I'm fascinated by a recent Ryan Lawler post at GigaOm, which argues that longstanding beliefs about how often people get a new TV have been upended in recent years. Apparently, conventional wisdom among TV manufacturers was that people got a new set every seven or eight years.
But over the last five years -- with the mainstream arrival of HDTV -- researchers note that half of US households have purchased an HDTV. That would suggest that the upgrade cycle has increased considerably over that timespan.
But what's fascinating is how Lawler's sources want to extend this accelerating trend into the future. Having recently discovered the conventional wisdom based on trend was incorrect, they immediately create a new conventional wisdom based on even more limited data.
First, they assume that TV upgrade cycles might end up looking like computing or mobile phone upgrade cycles. "HDTVs might not have reached the two-year replacement cycle that most consumers have for their mobile phones, but consumers are definitely making HDTV purchase decisions more often," Lawler writes. "Those decisions are happening more along the lines of the three- to five-year cycles that consumers have for computing devices."
Then, a shopping site spokesperson argues that it is the increasing quality of TVs that are driving upgrades and that higher quality will equal even faster upgrades.
"Low prices make it tempting for people to replace their TV sets more often. We estimate that today's household replaces it's TV set every four to five years. If TVs continue to get bigger, better and significantly cheaper, we estimate that people will replace them more often," Retrevo spokesperson Jennifer Jacobson wrote in an email.
Finally, Lawler offers an argument for how TV quality could continue to increase, even though visual quality has basically reached the limits of the human eye.
And as time goes on, there might be another reason for consumers to begin replacing their TV sets -- or at least, the TV sets in their living rooms: They might soon become obsolete. As more and more "smart" TVs enter the market, the applications and app development frameworks available on the first generation of Internet-ready televisions will find themselves eclipsed by more powerful and attractive options.
Now, I have no reason to doubt the idea that people have replaced their televisions more quickly in the last five years than in the years before that. But I do doubt all the rest of this reasoning.
First, HD is a special technology. It takes screen resolution from a low-quality easily perceived by the human eye to just about as good as anyone can notice. Perhaps that was a generation-level sales point that won't come around again.
Second, there's no reason to expect that TVs should follow the purchase cycles of phones or computers. Sure, they're both types of electronics, but so what?
Third, absent a compelling visual reason, it's unclear that people will continue to purchase televisions at a faster and faster rate. That people want a better picture is well-established. That people want a "connected" television is far less so, and there are many other ways to achieve the same service without having it built into the television.
Fourth, and most importantly, seeing a change in a trend shouldn't just get us to immediately adopt a new dogma. It should make us step back and question how well we know what's going on and whether we can trust our trend forecasting. Lawler's thesis may be correct, but it might not be What troubles me is the assumption that a nascent trend will hold. If a long-established trend can break down, what gives us confidence that a more recent one should accelerate?