A Home on the Range

Asking the right questions about local Internet service providers will get you the best service in the long run.

by Joel M. Snyder

Cyberspace is a shrunken, shriveled planet sucked dry like an old fruit by a few nerdy losers' need to be fatuously wealthy.
--Tony Hendra (GQ, October 1995, p.79)

The great Internet shakeout has begun. Hundreds of poorly capitalized, poorly run Internet access and service providers have fallen by the wayside, victims of their own success and their rivals' willingness to operate at a loss. In many markets, the competition has become brutal, with Internet access pricing dropping by the hour. I picked up a computer newsletter in Washington, D.C., this morning, and 17 Internet providers had advertisements, each offering deals too good to be true. Most of them probably were.

For some people, this low-price access is a bonanza. If you can get unlimited Internet access for $9.95 a month, it's worth the frustration of discovering some evening that the traditional busy signal has changed to a "this number has been disconnected" message every few months. Finding a new provider once a quarter becomes a fun exercise in market research and information gathering. Quick hint: Look for the phrase "low-price leader."

For others, like myself, stability counts for a lot. I don't mind change, but I prefer appropriate change. My e-mail address, the phone number I dial, the magic incantations I need to make my laptop connect up--I don't see a lot of benefit in changing those any more often than we change U.S. presidents.

Caveat Emptor

If you'd like to find a provider who's going to be around when you want it, try to get the answers to the following questions before signing on.

1) How likely am I to get a busy signal? If your provider is truly on the ball, it gets statistics from the phone company which tell it how often its phones are busy and how many times calls have been rejected because of busy lines. This amount of knowledge and coordination, however, is unusual. Don't make the mistake of asking how many phone lines the provider has; that's a meaningless number unless you also know the demand for phone lines.

You can try a different tactic by asking the ratio of customers to phone lines. If it's 5 to 1, you're in fat city. At 10 to 1, things are looking a little grim. Somewhere in between is probably a good balance between economy and convenience. If the ratio is more than 10 to 1, stay away: This provider has you on a one-way road to busy-signal city.

Of course, you will want to verify this information by dialing the access number yourself a few times a day for a week before you sign up. Count your busy signals: If you get more than one or two, stay away.

2) How likely am I to get help? You may think of yourself as an Internet guru, but everyone needs support once in a while. Your provider should be there when you have a question. Ask about support hours and make sure that the people on the other end of the phone know what they're talking about. I'd rather have to wait until the next morning to get an intelligent answer on the first try than get a hold of a sleepy know-nothing at 3 a.m. after 20 minutes on hold.

Try this experiment. Call whoever's on duty and say this: "I'm having trouble running traceroute. It looks like your router is limiting ICMPs. Is that right?" (If you really want to sound erudite, pronounce ICMPs as "eye-comps.") If the person at the other end says, "Yes," sign up immediately. If he knows what traceroute is but doesn't know the answer, that's average. If your question gets met by the telephonic equivalent of a blank stare, you're talking to an underpaid, undertrained headset and are better off conversing with your dog--and taking your business elsewhere.

Any provider who insists that you must send in support questions by e-mail to some anonymous address is an immediate blacklist candidate. E-mail is a nice option, but making it a requirement is a sign that the smart employees are in hiding and the confused ones get sent to the phones.

3) How much does the company know about running a business? This is a tough one to gauge, but there are some tell-tale signs. Watch for spelling errors, a sure sign of a business run by amateurs. Operating a successful startup doesn't require a college education, but it does require a certain attention to detail. Businesses that care about the image they project are much more likely to be around a year from now.

Advertising is another place to peer into the soul of your potential provider. Any business offering something too good to be true, such as unlimited access for 10 bucks a month, is betting you won't use what it has sold you--or has some hidden costs that will get you in the end.

Longevity, of course, is the best indicator of future success. But just because someone knows how to run a retail computer store doesn't mean he knows how to run a telecommunications service business. Another good exercise is to try to talk to the president. If you get through to his secretary but never can quite convince him to let you speak to the top dog, that's an excellent sign. If, on the other hand, you get a 20-minute sales pitch from El Jefe about your SLIP account, someone is not allocating his time very well.

4) Is the infrastructure sufficient to support you? The Internet is not made of expensive technology. However, there are limits to just how cheap you can go. If a visit to the premises is possible, look for good signs like equipment racks, neatly laid-out cables, and hardware manufacturers you have heard of. On the other hand, piles of modems, servers with their side panels removed, and Linux are sure signs of someone cutting costs too close to the bone.

It's not necessary for a provider to have backups for all equipment, but it's nice to know if it has uninterruptable power supplies, multiple lines to the Internet, and spares for some crucial gear. Zero downtime is probably too much to ask for, but any provider that cares about providing the best service will be happy to share with you statistics about system and network problems, downtimes, and upgrades.

5) Does anyone else like this provider? Business references are the best way to get inside information on a company. Post a note to a local newsgroup or bulletin board about the provider you're considering. (Warning: Do not post to a nationwide or, even worse, worldwide newsgroup!) You should be able to find customers who can tell you how well the provider has really been operating.

Be careful to find people who share the same standards you do. I ran into an old friend who raved about how much she loved her Internet provider and claimed it was the best in town. Why was she so happy? Because this was the first provider she'd ever used that "only lost a few mail messages a month." Thanks, but no thanks.

End Results

All of this research sounds like a lot of work just to get you connected to the Internet. Honestly, shopping for a new Internet service provider combines the intense excitement of finding a new dentist with the thrill of buying tube socks.

Avoiding the whole exercise is certainly worth some effort. Frustration, especially when I'm on a deadline or need to get some information online, is something I can do without. If it takes a day of work to find a firm that is not going to disappoint me, that's a good investment. *IW*

Joel Snyder is a senior partner at Opus One in Tucson, Ariz.
Reprinted from Internet World magazine Vol. 7 No. 3, (c) 1996 Mecklermedia Corporation. All rights reserved.