Establishing the Cost of Spam
(Part two of a series)

In the first part of this two-part series, IDC looked at the different kinds of technologies that are used to manage spam within an organization. In this part, we take the “next step”, and illustrate ways of quantifying the impact of spam – and the benefit of spam filtering technology – within an organization.

In response to the deluge of spam, IDC developed a tools-driven methodology to assess the business impact of spam within an organization. The “Spam Calculator”” identifies the cost of spam within an organization, and the impact that it has on employee attitudes and perceptions.

The first step in developing a Spam Calculator is categorize the ways in which spam impacts the operational costs of an organization. The most important source of spam-related cost (or lost revenue) is lost productivity – time spent handling spam messages. Our research has shown that employees handle spam is several distinct ways, and that each method has a different productivity-related cost to the employer organization. Generally speaking, it is more time-efficient to batch spam messages than to deal with each one individually; also, employees who take more than one action with a single spam message (for example, they read it, and then delete it) lose much more time to that message than employees who handle it once.

Although lost productivity is the most significant spam-related cost within most organizations, it is not the only cost. Some organizations have a significant second “hard cost” category – the technology and staff costs associated with deploying and managing spam filtering systems, which are introduced to reduce the productivity drain caused by spam. Within these organizations, we are able to develop an important set of financial metrics: the ROI, payback period, and net cash benefit of the filtering technology. This provides us with a unique perspective on the economic value associated with the deployment of an infrastructure technology.

The Spam Calculator” in Action

In the publicly-available case study* resulting from use of the Spam Calculator”, IDC worked with an organization that is using spam filtering technology. In this case, the system used is a heuristic system, meaning that it “learns” (by means of employees forwarding spam messages to the filter) how to block spam. Typically, these kinds of systems grow more effective over time.

Research with the users within the organization showed that most end-users were able to batch messages, either to delete them, or to forward them to the spam filter and then delete them. This was a particularly time-effective approach; 63 percent of messages were batched and deleted, at an average time per message of 2.43 seconds. Overall, however, the average time spent dealing with an individual spam message that reached a user’s desk was 7.92 seconds. Factors that increased the average time per message included time lost when the user’s ongoing work was interrupted by a spam message (the first step in a multi-step process for 14 percent of spam messages) and by the occasional reading of spam (an average of 11 seconds per message – generally followed by “individually deleting message”, which took an average of 5.3 seconds). Additionally, the filtering technology itself has an impact on the time that users need to manage spam, since it requires them to forward spam messages to the filter, and then delete the messages.

IDC then compared the user-reported information with data gleaned from the IS department. The IS department was able to identify total mail volumes, and the proportion of mail that is intercepted by the spam filtering technology. By combining these figures with the user-reported information, IDC was able to establish that 49 percent of all current mail volume is spam. Further, it is clear that spam volumes are increasing faster than legitimate mail volumes – an estimated 7 percent per month, as compared with 5 percent a month for legitimate messages. This means that within a year, 52 percent of mail volume is expected to be spam. Lastly, the increasing effectiveness of the heuristic system is apparent in historical intercept rates. Based on a combination of server and user-reported data, IDC concluded that the filter currently intercepts 61 percent of spam, but that it would intercept almost 85 percent within a year.

By combining the user and IS data with information on payroll, revenues, and the cost of hardware, software, and professional time used to manage the spam filtering technology, IDC was able to build a set of financial assumptions comparing the cost of spam with and without the filtering technology, and the cost benefit of the spam filter. Within this specific small business (approximately 60 employees), we found the following:

• Without the filtering technology, the productivity cost of spam over a one-year period would be $31,304; the total lost revenue opportunity would be $52,205.
• With the filtering technology, the productivity cost of spam over a one-year period is expected to be $11,770; the first-year cost of the filter itself is $9.984, bringing the total one-year cost of spam within the organization to $21,755.
• The filtering technology technology itself was shown to have a net cash benefit of $9,550 in its first year of use, with a one-year ROI of 196 percent, and a payback period of eight months.

Clearly, these are outstanding financial standards for an infrastructure technology. Coupled with other related benefits – reduction in user frustration, increased user confidence in the messaging infrastructure, reduced workplace exposure to distasteful materials – it is clear that spam filtering technology is worthy of consideration in all messaging-intensive environments.

* – If you’d like to download the case study – or if you want more information on using this method to investigate the cost of spam within your own organization, please visit www.spamcalculator.com.

For further information please contact Michael O’Neil at 416-673-2234 or moneil@idc.com.