Despite having two heads, three arms, and writing science fiction for a living, I am — contrary to rumour — an ordinary human being with ordinary preoccupations. Such as buying mobile phones.
I am not happy with my current mobile phone account from T-Mobile. Perhaps it's an accident of my local coverage, but when half my incoming calls don't reach me, it's time to look for a more reliable network. It's widely reputed that Vodafone focus on network quality more than their cheaper rivals (and I've got a few other reasons for looking at them) so I decided to look into it in a bit more detail. I need a 3G (UMTS) phone, with tri- or quad-band GSM for overseas travel, wifi a nice extra. Bluetooth too, so I can use it with my laptop. I'm unreasoningly prejudiced in favour of the QWERTY keyboard layout and I text a lot more than I talk, so I decided to look at some high-end phones, notably the Nokia N95 (yes, I know it doesn't have a QWERTY keyboard, but it's the current high-end flagship phone and it makes a good benchmark), the Nokia E61i, and the one I hanker after, the Nokia E90 Communicator.
(You will notice the lack of Windows Mobile phones in this list. I've tried it, more than once, but it's quite simple: Windows Mobile makes me break out in hives. I will have no further truck with the spawn of Redmond in matters telephonic.)
(You will also notice a lack of Crackberries. I have an unreasoning prejudice against phones that I can't run a Python interpreter and a word processor on. I don't know why, it's just the way I came factory-configured — if it's got QWERTY and a screen then it is a computer, and I get deeply uneasy when confronted by computers that I can't, in an emergency, hack out a new operating system on: it's like buying a car with no steering wheel or brake pedal.)
(You will notice the lack of the Apple JesusiPhone on that list, too. It doesn't do UMTS, its bluetooth stack is so broken that you can't get an HID-compliant keyboard to talk to it, and the operating system is locked down by default and Apple keep pushing firmware updates that bork the jailbreak exploits. Also, it's only available in the UK on O2 and I still remember them from the bad old days when they were Cellnet. Sorry, but changing your name in the hope that people will forget how crap you are doesn't work on me. And I didn't buy an original Mac 128K either.)
Anyway, this left me with just one cellco to look at, and a choice of phones and tariffs. I favour monthly accounts aimed at small businesses/sole traders; but how to figure out what the best deal is?
It's funny, but the phone company web sites I've looked at all refuse to tell you how much a bloody phone costs until you select a tariff. It's fairly clear that they don't want you to go comparing their tariffs, because then you might learn something useful — such as the total cost of ownership (TCO) and running costs per unit of time/text message consumed. There is some interesting marketing theory behind all this ... and probably several PhDs in game theory. I'm not a marketing expert, but I know what to do. The first rule of looking at a phone company website is to realize they're trying to game you. And the second rule is to game them right back.
Being a bit obsessive, and annoyed at being bamboozled with pretty websites that don't tell you anything, yesterday I attacked Vodafone's online shop with a spreadsheet I went through the site a couple of dozen times, with different tariffs — 12 month and 18 month ones — and fixed combinations of phones, trying to figure out what the best options were.
The first obvious conclusion I reached is that if you look at the total cost of ownership (TCO) of a phone, including both the phone cost and the monthly tariff cost multiplied by the term of the contract, there's surprisingly little elasticity in the bottom line until you get into the eye-wateringly high usage tariffs. The TCO for a sample phone on 18 month contract varied by only £102 between the Talk 75 and Talk 500 tariffs (75 included minutes and 100 included texts per month, versus 500 minutes and 1000 texts per month). The same pattern held on 12 month contracts, with a £60 spread. Which is, frankly, ridiculous, because you get so few minutes and texts on Talk 75 that the actual cost per minute is nine times higher, and the cost per text is eight time higher than on Talk 500.
But it's not a good idea to go up a level to Talk 800. Do that, and suddenly there's a big price jump. You get more minutes (but no more texts), and the end result is that the cost is barely any cheaper per minute used (and no cheaper per text) than on Talk 500.
What I had discovered looked weirdly like a classic bathtub curve — only plotting price against contract time, rather than the more familiar failure rate against time. It's a familiar curve: airline seat price allocations often follow the same distribution. At one end of the curve, you've got the chancers who want a flashy phone but no commitment to use it. Typically they'll sign up for a short, cheap contract with an expensive phone. Fashion victims, in other words. The cellcos are set up to recognize and fleece them, however. At the other side of the curve you've got the gabby heavy users, and they're going to throw money at you whatever you do, so you might as well take it. In between, you've got a highly price sensitive market, which you want to encourage to use their phones more (and graduate into being heavy users), so you dangle some promising discounts in front of them, weighted towards the heavier tariffs.
(Airline seats for long-haul flights: if someone books a flight six months ahead of departure, it is a Big Deal to them, so they value it, so you can price it high. If they book at two day's notice to go to Aunt Irma's Funeral in New Zealand, it's a coercion purchase, so you can price it high. In-between, there's a trough where people have time to pick and choose which carrier to use ... so seat prices are at their lowest in the period 8-12 weeks before departure. It's the same bathtub-shaped curve.)
In fact, the sweet spot on Vodafone's tariff curve (in the Anytime business packages) seems to be Anytime 500 on a 18 month contract. (By the time you hit Anytime 500 on 12 month contract, costs are beginning to rise; and anything less than Anytime 500 on the 18 month contract is in the "soak the trend-follower" category.)
And there's my second point: 12 month tariffs are weighted on the assumption that you're a trend-follower and may be part of the general customer churn. They invariably have a much higher total cost of ownership than the 18 month tariffs. How much higher?
Cost for a Nokia E90 with Vodafone, 12 month contract, Anytime 500: £187.23. Cost per month: £34.04. Cost of phone plus twelve months: £595.71.
Cost for a Nokia E90 with Vodafone, 18 month contract, Anytime 500: £127.66. Cost per month: £29.79. Cost of phone plus eighteen months: £663.88.
Read that again: the total cost of a twelve month contract costs nearly 90% of the price of an eighteen month contract. If you take the twelve month contract and stay on it for eighteen months, you'd be paying a whisker under £800. The mark-up for going for a short contract is huge; they're counting on your natural reluctance to be locked in for an extra six months to lead you to pay hugely over the odds.
(Want a twelve month contract? You might as well buy an eighteen month contract — if you decide to switch telco, the break-even point is thirteen months. At that point you might as well buy a new phone, set call divert on your old number, take the old sim out and cut it up so you can't run up any additional charges, and get going: you're still ahead of the game. The system is loaded insofar as it relies on customers fixating on the contract lock-in period and not realizing that they can "buy themselves out" at any point by cutting up a SIM and making a note on their calendar to remind them to close the account when the lock-in expires. And on most people not running the total cost of ownership through a spreasheet before they buy.)
They're also counting on your natural reluctance to pay for a "fatter" sounding tariff than you need to drive you to pick a "cheap" one like Talk 75, which is disproportionately more expensive.
Anyway, the moral of the story is: if you're planning on buying a new phone, it pays to hit your phone company website with a spreadsheet before you sign anything. And don't take anything for granted until you've run the figures. The TCO per minute for a phone purchased on the superficially cheap-looking Talk 75 tariff turns out to be two and a half times higher than the TCO per minute for Talk 200, and a ridiculous seven times higher than on Talk 500. They're gaming us. You should game them right back.