Ghost Lake

Ghost lake, originally uploaded by Unsettler.

Actually Blue Lake on the SW slope of St. Helens, and in fact Blue Lake is more of a lovely emerald green, rudely forced into grayscale by my morbid desire to bring out its inner ghostliness.


How Netflix Killed Itself with Kindness

During the dot-com boom of the early oughts, the New Yorkermagazine’s financial columnist  JamesSurowiecki wrote a piece called “How Kozmo is Getting Killed By Its Customers.”He explained the plight of dotcom delivery service, which lostmillions of dollars because its customers took advantage of its low deliveryfees to purchase one-off items, like a pint of ice cream. This caused Kosmo’sdelivery costs to outpace its revenue at rate of almost two to one. Surowieckirailed not against Kosmo’s flawed business model but against the bad behaviorof its customers; he called them “little terrors” and claimed that theirprofound sense of entitlement required a “New Economy mantra: Know when to fire your customers.”

I wonder if anyone at Netflix had occasion to readSurowiecki’s piece in the last month. Right now the once-beloved dotcom, if notexactly being killed by its customers, is at least being taken behind thewoodshed for a serious thrashing. Netflix’s former business model – a singlelow price for streaming and DVD rentals – was simply unsustainable. The companycouldn’t pay the licensing fees demanded by studios, win the rights to newstreaming content, and continue to ship DVDs at the same price it once chargedfor DVDs alone. But Netflix was in a precarious position precisely because ithad provided its customers with something that was too good to be true – anexcessively dangerous position that has sunk many more dotcoms than Kozmo. Ingame theory terms, Netflix’s predicament was the patsy position, in which thecustomer has very little incentive to continue the relationship if the branddoes anything to upset the delicate balance of cost vs. services. When Netflixdefected by raising their fees for the combined services, did their customershappily agree to it, reasoning that a price increase should be expected? Um,no. Naturally, they voted with their feet and left in droves; more than 800,000customers have given the brand the boot in the last quarter alone.
Netflix was missing a core insight than any behavioraleconomist could have given them: when pricing models in a given market – inthis case, the streaming content market – aren’t well established, you’reentirely at the mercy of your customers’ own perceptions, however irrational.Since Netflix customers had been given streaming content for absolutely nothingfor several years, their perception was that the streaming content is worth,well, absolutely nothing. If Netflix had never provided that streaming servicebefore, but instead introduced it in August for an additional six bucks amonth, they would not have lost customers, and their streaming service wouldhave taken off like gangbusters. Exact same price, vastly different priceperception. In the latter scenario, Netflix’s price increase would have beenseen as a form of cooperation, not defection – another triumph for a brandknown for delivering great service at low cost.
So what’s the lesson here? Online business models that keep prices low in order to fuel growth haveto pay attention to the end-game. Consumer perception accrues quickly to thestatus quo; that’s why it’s easy to sell taxpayers on the idea that letting atax cut expire on schedule is actually a “tax hike.” Above all, know what yourcontent is worth and charge accordingly. Believe it or not, your customers willthank you.

First frost

First frost, originally uploaded by Unsettler.

Probably *not* the first frost of the year at this spot, 5,000 feet up in the Blue Mountains in Central Oregon. Just the first one I’ve been around for. It’s a pretty area, lots of big mountain meadows, for which I am a sucker. This had all burned off an hour later, and I was fishing in shirt sleeves.

