Email Marketing Optimisation and Analytics
Email marketing remains one of the most cost-effective digital marketing tools, with research indicating a return on investment (ROI) ranging from £38 to 44 for every £1 spent. Despite its potential, the medium faces significant challenges, including low open rates and high unsubscribe volumes due to intense competition for consumer attention.
Analysis across multiple studies reveals three critical pillars for campaign success:
1. Subject Line Optimisation: The subject line and sender identity are the primary factors determining whether an email is opened. Predictive machine learning models, particularly Random Forest, can effectively categorise subject line quality using structural and content features.
2. Customer Behaviour Dynamics: "Email-active" customers are not always "purchase-active." Managing long-term profitability requires a nuanced understanding of latent relationship states and the avoidance of "over-emailing," which can diminish lifetime customer value by up to 32%.
3. Strategic Persuasion: Email content type (Promotional, CRM, or Alerts) triggers varying levels of "persuasion knowledge" in consumers. While promotional emails are effective at reducing shopping cart abandonment, CRM-focused emails are often superior for driving overall spending and building long-term brand affinity.
Dynamic Management and Customer Relationship States
A critical finding in email analytics is the disconnect between engagement (opens) and conversion (purchases). Using a unified Hidden Markov Model (HMM) and copula framework, researchers identified that customer preferences and relationship levels shift over time.
The Three Latent Relationship States
Customer-firm relationships can be categorized into three distinct states that govern behavior:
1. Low Open / Medium Purchase: Customers who may know the firm well enough to buy without needing to open informative emails.
2. High Open / Low Purchase: "Information seekers" who engage with content but have low conversion rates. Targeting these customers aggressively may be suboptimal for profit.
3. High Open / High Purchase: The most valuable segment, demonstrating high engagement and high transaction frequency.
Email Frequency and Profitability
The number of emails sent has a non-linear effect on both short- and long-term profitability. Finding the "magic number" of contacts is vital to avoid driving customers away.
• Optimal Frequency: In a study of a U.S. home improvement retailer, the optimal contact frequency was identified as seven emails per month.
• Profit Loss: Sending four emails (under-emailing) resulted in a 32% loss in lifetime profit per customer. Sending ten emails (over-emailing) resulted in a 16% loss.
Correlation: There is a positive correlation between email open and purchase behaviors, but they must be modelled jointly to capture the long-term impact of emails on the customer-firm relationship.
Comparison of Email Types
Research classifies emails into three functional categories based on their persuasive intent:
• Wear-out: The effectiveness of email diminishes over time as consumers receive repeated exposures. CRM emails show a clear wear-out pattern where recent emails are more effective than older ones.
• Non-monotonic Effects: Promotional emails often show a non-linear pattern of effectiveness. High levels of overt persuasion can trigger "coping strategies" such as ignoring the sender or disengaging with the message.
• Shopping Cart Abandonment: While CRM emails drive general spending, they can actually increase cart abandonment if used at the wrong stage. Promotional emails are the superior tool for converting a "saved" cart into a completed sale.
Consumers who "opt-in" to receive specific alerts are generally less resistant to persuasion. They respond more favourably to CRM emails than non-opt-in consumers, leading to higher spending. However, opt-in status does not significantly change the reaction to overt promotional discounts.
Strategic Checklist for Decision Support
1. Audit the Journey: Are you evaluating success by "Open Rates" alone, or are you tracking the full progression to "Abandonment Reduction"?
2. Calibrate Frequency: Are you at the "Magic Number" of 7, or are you currently accepting the 16%–32% profit loss of misaligned frequency?
3. Identify HMM States: Are you targeting "High Openers" who have "Low Purchase" intent with the wrong content?
4. Keyword Logic: Have you lemmatized your subject lines and selected keywords based on Arithmetic Average performance?
5. Sequence Execution: Is your hard-sell Promotion preceded by a 14-day Relational (CRM) lead-in?
References
Mahmoud, A.B., Grigoriou, N., Fuxman, L., Hack-Polay, D., Mahmoud, F.B., Yafi, E. and Tehseen, S., 2019. Email is evil! Behavioural responses towards permission-based direct email marketing and gender differences. Journal of Research in Interactive Marketing, 13(2), pp.227-248.
Paulo, M., Miguéis, V.L. and Pereira, I., 2022. Leveraging email marketing: Using the subject line to anticipate the open rate. Expert systems with applications, 207, p.117974.
Thomas, J.S., Chen, C. and Iacobucci, D., 2022. Email marketing as a tool for strategic persuasion. Journal of Interactive Marketing, 57(3), pp.377-392.
Zhang, X., Kumar, V. and Cosguner, K., 2017. Dynamically managing a profitable email marketing program. Journal of marketing research, 54(6), pp.851-866
