAMDAT IPOs, and several other events, 2004 was a watershed year for the industry. I've argued that the nature of value creation in the video game business, traditionally grounded on intellectual property ownership and retail distribution, was forever disrupted.

But something else of great importance happened in 2004 that often gets overlooked in critical reviews of the business. It was much more subtle and easy to miss -- a publisher launched a super high quality product and offered it for sale at a very low price. This relatively small act of busting the pricing cartel and using QPR (quality/price ratio) as a competitive weapon had profound implications for one of the industry's great companies.

The game was Sega/Take-Two's ESPN NFL 2K5. This American football simulation, developed by Visual Concepts, was first launched on the Sega Dreamcast in 1999. It was considered innovative, but hampered by it's exclusive association with Sega's poorly-accepted hardware platform. In 2001, the franchise went multi-platform, and in 2003 added the ESPN branding to its NFL license. Sega aimed the product directly at one of the most popular and successful video games of all time, EA Sports John Madden NFL Football, with limited results. In 2003, Madden outsold 2K4 over 10-to-1.

In 2004, Sega agreed to co-publish 2K5 with Take-Two. The companies, faced with a daunting entrenched competitor but possessing a high quality product, took an innovative approach: they offered the game at retail for $19.99, less than half the price of Madden. And they sold boatloads. They went from less than 10% unit share of the football category, to well over 30%, and expanded the overall market.

The success of the 2K5 strategy provoked an extremely aggressive response from EA. By the end of December 2004, EA announced a five year exclusive licensing arrangement with the NFL and the NFL Players' Association. By the end of January 2005, they added a 15 year deal with ESPN, effectively stripping the 2K franchise of its branding (and most of it's legitimacy as a football simulation).

But EA accomplished this at tremendous cost. At the time of the deal, there was speculation that EA paid in excess of $300MM for these rights. Let's do some math. Let's assume, for the sake of argument, that number is correct. It creates a royalty burden of $60MM per year for each of the five years of the deal, that must be amortized across units sold. Assuming 5 million units sold at $35 wholesale (it's likely more units at a blended lower wholesale), that's a royalty burden of $12 per unit, or over 33%. Based on my direct experience as an NFL/NBA/MLB licensee at JAMDAT, that's almost twice what we paid for non-exclusive rights to the league and players' association in a typical deal.

Play that out a little further. That's over $25MM annually that would have dropped to EA's bottom line, or around 8 cents per share. Just to give you a little perspective, EA earned 75 cents per share in fiscal '06 and 24 cents per share in '07. They lost $1.45 per share in fiscal '08.

EA accomplished their goal of eliminating the 2K series as a competitive threat (in fact, they eliminated it as a product altogether), but at tremendous cost to profitability from one their most reliable profit contributors, Madden. At the time, the analysts lined up to praise the deal, assuming that EA knew what it was doing and that they could grow the football market to accommodate the increased royalties. But I wonder how many of them would think it was anything other than a Pyrrhic victory now.

Friday, April 17, 2009

Strat Planning

Following up on my post about OODA, I thought I'd expand a little on the reason for strategic planning and some specific methods I've used in the past to facilitate planning.

One of the important lessons that Colonel Boyd derived from his study of military history was the importance of decentralized decision-making and field-level initiative. As the military has applied Boyd's concepts on the modern battlefield, this lesson has been further reinforced. Quoting from the Boyd Wikipedia entry:

"... [T]he most effective organizations have a highly decentralized chain of command that utilizes objective-driven orders, or directive control, rather than method-driven orders in order to harness the mental capacity and creative abilities of individual commanders at each level."

This only works if you have (1) creative, intelligent commanders (or managers in business); and (2) a strategic framework that establishes common objectives and helps establish a context for orientation, the second "O" in OODA. This orientation context is vitally important -- I like to call it a "shared hallucination" of the company's place and purpose. Managers need the flexibility to make quick decisions, but unless those decisions are consistent with the overall strategy of the organization, the results can be haphazard and uncoordinated.

