Pages

Pages

Sunday, June 20, 2010

Putting a value on customer data with reference to offload

There continues to be a ferocious discussion about AT&T's new data plans - and also, as I commented a couple of weeks ago, about the fact that its femtocell traffic counts against a user's monthly data cap.

Broadband Reports gives airtime to AT&T's rather unconvincing explanation:

"3G MicroCell is primarily intended to enhance the voice call quality experience in your home," AT&T's Seth Bloom tells us. "While it can carry mobile data traffic, that’s not the primary solution it provides," he says. "Wi-Fi is the optimal solution for home mobile data use. We encourage people to take advantage of Wi-Fi capabilities - that’s why all of our smartphones include Wi-Fi radios, and usage on Wi-Fi doesn’t count against your mobile data usage bucket."

Firstly, I rather pity Cisco, AT&T's supplier, for its client's diminishing of its carefully-engineered data offload capabilities. I suspect it hadn't expected to be relegated to providing voice-only coverage.

But the interesting thing here is that, unwittingly, AT&T is putting a value on collecting customer data. We all keep hearing that "subscriber data management" is a big deal, and that operators urgently need to find ways to monetise their customers' "social graphs" and aggregated web access records and behaviours.

But by telling its customers to prefer WiFi to femto for data traffic, it is sending out the message "Actually, customer data isn't really worth that much after all. They might as well connect directly to the Internet via WiFi - we're not interested in collating that usage for data-mining anyway".

Either that, or AT&T's core network is weirdly expensive, and they'd much rather have all the traffic bypass it and go straight to the web by the most direct route.

Again from Broadband Reports: Bloom goes on to insist that the Microcell "uses our core wireless network just like a call placed while driving down the highway uses the core wireless network." "The only difference is how that data or call gets there – via a MicroCell connected to a wired broadband connection instead of a cell tower."

Which is funny, because I always thought that the majority of an operator's cost base was the cell tower and radio network, rather than the core. And therefore "the only difference" for femto vs. macro is really just a cost-saving.

My gut feeling is that AT&T has either not thought this through - or has major problems in changing its charging / billing structure to accommodate femto vs. macro traffic differences. Perhaps it thought nobody would care.

3 comments:

  1. For femto cells, AT&T has to invest in network controller component in its data centre that integrates traffic from Femto to the packet core.

    AT&T is cunningly trying recoup this minor investment (compared to base stations and transmission) and send wrong message to customers - don't buy my Femto cells as you are better off on Wi-Fi.

    What it needs is a clear strategy for femto cells - acquire a ISP and then promote femto cells with free traffic in order to sell femto cells and lock in customers and also get their data.

    ReplyDelete
  2. For femto cells, AT&T has to invest in network controller component in its data centre that integrates traffic from Femto to the packet core.

    AT&T is cunningly trying recoup this minor investment (compared to base stations and transmission) and send wrong message to customers - don't buy my Femto cells as you are better off on Wi-Fi.

    What it needs is a clear strategy for femto cells - acquire a ISP and then promote femto cells with free traffic in order to sell femto cells and lock in customers and also get their data.

    ReplyDelete
  3. jonathan8:25 pm

    AT&T's decision to count femto traffic against the customer's overall usage limit is very difficult to understand. The cost basis for femto calls is vastly lower and both AT&T and the customer know it...AT&T explicitly and the customer intuitively.

    As all Femto traffic is handled through a separate gateway in the operator core it would be very easy to rate it differentially (much lower). As I understand it AT&T had (in the past at least) a byzantine billing system so possibly they lack the capability to easily rate and bill Femto traffic. Possibly. Personally that seems unlikely though.

    Most likely is that is it is either a poor marketing decision or a test run to see what happens.

    I work in telecom myself; I see far more value in offloading macrocell traffic than in trying to make a case for what is a dubious value proposition. With wide enough femto deployment AT&T could offload a major chunk of network load (a large percentage of mobile calls and data use is from homework where femto fits perfectly), go back to unlimited usage plans (reduced billing system complexity and CAPEX/OPEX, reduced customer Care OPEX), and compete better against Verizon (unlimited is cheaper/easier to market and very attractive vs. any bucketed plan).

    IMHO: sell the Femto gateways at positive margin, make the usage free, encourage femto use, switch back to higher cost but simple/fixed/unlimited plans

    ReplyDelete