Article Preview
TopIntroduction
Online marketplaces are gaining market share from brick-and-mortar stores everywhere around the globe (OECD, 2013). There are industries that make almost half of their sales online (OECD, 2012), and companies without any physical store such as Amazon are blooming (Bogaisky, 2014). Research shows that more often the customers visit an online store, more probably they buy (Moe & Fader, 2004). It is, therefore, essential to keep the online store in the customer’s mind. There are several means to do that, one of the most popular being email marketing. In Europe, 94% of the organizations use promotional emails as part of their marketing mix (FEDMA, 2010), with a good reason: the use of email is often profitable for the marketer. According to the Direct Marketing Association’s 2015 Response Rate Report, email campaigns provide the highest median ROI, of 21–23%, leaving behind telephone campaigns (19–20%), social media and direct mail (15–17%) mobile (12–14%), paid search (9–10%) and Internet display ads (6%) (MarketingCharts, 2015). When compared with other online channels, email is the most effective also in terms of purchase incidence and dollar sales (Danaher & Dagger, 2013) and has the best long-term effect (Breuer, Brettel, & Engelen, 2011).
While email seems to be viable in economic sense, there is room for improvement in terms of response rates. Compared with telephone campaigns (9–10%) and direct mail (3.7%), the 0.1% response rate for email is modest although at the same level as for mobile advertising (0.2%), social media (0.1%) and paid search (0.1%) (MarketingCharts, 2015). Therefore, finding ways to improve response rates would be beneficial for email marketers. One method of achieving this is to test the email newsletter to understand what kind of content the recipients find interesting and relevant enough to give the desired response. Testing also reveals content that makes email ineffective and even leads to its classification as spam, which again affects response rates.
Testing is no news to traditional direct marketers who have decades of experience in optimizing messages, targeting and timing in order to improve the response rates of their campaigns (Stone & Jacobs, 2008). Email, however, has a clear advantage in comparison to offline channels: clickstream data. It is fast, easy and unobtrusive way of collecting detailed information on individual activities (Bucklin & Sismeiro, 2009), objectively and economically. Clickstream data is collected in the user’s own environment in real decision-making situation, opening research opportunities into such topics as the modelling of decision making processes and stages (Bucklin & Sismeiro, 2009; Moe, 2006).
The article has three goals: to explore the recipient’s response process in the context of email marketing, to study how the behavior of the recipient can be measured with clickstream data, and to explain how multivariate testing can be used when aiming to generate a more positive reaction from the recipient. The purpose is to build a simple and easily understandable model for testing email newsletter design, and to demonstrate the feasibility of the model with multivariate tests studying regular, permission-based email newsletter sent by a European airline company. Internet sale is central to the business of airline companies today (Ferguson, 2014), making the case company a good example of a marketer that uses email newsletters to attract customers to the online marketplace.