Clear, simple articulation of corporate strategy is crucial. I have primarily used two tools to help me do this: a recursive goal-objective-strategy-metrics approach to writing the annual plan, and a threat map.

In my recursive goal-objective-strategy-metrics approach, the CEO establishes the top-level plan. This plan is comprised of a single, overall goal for the company (a thing to be accomplished, not a mission statement), some key objectives that support the goal, the strategies necessary to achieve the goal and satisfy the objectives, and some metrics by which success can be measured. This top-level plan should fit on a single page.

The recursive aspect of this approach is that each of the strategies from this top-level plan becomes a goal for the next level plan. So if one of the top-level strategies is "Finish the product before Christmas," the VP of production gets this as her goal and will develop objectives, strategies and metrics to reach it. This can get pushed down even further into the organization, so departmental managers get an annual goal and articulate their objectives, strategies and metrics.

The beauty of this approach is that it rolls perfectly back up into the master plan. If you want to know why the art department is outsourcing to China, for example, you can see that it satisfies their goal of delivering all art assets in September for $1MM, which serves the higher goal of shipping for Christmas on budget, which serves the top-level objective of reaching cash flow break-even by the end of Q1. And if one of the lower-level strategies fails, if China outsourcing can't deliver by September, you can change it to something else consistent with the overall goal. And that change propagates up and down the chain.

In practice, I only used this tool at the top-level and one level down, as a way to coordinate strategy at the VP level. But it worked to keep everyone on the same page. Actually, the process of writing and critiquing the plan as a team was more important than the resulting document, because the process surfaced tensions, fostered debate, and led to a better plan.

The other strategic planning tool I really like is the threat map. The way I create a threat map is to put my business in the center of a series of three concentric circles. The first circle closest to the center represents the immediate, near-term competitive threats, the next circle represents the medium-term threats that might arise in certain circumstances, and the outer circle represents the long-term threats.

Radiating out from the center I draw several axes, representing the things that will make the company successful (or its key vulnerabilities). I then place the competitive companies and market trends on the map, in the temporal bands, near the axis along which they are attacking my company. In addition to clarifying competitive threats, this was actually a great way to see opportunities, set priorities, and think about candidates for alliance, business development, and acquisition.

For example, in the case of JAMDAT, we had three axes: intellectual property, distribution, and game quality/innovation. On our map, we identified EA early on as a medium-term threat to us via intellectual property, so we sought to license the most dangerous of their properties from them, effectively neutralizing their threat and co-opting their advantage against us for several years.

Another example: we identified content aggregators as threats to our direct distribution relationships with carriers. In response, we carefully avoided using them as an alternative to the lengthy carrier sales process, even at the cost of immediate and incremental revenue, so we wouldn't inadvertently strengthen the aggregators' collective market position against us. This is a perfect application of strategy to value creation; the default response of the sales team would have been to simply maximize revenue, but that would have diminished our long-term competitive advantage.

Ultimately, the threat map and the annual plan are just tools. Without intelligence, creative thinking, and decisiveness in execution, the best laid plans are worthless. These tools can help direct a good team, but they can just as easily reveal weak organizational design and bad hires. Caveat emptor.

Thursday, April 16, 2009

Champions League Quarter-Finals

The UEFA Champions League semi-finals are set: Man. United v. Arsenal and Barcelona v. Chelsea. The winners will play the final in Rome on May 27th. There's a chance we will see a rematch of last year's Chelsea-ManU final, although I don't think that's the way it's going to go down. This is no symbolic tournament about national honor -- this year's winner could take home nearly $60MM in prize and TV money.

Once again, there are three English Premiership sides in the final four. It's the third year in a row that we have a chance for an all-England final -- Sepp Blatter's head must be exploding. Not only does it reinforce Blatter's rants about the excessive strength of the Premiership, but it also increases the English clubs' claims on the $300MM in Champions League TV revenue, giving these top clubs an even larger war-chest with which to acquire the foreign players that Blatter has been trying to limit.

