An inside view of junk eMailing: How spamming "works." Well...sort of works.

The economics of spamming.

Thanks to a research team at UC San Diego--with support from the International Computer Science Institute--we have a better understanding of the economics of spamming.  In fact, we now have an equation that shows how it all goes down--down to what amounts to a pretty small pile of money in the scheme of things:

G = [E] x [M] x [D] x [F] x [V] x [B] x [P]

Where...

G = Daily gross revenues [about $7,000]
E = Number of spam emails the botmaster ordered the zombies to send per day [1.7M]
M = Number of middlemen computers [550]
D = Percentage of the messages that are delivered [23.8%]
F = Among the messages delivered, the percentage that make it through spam filters [varies]
V = Among the users who receive the message, the percentage who visit the site [0.0127%]
B = The percentage of those who buy something [0.00266%]
P = Average order price [$100]

The research and how it worked.

Here's how this formula came about:  A guy by the name of Chris Kanich and associates intentionally infected eight computers with software they'd found in the "wild" which was already doing spamming by infecting a network of computers into eJunkmail-generating zombies.  What they did was simple:  They merely changed the orders, shifting the operation to serve their own research purposes.

After Kanich and his team worked their transformation, they introduced spam ads directing responders to a site that Kanich's team had erected; the site had the appearance of an ordinary internet pharmacy.  However, instead of taking credit cards for the offered items, the site produced, in the end, an error message, but not before counting the clicks and order behavior of the hapless buyers!

As you can see, the expectancies for buyers' responses are very, very small [i.e., only 0.0127 percent visit the site and the percentages of those who do, only 0.266 percent actually buy]; what's more, these diminished values only become economically interesting when the starting number of "hits" in the first place end up being very large!  

Why spamming and marketing are very different animals.

If you push the numbers, 1,700,000 spam messages were sent; about 405,000 were received at people's computers; if spam filters kept out 50% of those messages, that means that about 200,000 actually saw the junkmail item in their inbox; only 25,692 of those 200,000 people decided--for whatever crazy reason--to go to the spammer's site; only a tiny fraction of those ended up placing an order.  So what's the yield of disturbing about a half-a-million people to start with?  It's seven actual transactions.

And, that's the problem with why spamming is a bad thing and--here's my important point--not really a part of marketing, per se, despite it's reputation for being so.  Let me explain...

Sound marketing theory and practice--and that of modern business more generally--are predicated on starting-then-establishing transactional relationships [i.e., reasonable, feasible selling-buying encounters] that are targeted, that is to say, ones that are focused on specific markets with specific needs, interests, and problems; in other words, sellers matched with likely prospects.  Spamming completely ignores that key principle.  

Riding the system for free.

But because eJunkmail--and spamming more generally--are quite literally free--there are people who act as the equivalent of parasites on the free-enterprise system to harass millions of people, only to consummate a transaction with only a handful of them.  

You can see the same thing operating in computer-generated random telephone dialing:  over a weekend, thousands of people's lives can be interrupted and inconvenienced by an electronic, automated call program, all so that on Monday morning a enterprising but misguided local business can follow-up on one or two lousy leads.  

The only thing that spamming--or, for that matter, computer generated phone calling--have in common with marketing is the technology that's used--and little else.  Beyond that there's little basis to believe that they belong in the same business category.  It's not even close. 

[The specific research cited here can be seen on the blog's Facebook Page--check it out of you'd like to get the story as reported by Wired Magazine.] 
_______________________________________________________________

Follow Keith's biz blog on Twitter for updates and see more of what he's reading about on his Facebook Page. If you are inclined, you can write him at kmurray@bryant.edu.