The TV and web retailer will bring its e-commerce and video programming to France in 2015. QVC already sells in Germany, Italy and the ...
SPONSORED SUPPLEMENT: E-mail marketing
(Page 2 of 4)
Once retailers know the types of marketing communications in which their customers are interested, the next step is to craft short, concise messages. By getting to the point, especially in the subject line, marketers can cut through the clutter and create a purposeful call to action that leads to customer interaction.
"There is a link between the purpose of the message, its length, and the content," says Mike Hilts, president and general manager of YesMail, an infoUSA company. "The purpose of marketing e-mails that have a lot of fluff can be confusing. The objective of the campaign has to be clearly set and executed in the mailing. If a mailing is intended to support the publication of a new print catalog, that`s what it should do."
Once retail marketers understand how to create pointed relevant messages, they can then begin to focus on creating personalized messages. This means more than using such benign techniques as incorporating the recipient`s name into the greeting. Personalized messages are crafted on a thorough understanding of customer behavior patterns, purchases and preferences.
Sources for these data include Internet cookies that attach to a shopper`s personal computer when visiting a specific site and track movement across the site, entry and exit points to the site, and when during the session the customer abandoned the shopping cart.
"Sending a coupon to a customer who abandoned the shopping cart won`t necessarily get them to complete the sale if the item they were looking at needs to be touched and viewed in person," says Pollard. "Marketers need to look at what might trigger customer behavior before sending an e-mail message."
Marketers should also consider the types of products purchased when sending e-mail intended to create a cross-sell or upsell opportunity. In this case, timing is important because merely pushing a marketing message after the sale can fail if it comes too soon or too late. Missing the mark reduces the relevancy of the message and feeds the perception that the retailer knows little about its customers.
"Shoppers who buy a printer and toner cartridges are sure to need replacement cartridges down the road, but it may not be for a few months," says Pollard. "The retailer knows how long a cartridge will last and how many were initially purchased and can time the follow-up message accordingly so it is relevant and thoughtfully proactive."
The segmentation challenge
Segmentation of mailing lists along customer interests is another way retailers can personalize e-mail campaigns. This can be tricky for highly specialized retailers, because their customers all have the same interest, which is what the retailer caters to, as opposed to larger retailers carrying a broader array of goods.
"If a retailer`s catalog is highly targeted in the first place, it becomes more difficult to personalize messages and offers because their catalog is so specific to a particular interest," explains Arial Software`s Adams. "Besides, most specialty retailers don`t have the resources to gather data that granular. In this case, the best way to get personal is to ask the customer during a consensual contact what they want in their e-mail messages."
This can be done via a quick survey during a shopping session or that is delivered through an e-mail. "Customers will take the time to make their opinions heard, even if there is no reward," adds Adams. "If a small sample replies, it still provides insights marketers might not otherwise get."
Using list segmentation to create personalized messages can also be done on a broader basis, such as by season or type of event. While such campaigns are less complex to construct, they are still highly effectively as long as the marketer avoids doing an e-mail blast to the majority of the list.
"There are lot of complexities to creating personalized marketing messages and sometimes it`s better to move away from those complexities and personalize messages based on broader lifecycle behavior," says YesMail`s Hilts.
Another personalization option is adjusting e-mail campaigns in real time. This involves sending a small test sample, tracking what customers respond to in the message and moving those elements up or down accordingly or replacing them altogether. Once the right formula is hit on for specific customer segments, the campaigns can be rolled out.
"The challenge is that retailers are conducting battlefield marketing, which means they need the tools to analyze the data in real time so they can make the right recommendations and create purposeful mailings," says Hilts. "If a message lacks purpose, it lacks relevance."
While YesMail is planning to add an artificial intelligence component to its predictive models, it is at least a year way from doing so, as artificial intelligence is only now seeping into e-mail marketing.
Hilts sees artificial intelligence as a way to help marketers better manage the rules placed around e-mail marketing campaigns of which there can be hundreds. "The rules around personalization can make execution pretty complex and that is leading some marketers to personalize on a simpler level around events or seasons," he adds.
Creating the right message to be read by the customer at the right time is only part of the battle. Retailers must still get them past increasingly stringent spam filters used by Internet service providers.
Becoming certified with an ISP as an e-mail marketer, a process also known as white listing, is the starting point to ensure delivery. Still, mail can get bounced by filters that ferret out suspect words in the subject line or the body of the message. Some filters, such as Bayesian filters, are even capable of detecting whether the content matches up with the subject line.
What separates Bayesian filters from other spam filters is that they learn over time to identify new types of spam the more they analyze incoming e-mail, according to Got Corp.`s Vezina.
Filters that learn