Would you buy a complex wireless data management system from Carphone Warehouse?

by Guy Kewney | posted on 19 April 2002

There are two ways to introduce a new technology. One is to offer a premium service and charge high, cutting prices as volume goes up. The other is to give it away, in the hope people will get addicted. Which is appropriate for GPRS data over wireless?

Guy Kewney

On the same day as Mobitex provider Andrew Fitton cast doubt on the effectiveness of GPRS as a data carrier for users of BlackBerry devices in the UK, the FT put the wind up the GPRS community with an interview with Carphone Warehouse chief Charles Dunstone, who has criticised the Emperor's New Clothes; he says: "Only 0.6 per cent of the GPRS-enabled handsets sold in our stores have been activated."

Everybody has had their two cents worth in responding. Typical of the postbag here at Newswireless Net was this from Fredrik Naslund, MD Europe for mobile infrastructure specialist WaterCove Networks: "We believe that retailers like Carphone Warehouse must rethink their selling strategies if next-generation mobiles are to be a success."

"Too much emphasis is placed on selling phones simply as fashion accessories - consumers cannot be expected to invest in GPRS services if people aren't told what they can do with them at point of sale," said Naslund. "All retailers should encourage potential customers to fully test the phones and staff should be able to demonstrate all available services."

Naslund concluded: "GPRS is perceived as too difficult to set-up and use, but that is only because vendors are not providing phones and services that work straight 'out of the box'."

Dunstone blames tariff levels: "The problem is that you have to choose a more expensive tariff to use GPRS and people just don't think it's worth it."

Hm ... and perhaps, there are even more problems? What lessons can we learn from the past in the wireless data industry, for example? And what useful parallels are there?

First off, let's try to remember that huge success, SMS. It actually started right at the beginning of the first PCNs, with Orange; it was always part of the GSM specification. It was, nonetheless, years before it hit critical mass and became the fashionable way to chat up members of the appropriate sex. And it was dead cheap.

The thing to remember, surely, is that nobody knew what to use it for.

What about the "learning curve"? It's a fact which is now taken for granted that you introduce something which is expensive to roll out - because has a huge investment behind it - to the wealthy user first. It's also a fact, which seems to have been forgotten entirely, that it has to be something they all want, desperately. Typically, a camera with twice the pixel count, or a computer with three times the disk space, or a phone with four times the battery life.

What is a "loss leader"? Well, there are lots of examples of products which were given away at first, and then, when people couldn't do without them, the price started to rise. Just watch this Web site ... and I'm sure you can think of others!

Which model applies to GPRS? Clearly, there was no huge, pent-up demand for slightly faster SMS at hugely expensive rates, with dauntingly difficult installation, setup, and user learning problems.

There are companies like Peramon and Commtag and Sirenic - who will undertake to set these things up for you, including managing the email side, and installing servers in your office LAN - and charging you like a wounded rhino as well; and there are also people doing sophisticated integrated packages like Information Builders who will provide a full Web Server solution to several business process problems for mobile executives, charging like a herd of wounded elephants.

However much value these may be for money, they aren't a quick sell.

It isn't hard to see that people buying these solutions won't all appear on Carphone Warehouse stats: you won't go to the Stack-em-high Trendyfone store for a complex business solution.

And it's equally simple to understand that they don't rush out and buy a thousand devices in the first week they're available, either. They do a trial, they go back to the Board, they set up a committee to evaluate and negotiate; they go back to the IT director with a request for estimates of the amount of new hardware they require; they dispute the resulting figures; they then prepare a six month RoI spreadsheet and go back to the Board ... and about a year later, you start to see revenues.

But even if you allow for all this, Dunstone is probably right. GPRS is still written about as if it provided really fast access. It doesn't. It's written about as if it is what it will be in a year's time, if the networks install an awful lot more equipment. It isn't and they probably won't.

You can get bursts of high-speed data, yes. You can get as much as 40K a second, for a second or so. This is a bit like saying that you can cross the M4 motorway on foot in about ten seconds if you pick your moment.

But if you want to send and receive more, you get your share of the gaps in the rest of the traffic. One lunatic may be able to sprint across a motorway; but if you want to take a platoon of soldiers across, they better be armed and able to shoot the tyres out at 500 yards; and if you have to escort a convoy of schoolkids ... well, not only are you going to lose some of them, but it's going to take quite a while.

GPRS is a stop-gap. It's a way of testing what sort of software we might want to use on networks which give us 128 kilobits of data traffic per second, at trivial extra cost.

One day, nobody doubts, this will be what we take for granted. We don't know if it will come through a hugely expanded GSM phone network (well, yes, we do; it won't) or through an ad-hoc federation of WiFi and Bluetooth "hot spots" or through a cheap and efficient UMTS third generation network, or some other invention; but the trend to higher bandwidth and lower cost is probably unstoppable.

To try to sell GPRS as if it is already this high-speed, cheap data link, is to guarantee failure.

The full FT interview with Dunstone is in their archives.