The home improvement chain also said the malware responsible for the breach has been removed from all stores.
(Page 3 of 4)
"Bayesian filters are capable of learning to stay ahead of spammers," Vezina says. "These filters will calculate the density of the content and cross match specific words against other e-mails."
That makes it tougher for marketers that do not keep up with the latest spam-screening techniques to create campaigns that can get past such filters. "A marketer`s relationship with an ISP is just as important as creating mail with content that is not considered spam because they will base their filtering in part on the retailer`s reputation," Vezina says.
Becoming certified can reduce the scrutiny under which ISPs place a retailer`s e-mail campaigns, as well as provide tips for helping to prevent their mailings from being bounced by Bayesian filters. "It`s also best to talk to the ISP about delivery rates, do`s and don`ts, and what level of accreditation with them makes sense," adds Vezina.
Nevertheless, getting past a spam filter is in large part a guessing game as ISPs do not reveal the specifics behind their filtering process. "Unless a marketer has dedicated resources to understanding e-mail marketing campaigns there is no easy way to understand the filtering process," says CheetahMail`s Seeley. "The most effective method is to track delivery rates and trace any problems back to what made a message undeliverable."
The simplest method is for retail marketers to set up mailboxes through the ISPs that deliver their mail and send test messages to themselves. If a message is not received or goes into the spam filter, the retailer knows the message needs to be tweaked. Such tests can also show retailers how their mailing will look upon receipt by customers.
That is key since many consumers still use dial-up and other low bandwidth connections to the Internet that can hinder receipt of graphically intense messages, according to Seeley. "Monitoring inboxes for receipt and errors around delivery is important," he adds. "Conforming to spam filters requires a lot of vigilance and testing."
In the event a message proves undeliverable when a campaign actually launches, retail marketers ought to have a backup e-mail address available, especially if their message is time sensitive.
"If a message isn’t delivered, retailers must be able to default to another e-mail address or delivery channel," affirms StrongMail’s Lewis. "The more time-sensitive a message, the more limited the opportunities to get the message re-sent in time if it goes undelivered."
Lewis recommends regular testing of e-mail templates and content to keep pace with changes in the filtering process. "Your reporting must take into account deliverability as well as which messages pull response," he adds.
The risk of a message being swept up by a spam filter is why more retail marketers are integrating analytics into their e-mail campaigns. "Reporting around transactional e-mail is crucial, because it helps put rules in place that say what type of message to serve up under specific conditions," adds Lewis. "Different marketing messages will be associated with different types of transactional e-mail and even customer segments. These rules will change over time and retailers must be able to make adjustments as needed. Again, whether it’s a purchase or registration confirmation or some other form of notice, the marketing message must fit the transaction or purpose of the mailing."
The key metrics that matter are open and conversion rates as they directly impact sales and tell marketers whether their strategies are working. "There is a tendency to get overly complicated in measuring the effectiveness of e-mail," says Adams.
The objective view
In many cases the best check for the effectiveness of a campaign is for retailers to step back and view their messages on a personal level. "Marketers need to ask themselves, `Would I respond to this message` and on what level?" continues Adams. "Marketers come up with a lot of great offers, but if the copy talks down to customers or fails to get to the point, then the message gets lost and the campaign won`t fly."
When interpreting metrics, retailers ought to keep the specific metric in mind. In some cases a marketing campaign pitching a high-end item might have a low open rate, but the campaign can still hit its sales targets. "Marketers need to keep in mind the end goal of the campaign at all times," cautions Adams.
Indeed, some retail marketers send e-mails to drive traffic through other channels, based on the type of product being pitched and the recipient`s preference for making a purchase. "Some consumers use the Internet purely to shop and then buy in-store, but that does not mean they are not receptive to e-mail," says CheetahMail`s Seeley. "If an e-mail campaign is sent Monday and residual in-store sales go up throughout the week, it`s a good chance the rise is directly attributable to the mailing."
Tracking sales made through offline channels, but generated by e-mail campaigns is tricky, but it can be done by attaching tracking codes to coupons or tracking in-store sales of a specific item promoted in a campaign. "E-mail is just one component of a retailer`s overall marketing strategy and as they begin to understand this concept, retailers are closing the loop," says Hilts.
As this loop closes, retailers are becoming more intimately knowledgeable about their customers and are using that information to create links within their e-mail messages that transport their customers to dynamic landing pages customized to their shopping preferences. In many cases, these pages can be variations of those already browsed by the customer, but tweaked in their design to have better appeal, according to EmailLabs` Pollard.
The holistic view
"Every e-mail service provider is getting more integrated with analytics to better understand their customer," says Pollard. "Linking consumers to the home page through an e-mail is no longer effective."
As retailers take a more holistic view of their customer base through analytics, they gain more clues as to why certain segments of their marketing list may or may not have responded to an offer.