Death, taxes and spam

I turned off the trackback feature of this blog a long time ago. It’s not that I don’t like trackbacks. It’s just that the spam became too much – hundreds and hundreds a day. And, let me tell you, trackback spam is the worst. It seems to be produced in hell by a gang of very efficient and very warped demons.

I also had to start forcing people to register before commenting. I know it stinks but, again, I can’t deal with the spam (even with a filter). I turned off the comment registration requirement a few days ago just to see what would happen. I had 50 spam comments in ten minutes. I noticed, too, that their quality is up in general – quality in spam being measured by the degree of deviousness. The new thing seems to be to start with a couple of sentences about something in the news in order to slide by the filter. It works pretty well, too. Congratulations, guys.

And email spam? I’ve been under the impression that it’s been getting a lot worse recently. Hundreds get caught by my spam filter every day, and a few dozen more get through. Stock tips, mainly. Relentlessly annoying. Apparently my impression is right. A new study says that nine out of every ten email messages today are spam. When I first saw that figure, I said to myself, in a loud voice, “No way!” Then I thought about it a little more, and I shouted back at myself, “Way!”

Nine out of ten.

That’s a lot of spam.

The New York Times reports that spammers have successfully defeated the four main strategies used by filters: evaluating where a message is coming from (useless with botnets that take over innocent people’s PCs), checking the text for tell-tale keywords (text has been replaced with images of text), scanning for multiple copies of the same message (spammers use software to automatically alter a couple of pixels in every message), and checking embedded links (you don’t need a link if you’re just trying to get someone to buy a stupid penny stock).

So how many suckers are actually going to buy some obscure penny stock based on a ridiculous spam tip? The answer: enough. The Times gives the lowdown:

Spammers buy the inexpensive stock of an obscure company and send out messages hyping it. They sell their shares when the gullible masses respond and snap up the stock … Though the scam sounds obvious, a joint study by researchers at Purdue University and Oxford University this summer found that spam stock cons work. Enough recipients buy the stock that spammers can make a 5 percent to 6 percent return in two days, the study concluded.

It’s all about zero costs and unlimited scale. Given enough eyeballs, no scam is too shallow. If you want a lesson in the new economics of abundance, there’s no better place to look than at the business of spam. It’s all there.

7 thoughts on “Death, taxes and spam

  1. John Baschab

    The spam problem has been a back-and-forth war of escalation over the past few years. We filter about 10,000 email boxes. Here are the spam stats that I received last night (nightly email from our filtering box):

    Junk – 148,203

    Legit – 21,660

    14.6% of the inbound messages are legitimate! I am always astounded by this stat. I don’t know how anyone does without a spam filter.

    The top spam recipient in our spam filter receive a whopping 537 junk messages, and 3 legitimate messages.

    A while back an article (WSJ) I think, paraphrase: “Spam will always be around so long as it works.”


  2. IanKemmish

    I think the New York Times article is a little behind the times, if you’ll pardon the pun. One particular pump-and-dump scammer has been forging my email address for some years now and these scams make a LOT of money. You can see a history of nearly 50 scams in two years at These started at just a few tens of thousands of dollars per scam, but recently (just since PartyGaming was outlawed in the US – is this relevant?) some scams have resulted in more than ten million dollars changing hands. Part of this will be the result of scammers hacking into online trading accounts and emptying them by buying worthless shares, but a lot will be from people responding to both the spam and the price spike from the hacked accounts. When spam is this profitable, you’ll never win until the law enforcement agencies start to take it seriously.

  3. Thomas Otter

    Hello Nick,

    You may be interested in the Frieder and Zittrain paper on touting. (Frieder, Laura and Zittrain, Jonathan, “Spam Works: Evidence from Stock Touts and Corresponding Market Activity” . Berkman Center Research Publication No. 2006-11 )

    Herewith a quote:

    We find a significantly positive return on a day of heavy spam touting of a stock, along with the day preceding our detection of such touting. Volume also responds positively and significantly to heavy touting. Returns in the days following touting are significantly negative. Though we have no way of directly knowing if the spammer actually has holdings in the spammed stock apart from the spammer’s own admission, when it appears, and the surmise that some pecuniary motive inspires sending the spam, the evidence accords with a hypothesis that spammers tout stocks in order to increase trading activity and price enough to unload their positions at a profit. Selling pressure on the part of the spammer then results in negative returns following heavy touting.

    Spamhaus deserves more PR and support. It is amazing that this 30 person company is so vital in the fight against SPAM, yet it receives almost zero funding…

    I discussed this briefly here…

Comments are closed.