Consider the case of James McGrath Morris, who publishes an email newsletter called “The Biographer’s Craft.” When some of his readers were not receiving the newsletter, he ran his copy through a spam checker. Use of the phrases “young adult” (as in literature), “getting nasty” (referring to a legal matter) and “hot” (in reference to what’s popular in books) were red flags, so to many spam filters, his content was questionable enough to block it from those at the other end of the email stream. Context was irrelevant.

Or consider Mike Fratto, a writer at InformationWeek.com who reported that one of the site’s visitors was having trouble forwarding spam to the FTC’s spam reporting email address (spam@uce.gov) because the forward was — what else? — blocked by a spam filter. The same email, when forwarded to Mike himself, also ended up in the junk email folder.

It’s reasonable to expect some normal emails to fall into the spam filter through unlucky phrasing, but the problem has begun to increase. When Morris asked a professional about lowering his spam-check score, the answer was simple: he just had to censor himself and change any questionable phrases to different ones. But that, as he notes, creates a slippery slope as spam filters try to keep up with more sophisticated junk email onslaughts. “If I surrender those words now,” he writes, “what might I be asked to give up next month?” He muses about ordinary writers becoming stymied in their craft when spam filters trip them up for using phrases like “beastly behavior,” Lolita” or “swelling ranks of investors.” Those who write and distribute email newsletters are fighting a battle of words against software, with strict self-censorship as a result. Writers have had to begun to write for the filters, not for the audience.

In addition to screening for key words and phrases, some filters also check for “bad reputations” from mail servers and IP addresses. If a given server or IP address has been used by spammers, it could end up on a list that makes content from that server or IP address automatically questionable to spam filters. These lists change constantly, and as Fratto notes, it’s hard to get one’s server or IP address removed from the list. Morris experienced this sort of frustrating filtering when his IP address turned up on a list of questionable sources at www.spamhaus.org. The Spamhaus Project claims to maintain the list free of charge to keep email administrators in the loop on spam sources, but Morris hadn’t sent any spam from his IP address. Later, when he checked his IP address again, it was no longer on the list, or on any other lists he checked.

These sorts of restrictive filtering and quiet blacklisting are an obvious problem with spam filters today. What complicates matters is that the person sending the email might not know that their IP address or server is on a black list (the list keepers don’t notify those who are listed; they just maintain the lists), and those who receive an email that is labeled a spam message may never see it, as it’s usually diverted into a junk mail folder or deleted entirely without notifying the recipient. In my own case, Yahoo! puts my spam into a junk mail folder and tells me when I have new messages there, but it’s up to me to wade through the hundreds of spam emails I receive in a day to make sure that no legitimate email is being sent there in error. If I delete the contents of the spam folder, they’re gone, bypassing the “trash” folder completely and going out into the ether. 

And to add insult to injury, many spam messages still slip by the filters and into my inbox.

What can be done to make spam filters and blacklists more aware of context and intent in email? Not much, unless we’re willing to open up the restrictions and allow more spam to reach our inboxes as a result. The price of protection from unsolicited advertising in our email is that some emails will be filtered that shouldn’t be. So which is more important: access to the information, or protection from the noise? For now, that’s an individual choice. You can help the filters perform at their best by putting desired addresses on your email “approved” list, removing yourself from as many spam lists as possible, and checking your junk mail box periodically to see if any legit messages fall through the cracks.

The vast majority of emails sent today are spam… We just have to do what we can to make sure that the filters we use don’t eventually consider ALL email spam.

Sources for this article: InformationWeek, The Hartford Courant, The Spamhaus Project

The answer is as simple as it is frustrating: Spam works.

Up until recently, the general belief was that spam received low response rates, meaning that a million spam messages sent might result in 10 purchases, tops. But on August 19, the Internet security company Marshal released a study regarding the success of spam email marketing. It found that not only were people reading their spam emails, but that 29% of Internet users surveyed admitted to actually BUYING something from a spam email. Perhaps not surprisingly, the most popular items purchased were sexual-enhancement products, adult material, pirated software and luxury items, many of which are knock-offs.

And before we can assume that “regular” people don’t buy anything from spam, we must remember what the spammers are selling. Marshal’s VP of Products, Bradley Anstis, said, “The Internet provides convenience and a degree of anonymity to people who want to buy illegal or restricted goods. It is a black market and spam has become a conventional means of advertising to a willing audience of millions of people who are purchasing from spam.” Worse, most of those who admitted to buying from spam also admitting to buying multiple times from spam. A similar poll by Forrester Research in 2004 found that fewer people (20%) were buying products from spam emails, which means that, if this recent survey is accurate, the percentage of spam-buyers has gone up significantly in 4 years.

Can this be true? Can almost one-third of Internet users actually be misled enough to buy things advertised in spam? That doesn’t inspire a lot of hope for the fight against unsolicited email marketing. After all, maybe the spammers are actually giving people what they want: an anonymous way to buy questionable stuff they can’t get elsewhere.

Before we get too despondent, it helps to keep in mind that the poll surveyed just 622 people, hardly a representative sample of Internet users. In fact, that’s a less than .0002% sampling of the roughly 360 million people using the Internet (according to Download Squad). Add to this the fact that there is some skepticism regarding the veracity of these numbers because of the small sample and the data gathering method in general (Terry Zink’s anti-spam blog, for example, raises questions about whether so many people would actually make purchases from their spam folders). Even aside from the skepticism, one can find instances of general “who cares?” attitudes about the number of purchases made through spam; as Lee Mathews of Download Squadput it, “People decide to buy things from all kinds of unwanted sources: flyers, stickers, magazine insert cards, bumper stickers, board signs at hockey games. Why is it big news that people buy products advertised in spam?” He’s got a point; we assume spam is widely disdained, but every unsolicited marketing campaign can find a handful of people willing to buy from it. 

Still, spam is annoying to most of us, and in recent years, spam messaging has grown by leaps and bounds; the Marshal TRACE (Threat Research and Content Engineering) team found that global spam volume doubled for the year ending in June 2008, with approximately 150 billion spam messages sent per day. That, according to Marshal, accounts for more than 85 percent of the total emails sent around the world, and because it uses bandwidth and resources (and also because it’s increasingly becoming a means to spread malware), it’s a major problem for Internet security professionals and for the public in general. It’s true that most of the spam ends up filtered, so it’s just a small percentage that makes it into a user’s inbox. But once it gets to the inbox, the response rates go up much higher than the 10-purchases-per-million-messages estimate.

Thanks to the advent of botnets (which infect regular people’s computers and allow criminals to send messages without the need for their own servers) and the sheer cheapness of spam messaging (Marshal estimates the cost can be as low as $5-10 US for a million messages), spam is a very lucrative endeavor for those selling less-than-legitimate products and services. But it wouldn’t be worthwhile at all, of course, if people didn’t buy what the spammers were selling. As the website SpamDontBuyIt.orgpoints out, “if you buy products or services from spam email, you are just as guilty as the spammers for creating the problem.” It’s a simple supply-and-demand equation: if spam didn’t pay, spammers wouldn’t do it. Which means we, as Internet users, must take a little responsibility for the ever-increasing pile of spam emails in our inboxes.

Seems obvious, but up to 29% of Internet users don’t seem to understand. Or maybe they just don’t care. Either way, you can take ownership for your own role in the spam problem: Don’t buy what you see in spam.

Sources for this article: Marshal, SpamDontBuyIt.org, Download Squad, Terry Zink’s anti-spam blog