Counting the Hotspots that Count

By Adam Stone

January 15, 2003

The number of Wi-Fi public access hotspots in North America is a number that is constantly in flux as the many providers and aggregators struggle to show off their coverage -- but the number of locations probably does not matter much in the long run.

Counting hotspots should be easy, right?

Wrong.

As network aggregators and network operators strive to position themselves within the Wi-Fi market, the pressure is on to show big numbers. Whoever has the most hotspots will be best positioned to capture market share, or so the logic goes. Trouble is, reliable numbers are hard to come by.

"There are about 2,000 [public-access] hotspots out there -- and that's everybody," said Iain Gillott, an independent consultant with iGillott Research in Austin, Texas. "Yet I have heard everything up to about 5,000 hotspots. It is just all over the place."

It's not that the aggregators are making up their numbers. Analysts say the big problem here is duplication. Take for instance Wayport, which has hotspots in many hotels. Now, a user who subscribes to the aggregator Boingo can also access Wayport hotspots, because of agreements between Wayport and Boingo, explained Chris Kozup, an analyst at Boston-based Meta Group.

If both Boingo and Wayport count that hotspot in their tally, the numbers will quickly get skewed. Apparently, that is just what has been going on. "There are so many different relationships and crossovers in terms of roaming agreements," said Amy Cravens, industry analyst at In-Stat/MDR in Scottsdale, AZ.

To help clear the air, Cravens recently put out a comprehensive report tallying up hotspots among the major WISPs and aggregators. By her count, the November numbers for the top three came up as follows:

  • Boingo: 700 operating Wi-Fi locations
  • GRIC: 650, with 1,000 expected by year's end
  • iPass: 375 Wi-Fi locations

Plus, as T-Mobile pointed out when we first wrote about this report, that doesn't even count the ubiquitous Starbucks Coffee shops with Wi-Fi, which they say was up to 2000 at the end of last year.

This business of counting hotspots is a hot issue. In a typical press release, for example, GRIC went out of its way on December 3 to state that its "total broadband access locations -- both wireless and wired -- topped the 1,000 mark for the first time." This was not a side note: Rather, the hotspot count was the subject of the release.

So why all the fuss about numbers? After all, 2,000 (or is it 3,000?) hotspots across all of North America doesn't amount to much. Yet the operators have taken pains to assert their numbers, each claiming their hotspot count as a competitive advantage.

Some analysts says it is because the operators have little else to count right now.

"Here is what it boils down to: There are not enough hotspots out there. There is not enough coverage out there. It is like cellular in the early days, when people used to tell you they had so many towers. They were flexing muscle to show how big they were," said Gillott.

"Now that they have much better coverage, they don't talk about towers any more. In fact they keep it secret," since the tower count could be perceived as competitive information, he explained. "Now they talk about customer service and how many customers they are adding and how many minutes you can have for $30 or whatever."

As Gillott sees it, the quibbling over hotspot numbers is simply an indication of how far the industry has not come. Plenty of others back his view.

"To get subscribers, you want to paint the picture that you have a broad coverage, and that is what these numbers are all about," said Kozup. He added that the hotspot count at this point merely serves as a diversion from what will ultimately become the real competitive issue among the operators. "Numbers are one thing, but in my mind what is important is the business quality of the location."

What kind of traffic does a location encounter? It is generally understood that a hotspot in a coffee shop has more value than a hotspot in, say, a grocery store. In this sense, the mere counting of hotspots has very little value -- or, as some would see it, no value at all.

"It does not matter how many hot spots you have," said Eddie Hold, a wireless analyst with research firm Current Analysis in Sterling, VA. "If you have hotspots in the most irrelevant places, no one is going to use them. So it is the quality of the hotspot that counts. If they really wanted to take the argument to the next stage, then the question is: How many users are using that hotspot, and how much revenue is that hotspot generating?"

To Hold's way of think, the fact that they are not saying this only shows that "it is really early days, and people are still just getting into this idea."



Comment and Contribute
(Maximum characters: 1200). You have
characters left.