3. Enterprise campaign testing strategy: A third method of enhancing campaign effectiveness is the development of an enterprise testing strategy. This entails the creation of standards for key areas, such as: a) campaign dimensions and taxonomies; b) cross-channel evaluation metrics; and c) control groups.
- Campaign dimensions and taxonomies: Identifying a consistent set of dimensions and values will help evaluate results by campaign category, compare new campaigns against category averages and enable the development of category-specific benchmarks for evaluating relative campaign success. It also helps set marketing investment levels based on prior performance. For example, the campaign objective dimension may be categorized into informational, promotional, transactional and loyalty campaigns. Campaign type could be classified as holiday, upsell, cross-sell, refer-a-friend, reactivate, etc. Offers may be grouped as percent vs. dollar off, limited time or quantity, free vs. discount on general, specific department or stock keeping unit. Frequently used test dimensions for e-mail campaigns include: customer segment, lifecycle, timing, frequency, headers (from name and subject line), call-to-action, wave, personalization (name, geography, interests, browsing, transactions), creative, copy, content, graphics, layout, format, landing pages and advertisements.
- Consistent cross-channel metrics: Developing consistent and relevant metrics for each channel, as well as methods for aggregating and standardizing metrics across channels, is another important success factor. As an example, e-mail campaign evaluation should include individual metrics such as deliverability, bounces, unsubscribes, spam complaints, open rates, clicks, conversions, sales, net sales and satisfaction. In addition, it would be important to compute an aggregated campaign profit metric that balances net sales against the negative impact of spam complaints, unsubscribes and satisfaction ratings; as well as product margins, promotion and fulfillment costs. The fully loaded campaign profit metric could then be used to drive investment decisions; whereas, the individual metrics could be used in a diagnostic fashion to identify weak links, tweak micro-level campaign attributes and create differential strategies by segment. For example, a high percentage of opens but no clicks points to header issues; whereas, a prevalence of customers with clicks but no conversions may point to offer issues. Finally, a way to standardize metrics across channels is important, because it can help ascertain the relative value of each channel in driving returns, both individually and in combination. Multichannel campaigns will often demonstrate that the whole is greater than the sum of its parts.
- Control groups. Control group standards should range from minimum size requirements, statistically significant quantities based on expected response rates, differences and related factors, as well as weighting guidelines. Our assessments have shown that invalid control groups (for example, those that are not appropriately weighted by segments) appear to be a prevalent reason that marketers end up drawing inaccurate conclusions pertaining to campaign success and ROI.
4. Capturing and integrating referral network value within campaign metrics: Many organizations continue to underestimate the value driven by their marketing campaigns, because they only measure the direct sales impact on promoted customers. The recommendation is to compute a fully loaded customer profit metric that includes not just the revenue directly generated by a customer, but the total profit realized from the customers’ direct and indirect referrals, networks and advocacy/influencer status. With the rise in social networking, word-of-mouth and viral components in both traditional and emerging channels, this is becoming an increasingly important dimension.