Would you pay a postage fee to send email?
Fri, Feb 17 2006
Sorry I haven't done an eBusiness News update for more than a month: it's been a crazy time recently with many new projects, a trip to New Zealand for the annual linux.conf.au, about half a dozen media appearances, and working frantically to meet a publishing deadline on my next book.
But the world doesn't stand still and there have been some interesting developments recently in relation to spam control. The hottest topic in the last week or two has been plans by AOL and Yahoo! to implement a "pay to send email to our users" scheme, but before I explain it you need to understand just a little bit about the economics of spam and email in general.
Because you don't have to pay for a stamp every time you send an email it's easy to think that sending it is "free". Of course everything has at least some cost even if it's negligible, and in the case of email it's in the cost of maintaining mail servers, running spam filters, paying for internet connections, and all the other ancillery paraphenalia that goes along with it. If you run an internet connection to your business and it's used mainly for email, you could make a reasonable estimation of the cost per email by dividing your monthly access plan by the number of emails you send. If you're sending a lot of email it probably works out to somewhere in the region of a few tenths of a cent to maybe a few cents per message.
In any case it's so small that you could consider it to be negligible if you send a few hundred or even a couple of thousand emails per month. At that level it's just not even worth measuring.
If you're a spammer it gets even cheaper per message because they typically use armies of hijacked machines to send out huge quantities of mail on their behalf. Delivery runs in the range of tens or hundreds of millions of messages is typical for a spam run, at a cost in the region of thousandths of a cent per message.
Compare that to printing and delivering postal spam. Here in Melbourne it costs in the region of several hundred dollars to deliver to 5000 households, plus the actual costs of printing the material which could be another few hundred dollars. That amounts to somewhere in the region of say 6 to 10 cents per recipient.
So spammers are working the statistics and the economics of the situation. They know they can get by with a ridiculously small response rate because their cost per message is also ridiculously small. Everything works in their favour.
But what would happen if there was an "email delivery fee" or similar, with providers refusing to deliver mail to their users unless you pay them for each message you send to them? What if the delivery fee was very small, say 1 cent per message? To most users it wouldn't impact them at all. One cent per message if you send 100 emails / month is a whole $1. To a typical user it wouldn't even be a glitch on their access fee.
But think about what that does to spammers: instead of sending messages at a cost that's so small you need to use scientific notation to represent it, the cost goes to one cent per message just like for regular users. If you're sending 100 million spam messages in the hope of getting a few dozen responses that one cent suddenly becomes a really big deal. The entire economic basis of spam goes right out the window, putting it more in line with regular postal spam. Sure, you could send it if you wanted to: but it would have to be much more targetted, and it would have to produce a direct result that makes the exercise economically viable.
Back to Yahoo! and AOL, and that's basically the concept behind their latest move although they haven't actually taken it as far as the scenario I explained above. They're not saying you have to pay them to deliver email to their customers, but what they *are* saying is that if you pay you'll be able to bypass their normal spam filtering and get your messages straight into their user's inboxes. It's more like a "delivery guarantee fee" than just a "delivery fee". Unpaid email is (maybe) still delivered, after going through all their usual spam countermeasures. But if you want to be sure your email gets through, you have to pay.
Overall it's a very interesting approach. For years I've said that spam won't stop until the economic basis for it changes. Legislation is useless. Technical countermeasures are just an arms race. The only thing that will really stop spam dead is if it costs real money out of the pockets of the people sending it. The approach taken by Yahoo! and AOL could do exactly that, and only time will tell whether it's successful.
Personally I don't like the fact that this approach penalizes all the "good" users out there in order to hurt the "bad" ones. I don't like the fact that it could make email-intense but legitimate operations such as newsletters sent by non-profit organisations too expensive to maintain. In short, collateral damage sucks. And if this approach becomes widespread and generally accepted there will definitely be major collateral damage.
But maybe that's just the price we have to pay to make spam go away for good.