Verizon's iPhone 4: I Was Wrong. Here's Why.

"The surest way for an analyst to generate attention is by making an Apple prediction. Apple has a cult following, and its product development and launch strategy is famously secretive, so the fact that your source is the lunch counter guy across the block from the Hon Hai factory in Taiwan won't be discovered (or may even be considered authoritative!)"Those were the opening lines in my first column for SlashGear back in October of 2009. I went on to note that while I don't usually make specific product predictions, I would go out on a limb and provide a guess on when we'd see an iPhone on Verizon Wireless: in 2014, when Verizon Wireless completed its LTE rollout.

By then, Apple could provide a single iPhone that worked across both AT&T and Verizon Wireless LTE networks. I later adjusted that estimate to 2012 once Verizon Wireless accelerated the expected completion date of its national LTE rollout.

Proof that I should never had made specific predictions came a few weeks ago when I attended a press conference in New York and got hands-on with an iPhone 4 specifically designed for Verizon Wireless' CDMA network. It started shipping earlier this month.

Apple isn't building a CDMA iPhone just to sell a lot of phones. There are certainly financial considerations here – Apple was leaving a lot of money on the table by offering its phone through just a single U.S. carrier (Current Analysis doesn't create sales forecasts, but if we did, it would be a very big number). However, Apple's business model is to build a hardware platform once and then focus on software and services.

Other handset vendors build both CDMA and GSM devices, or even multi-mode devices, but for Apple to justify building an entirely separate hardware platform, it needed more than just the promise of additional device sales. Verizon Wireless admitted as much when it noted that it was never in the running for the original 2007 iPhone, and that it had to approach Apple in 2009 rather than the other way around. In addition to being an exceptionally profitable endeavor, Apple had three reasons to build a CDMA iPhone:

1. Blunt the rise of rivals. Without an iPhone, Verizon Wireless was forced to back other operating systems. The carrier proved to be a powerful backer, first fueling sales of miserable touchscreen products from RIM and then increasingly strong Android devices. To some extent, the damage has already been done with Android, which evolved extremely quickly from an OS only a geek could love just eighteen months ago, into a polished user experience with 2.2 and later. Still, an iPhone at Verizon Wireless should slow Android sales going forward and prevent Verizon Wireless from overinvesting in webOS or Windows Phone 7.

2. Offload some of AT&T's network traffic and disassociate Apple's brand from AT&T in the U.S. AT&T's network has repeatedly failed in three locations: the San Francisco bay area (where Apple is located), New York City (where much of the media and most of the financial industry is located), and anywhere a few hundred journalists converge (such as CES press conferences at the Venetian). AT&T has made strides in improving the network performance in Manhattan and the AT&T-Apple relationship remains strong, but Steve Jobs is a perfectionist, and a large part of the consumer experience of a phone is the network it is on.

3. Finally, it turns out that waiting and doing a single LTE/HSPA iPhone is not really an option. Verizon Wireless and AT&T are both migrating to LTE for their 4G network deployments, but, at least initially, both companies are using LTE just for data, not voice. This means that Apple cannot wait for the national LTE rollouts to be complete and then create just a single LTE iPhone, because it still will not function as a phone on Verizon Wireless without CDMA in there, too. (A single LTE device may not work properly on both AT&T and Verizon Wireless in any case; while both carriers are deploying on the 700 MHz band, the specific frequencies used within those bands is far enough apart that the antenna may have to be tuned to one network or the other for best performance.) Given these circumstances, building a CDMA/EV-DO iPhone is a necessary prelude to any LTE iPhone in the future.

So, now that we have a CDMA iPhone, when will Apple build an LTE iPhone? I think I should probably quit making these sorts of predictions once and for all. Sorry.