VoIP

ISPs Tie Regulatory and Voice Strategy Into Softswitch

ISPs can join the gorillas by buying a $170,000 box and taking on the challenges of a vast array of new services.

by Max Smetannikov [December 19, 2003]

The difference between being a run of the mill dialup ISP and a worthy competitor to the likes of Time Warner, AOL, and the RBOC may be no more than $170,000 and a major business overhaul that gets an ISP exposed to all kinds of new businesses, a stressful change but one that many service providers may deem the key to survival.

After a lengthy pause in telecom spending, a number of softswitch manufacturers are pursuing carrier customers with new products. Vendors say the pickings are slim so far because the only carriers buying gear are rural RBOCs and UNE-P CLECs. After tracking one deployment by one vendor I am not so sure this is entirely accurate—small and regional ISPs could be a niche category overlooked by some manufacturers.

The deployment I have in mind is that of sentitO's NEO Services Switch at Wisconsin ISP Merrimac Communications. Traditionally softswitch vendors approach customers like Merrimac with a pitch that goes something like this. Voice is the next big service, and you are stuck with data access—this appliance is easy to manage and deploy and you can be twice the company after buying it.

This pitch looks very different through the eyes of Bart Olson, Merrimac's president. Let's consider Olson's business reality for a second. Even though Merrimac is not a small operation, with its own cable service on top of "regular" ISP services (mainly dialup), it has to match offerings and coverage within its territory with those of much larger competitors—SBC, the local RBOC, Time Warner cable, and large ISPs like EarthLink and AOL.

Keeping up with the big boys is expensive and sometimes not possible. Upgrading to V.92 to offer basic services like incoming call alerts means upgrading all access concentrators, a sure way to dip into red on the dialup business. DSL as a facilities-based business appears a loser proposition given the interconnection prices ISPs get for their DSLAMs and retail DSL prices from the RBOCs.

And forget additional services for a second—adding more dialup customers while leasing more T-1s creates an unfavorable cost curve where margins on those accounts are getting thinner. The solution is almost entirely regulatory: becoming a CLEC gives an ISP like Merrimac access to lower cost tariffed interconnection facilities from the RBOC, as well as the right to colocate and get access to unbundled elements for the DSL service.

The government made me do it
Just one twist here—the FCC and PUCs want to see facilities-based CLECs these days, and data-only CLECs have less eligibility for lower tariffs. Which means a company like Merrimac is basically looking at becoming a voice carrier first in order to get to the next step with its data business.

Offering all-in-one softswitches that basically convert data-only carriers to CLECs with one deployment is the business plan of a range of softswitch makers that I cover—from very large players like Sonus to smaller platforms like those made by Taqua and sentitO, a total of 20 or so vendors. Many of these companies go after two markets—carriers with large backbones and enterprises seeking PBX replacements. Merrimac's example is a good illustration of how much you need to understand about carrier niches to really make these platforms work to customers' benefit.

Olson spent a couple of years looking at softswitches and decided against the likes of Taqua (market leader with rural RBOCs and a good size for large ISPs) because he is not interested is offering POTS telephone service, which many softswitch vendors support. It came down to two vendors offering all-IP telephone service—CopperCom and sentitO. CopperCom's final configuration ended up costing $300,000, where sentitO came in at $170,000. Not only cheaper, sentitO had a unique ISP hook—an answer to the V.92 dilemma.

sentitO's NEO has a built-in V.92 remote access concentrator, which means Merrimac can retire all of its 3Com Total Controls. Next, Merrimac plans to colocate Senitito's switch (which, by the way, is a class five platform) with SBC, and, based on that, get lower interconnection rates for its data traffic. The third stage of deployment is trying to offer Vonage-like service to Merrimac's own cable modem customers, and explore a UNE-P-based DSL offering bundled with a voice service as an alternative to existing dialup customers.

Does Olson wish for the old days? Is he fixing something that is not broken? Hardly.

"I don't know about where you are but over here it is like we have a riot on the streets," said Olson referring to competitors stepping up bundles and feature by feature marketing. "With what we've got now, there is not much to go to besides dialup—and this sentitO deployment we are getting to be on the same level as the big boys."

Olson is hardly the only ISP going through this transition. It all starts with the expensive V.92 upgrade which many small and medium ISPs can't afford, but which the AOLs and EarthLinks of the world completed a while back, believes sentitO's vice president of product management and marketing William Flannagan.

We have to take this statement with a grain of salt—the vendor has only four public customers so far, with Merrimac the only ISP. For this business to work out for sentitO companies like Merrimac have to do a lot of heavy lifting—like making colocation with the RBOC happen.

So far, Olson says things are moving very slowly—six months of negotiations got him only an exploratory meeting with SBC—given that the PUC cycle of getting all the right stamps and permits is done. Haven't we seen it all before with DSL?

Operating issues aside, ISPs do appear a fertile ground for softswitch vendors if you consider Olson plans to pay back for the switch with savings from T-1 rate decreases—making launching additional services less stressful. Essentially, once all the bills associated with new launches are paid, the new services will be gravy on top of this new infrastructure, which was deployed for purely regulatory reasons.

This is important because having talked about ISPs doing voice for so many years this is the first case study that I have seen that at least adds up. Many things could go wrong, of course, but the bottom line is that Olson has to make voice work in some form to maintain the CLEC status going forward (see Triennial Review Part IV: A Game Played Every Year), and now he has the tools to do it.

"If you can go in there and offer DSL and VoIP over the same line with unlimited long distance and additional lines and call forwarding—and throw some more stuff into that bundle—there is market for it," Olson said.

If he is right, we expect that market to emerge in the next two years or so.

—End