There was some great drama in the quarter-finals, at least in two of the ties. In the two that weren't particularly competitive, Arsenal dismantled Villareal 4-1 on aggregate, and Barcelona absolutely shelled Bayern Munich 5-1 on aggregate. Barcelona's 4-0 win at home was particularly eye-opening; they dominated Bayern so completely, and looked so dangerous in attack, that match could have been 8-0.

The Man. United-Porto tie was surprisingly competitive. Porto came to Old Trafford and toughed out a 2-2 draw, grabbing two crucial away goals. That put United in position to have to win on the road in Portugal. Which they did, but only 0-1 on an early goal from Christiano Ronaldo. If Porto had managed a late goal and a 1-1 draw, the Portuguese side would have gone through on away goals.

But the most enjoyable tie was certainly Chelsea-Liverpool. Chelsea was on the road in the first match and punked Liverpool 1-3, setting a seemingly impossible bar with three away goals. Liverpool had to win 0-3 on the road to win the tie. They opened the match at Stamford Bridge with two first-half goals and Chelsea was clearly sweating bullets down 0-2 at the break. Chelsea roared back in the second half and went in front 3-2 with 10 minutes to go. But then Liverpool scored twice, making it 3-4, pulling to within a goal on aggregate, and, with 4 away goals, seizing the tie-breaker advantage. It was down to the wire until Lampard scored in the last minute to clinch it for Chelsea, 4-4, 7-5 on aggregate. Classic.

Predictions? I think Arsenal can beat United (they beat them 2-1 in their league match back in November), but I just don't think it's in the cards this year. I pick United to go through to the final again by a slim margin on aggregate. Barcelona v. Chelsea will be highly competitive, but this year the Spaniards are just too strong to bet against. They've scored 87 goals in 30 matches in La Liga, an average of almost 3 per match, and have an astonishing +63 goal differential. They are just humiliating teams this year. I think they go all the way.

Tuesday, April 14, 2009

Disrupting the OODA Loop

I got a lot of re-tweets over the weekend regarding a blog post by Kas Thomas on "The Principle of the Last Responsible Moment." Thomas' post has mostly to do with project planning and the fallacy of the complete specification. But the actual principle he's talking about is a military tactic which is oriented toward threats, or in the realm of business, competition.

I've always been a big believer in strategy, but I'm an equally big believer in execution. They are the yin and yang of business competition; execution limiting and informing strategy, strategy creating the pathway for execution. I've always thought it absurd when somebody describes a business as being either a "strategic play" or an "execution play" -- business competition is inherently always both.

The military has been dealing with the problem of strategy vs. execution for a long time. Everyone has heard the truism that "no plan survives contact with the enemy," but the military has a doctrine to deal with that fact, called "OODA" -- Observe, Orient, Decide, Act. OODA was first articulated by Air Force Colonel John Boyd, and it describes a continuous command and control feedback loop, in which unfolding circumstances inform changes in strategy and execution. It is to the military what agile development is to the software team.

To me, Boyd's key competitive insight is the idea of "tempo" -- the speed with which you can run your feedback loop compared to your enemy. The most profound disruption occurs when you can strategize and execute inside the decision-making window of your enemy. As Boyd wrote about aerial combat:

"Time is the dominant parameter. The pilot who goes through the OODA cycle in the shortest time prevails because his opponent is caught responding to situations that have already changed."

This is a massively important idea. Getting your opponent to respond to your moves prevents them from seizing the initiative and surprising you. I've seen this play out in business many times. In traditional video games, long development cycles created numerous examples of teams that conceptualized new products and delivered them to market while their competitors were mired in delays, grabbing market share and reinforcing brand. We've all seen supposedly weaker technologies vanquish more established ones through rapid, directed innovation in response to market forces.

Time is of the essence. If we've learned anything from the media and technology revolutions of the last 30 years, it's that things happen a lot faster than everybody thinks.