# Solving the spam equation

I have received quite a few responses from IT folks about the cost of spam.

One company sent me its entire analysis in an updated spreadsheet and the figures were fascinating: 8,000 employees with e-mail, a fully loaded per-hour cost of US\$27, and each user averaging 10 messages per day of which 88 percent are spam. Those parameters ring up a total of \$611,111 dollars of lost productivity annually - that's a per-user cost of about \$76 per year.

Add in the bandwidth (\$18,000, with spam taking up 7 percent of the total), storage costs (\$11,600 with 6K-byte average messages and \$2.50 per gigabyte) and support (\$65,280 for spam) and the bottom line is that spam costs this company a total of \$705,748 per annum, or about \$88 per employee, per year.

Now this company's employees are pretty light e-mail users. Just consider what would happen if they were to increase e-mail use to 100 messages per user, per day, a typical volume in many corporations: They would incur 10 times the cost, or about \$6.3 million per annum. That's serious money, with the majority of the expense being lost productivity.

But some of you disagreed with the concept of losing productivity to spam. One reader responded with a revised spreadsheet analyzing the cost of bathroom breaks. He concluded that the total cost of bathroom breaks per user, per year for a 1,000-employee operation would be \$4,484,000 - almost \$4,500 per employee! He suggested that, "The real intent of my parody is to show how small the cost of spam is compared with other time-consuming non-productive work activities. If we keep it in perspective, is the time and money spent on spam filtering really worth it?"

Very amusing, but might I be so bold as to suggest that the comparison is wrong for a couple of reasons. First, bathroom breaks are not an optional expense. Not only do most people function poorly when they are prohibited from taking a bathroom break, they also get pretty cranky (and the law seems to have a few opinions on the matter).

The second and biggest reason is that spam, unlike bathroom breaks, has a profound effect on the signal-to-noise ratio of e-mail, which for many organizations is at least as important as the telephone.

When users receive 100 or more messages per day with a significant percentage of spam, it becomes highly probable that they will miss critical messages.

Look at it this way: Would you tolerate random outsiders doubling the number of records in your databases with irrelevant and erroneous data? No way!

So what are we going to do about spam? There are three things. First, refuse to do business with companies that spam. Because spamming is becoming commonplace, some companies think it is therefore an acceptable way to conduct business.

For example, after the 10th spam from Maxgroup.com I spoke to a representative who told me that the company used a bot to scour Web sites for e-mail addresses, and that spamming was the easiest and cheapest way to build a reseller base.

He proceeded to go through the usual arguments that it would only take me a few seconds to delete the spam (as if it were the only spam) and that what the company was doing was perfectly legal (quite true for now). He offered to take me off the list and yesterday I received spams 11 and 12 from them.

So not only do we not do business with these companies but we need laws to rein them in. That's the second thing. While laws are not the ultimate solution to spam, they are needed to stop companies that are simply taking advantage.

Third, we need technology. There are all sorts of interesting and effective solutions, and next week I'll tell you what you can use.