Compared to many of Google's latest initiatives, delivering wires into homes would seem to be a lot more capital-intensive. But Google is up to something with its fledgling TV-broadband access business in the Kansas City area. Back in 2010, Google called the service an “experiment” and said the aim was to make it available in up to 500,000 homes. Since then, the rollout has had considerable bumps, and Google likely has a ways to go before it approaches that figure. Still, as it tries to make broadband and TV up to date in K.C., speculation continues whether it will stop there or expand to other markets. It’s Google, after all. It doesn’t think small. And it has a new deal to move into suburban Olathe, Kan. But since no one outside Mountain View knows the end game, Bernstein Research is probably a good source to turn to for insight. Its prediction? Google will look to go elsewhere, but isn’t seeking a footprint anywhere close to the largest cable and telco operators. The expansion would be a “few million homes” over time. Perhaps good enough to be one of the top-10 providers, but not enough scale to “make a difference in shaping the structure of the broadband access industry in any significant way.” When Google was prepping Fiber, some 600 communities fought to entice it. The Bernstein analysts wrote late last week that they can “almost envision” top Google executives -- including CEO Larry Page -- “announcing a competition for the next three, five or even 10 cities” to compete for the service. But Bernstein indicated that any expansion is predicated on establishing the “economic feasibility” of the K.C.-area operations. It estimates that Google is investing some $100 million in the project, where it is an over-builder -- businesses that have a checkered history. Google may yet mount a significant challenge to Time Warner Cable, AT&T and now Comcast in Greater K.C. But it won’t offer phone service, so it won’t have the triple play as a marketing tactic. (Then again, in 10 years, won’t landlines have gone the way of Netscape?) Google does have a virtually free broadband offer (there’s a $300 start-up cost). Most prominently, paying customers have access to speeds that are much faster than existing providers. There's also the Google name in a business where cable providers don’t exactly engender affection. But reading the Bernstein report, the overriding goal of Google Fiber really isn’t really a surprise: research and testing writ large, not stand-alone commerce. Bernstein says based on an estimated installation investment of $2,400 per home, it could take more than three years to break even on a dwelling. With Google Fiber, the company wants to get under the hood of a business that has a major impact on its future. It wants to understand the financials, technology and consumer behavior. Time Warner Cable, AT&T and their cohorts are Google gatekeepers. Bernstein says Google “should be” worried about cable operators having an “almost-monopoly” in many areas. That could open the door to charge Google fees with the threat to “prioritize traffic” to other sites -- such as a Facebook or a Yahoo. (Hard to believe Congress wouldn’t get involved there, but who knows? Bernstein seems to think Google would lose over the long haul.) Also, with Google’s core operations increasingly reliant on faster broadband speeds, Bernstein suggests that an entry into the access business may be in response to some fears that incumbent providers will curtail investments there. If Google proves there’s a hunger for more speed in K.C. and Time Warner Cable and AT&T feel they have to respond, that may encourage them to expand the ignited service elsewhere. Sometimes lost in the Google Fiber fascination is its TV service. That's certainly understandable, considering that Google isn’t really gambling with different channel lineups more than competitors. A “Google Goes Without Sports Channels” headline would focus more attention, although Bernstein says it is offering a Netflix opportunity that other operators are not. But Google may look to innovate with the user interface, as operators believe consumers want improved search and recommendation engines. It’s also likely, Bernstein says, to offer some unique features, although operators are innovating quite a bit themselves. This isn’t a decade ago. Google is such a data-driven company that it’s no surprise it’s also using the K.C. move to gain insight into “consumer behavior, particularly when it comes to video consumption and advertising effectiveness,” Bernstein says. As a TV operator, Google presumably has a couple minutes an hour on cable networks to sell ad time, try things with interactive and addressable advertising and monitor results via second-by-second data. More broadly, Bernstein says the data could offer clues to how TV spots work in conjunction with ads on YouTube as well as search and display platforms. This could all lead to metrics that allow Google to persuade major advertisers to launch integrated campaigns that rely heavily on its core businesses. Any shift in ad dollars from TV to YouTube and search would be a boon. So Bernstein suggests Google Fiber will prove to be a loss leader -- one well worth it.