A war of words over mobile broadband speeds heated up last month in Hong Kong. In July – four months after Hong Kong CSL launched the city’s first HSPA+ network with speeds up to 21 Mbps – SmarTone-Vodafone CEO Douglas Li held a press conference claiming CSL’s Express 21 HSPA modem did not in fact deliver the 21-Mbps download speeds claimed in its marketing materials.
As you might expect, CSL is not amused and denies all charges. And now local regulator OFTA is investigating the claims.
In a sense, the CSL/SmarTone feud is a textbook case of the risks of relying on peak speeds to sell broadband services. That said, the feud is also admittedly somewhat mystifying – at least from an industry insider point of view.
After all, promoting peak speeds is nothing new. The wireless sector in general has a long history of marketing data throughput capabilities in terms of theoretical peak speeds, which are (as far as I know) never the speeds that users will get in real life unless they’re the only person online and parked right next to the base station.
In 13 years of covering this business, I have not seen a press release or advert (whether it was aimed at operators or consumers) proclaiming real-world speeds on a wireless system – unless it included a disclaimer in small print marked by an asterisk saying that speeds may vary due to network conditions.
But while this is standard industry practice that pretty much everyone in the supply chain understands and takes into account when making purchasing decisions, consumers are not nearly so savvy, and regulators may be less likely to simply shrug it off as a case of caveat emptor.
The catch is that mobile broadband speeds really do vary drastically under different conditions, especially in a mobile-saturated urban canyon like Hong Kong. It would be asking a lot of cellcos to make more accurate claims when the speeds can vary that widely. As an example, potential speeds on Telstra’s HSPA+ network in Australia range from half a megabit to 8 Mbps. (Source: Telstra’s Next G web site.)
Perhaps a better benchmark is the user experience. After all, most mobile apps – for now, anyway – perform just fine at 3 Mbps or less. So instead of asking if customers are getting the throughput they were promised, the question should be: are the speeds they’re getting impacting the quality of the service?
Granted, UE a more subjective benchmark. But an argument could be made that as long as the customer’s happy with the service, they’re getting their money’s worth. And if they’re not happy, they can complain to customer service and to the regulator, and – in Hong Kong’s notoriously competitive market where mobile number portability has been in place for years – they can take their business elsewhere (if they want to pay the penalty fees of breaking a contract early, that is).
Whether or not that’s a factor in OFTA’s ruling, now is as good a time as any for cellcos everywhere to stop crowing about how fast their networks are and focus more on promoting an awesome user experience.
This doesn’t have to go as far as SLAs (at least not for consumers – enterprise users may be another story, especially once SaaS services gain traction). But we already know that most mobile users don’t care what runs under the network hood or how fast it runs – so long as their services run smoothly. At the end of the day it’s the QoE you’re selling. Put that in your adverts instead and leave the throughput specs to the engineers.