What Digital Media Could Teach NPR About Pledge Drives

I’m a little grouchy this week. Morning is not my friend onthe best days, and the worst days are the ones in which the soothing warble ofNPR’s Steve Inskeep on my clock radio has been replaced by the screechy auralassault of an NPR station manager exhorting me to pledge. The pledge driveconcept appears to be modeled after a dog whistle – it emits a shrill blastthat only certain species can hear, and it draws us in not because we like itbut because we desperately need it tofriggin’ stop.
In the final days of the pledge drive, the reporters andstation managers start getting punchy from endless hours of pimping tote bagsfor cash, and their rationales for why you should pledge come unhinged: “You wouldn’twalk into a book store and steal books,then expect the other customers to pick up the tab for your stolen books, would you? Well, that’s what you’redoing when you don’t help pay for all the great content on NPR.”
I admit to getting perverse enjoyment from listening to theseweird, caffeine-fueled analogies, because they’re actually probing some hardtruths about how we get people to pay for things they might otherwise get forfree. The subject is of great interest to me because it’s one of many areas ofhuman behavior that can be illuminated by gametheory, which concerns itself with the incentives and disincentives thatdrive cooperation.
From a game theory perspective, you’d be hard-pressed todevise a worse system of incentives than the NPR pledge drive. Its fatal flawis that it rewards contributors and non-contributors equally – with freecontent – and punishes them both equally with jarringly interruptedprogramming. In game theory, this is known as the volunteer’s dilemma: all NPR listeners have a collective interestin keeping NPR on the air over the long term, but in the short term, you cancontribute nothing and enjoy exactly thesame benefits – minus the tote bag – with very little risk that yourpersonal failure to contribute will doom NPR. It’s a system that encouragesfreeloaders, and it gets them; only 1in 10 NPR listeners is also a contributor.
I have a friend who runs fundraising for public broadcastingin central California, and on a recent visit, I subjected him to my pledgedrive critique. He listened patiently, then said, “That’s a good point. So howwould you fix it?”
Um, OK, you got me there. NPR is part of a broadcast medium,and all my ideas about incentives and access to content are based on digitalmedia, which can be as narrowcast as you want it to be. And of course, as ataxpayer-funded entity, NPR is legally obliged to provide open access tocontributors and non-contributors alike. So any system of incentives would haveto start with that premise.
But in my fantasy pledge drive world, NPR could constructits incentives around whether I’d be subjected to their soul-deadening pledgebreaks. They would give me a strong incentive to make my contribution at the start of the pledge drive, because immediatelyupon doing so I would receive a code I could punch into my clock radio or carradio to shut off the pledge drive pleadings and return me to my regularlyscheduled programming. It sounds sci-fi in practice, but in principle, it’s notmuch different than me subscribing to so I can shut off the ads.
The increase in consumption of “traditional” content viadigital media offers the potential to solve classic problems like thevolunteer’s dilemma: in digital media, we have the ability to construct finelytuned incentives and pricing tiers because content can be distributedone-to-one.  In short, we can allowpeople to decide how much hassle they want to go through for access to content,and we can allow them to pay their way out of the hassle. This means we can rewardsubscribers without completely shutting out non-subscribers, who can accept thetrade-off of inconvenience for access. It’s taken us a decade to figure out howto do this well, but successful models like the New York Times’ new paywall system reflect a growing understandingof the behavioral economics governing content consumption.
That’s one of many reasons I’m excited about the amount ofcontent delivery converging around the mobile medium. Convenience is a bigincentive in getting consumers to pay for content, and mobile is all aboutconvenience (or at least it should be). NPR may not have figured out how tocrack this code, but some of their shows have.
I’m a fan of ThisAmerican Life, but I’m never near my radio during its scheduled airtimes.The TAL mobile app solves that for me, and it offers finely calibrated levelsof convenience based on what I’m willing to pay. If I’m willing to pay nothing,I get an ad-supported app that streams the latest episode. If I fork over threebucks, I get access to the episode archive, but I can only have one episodedownloaded at a time. If I want more, I can purchase individual episodes viaAmazon through the app.
That’s a smart content strategy, and it deftly avoids thevolunteer’s dilemma by giving me the short-term rewards of greater choice andconvenience in exchange for my cooperation. (Although I should acknowledge thatIra Glass still works a pledge pitch into his intro, and the app is prone tocrash. Nothing’s perfect.)
We may be a long way from NPR being able to unshackle its loyal contributors from the ear-manacles of the pledge drive, but I think we’re heading in the right direction in digital. In game theory terms, a state of equilibrium exists when content producers are getting paid (or funded), and consumers are getting the content they want at a reasonable price.  An equilibrium is a stable marketplace: customers tend to stick around, until the content producer does something to upset the balance. Sound familiar, Netflix?  Nope, I’m not even going to touch that can of worms yet – I’ll save it for next week, when the pledge drive is over and my mood improves.