A Customer-Centric Approach to Digital Marketing Analytics2HANNAH HILLERA NEW VIEW ON MEDIA: A CUSTOMER-CENTRIC APPROACH TO DIGITAL MARKETING ANALYTICSinteraction with multiple channels and media across various customer engagement points characterizes a customers actual experience. As the complexity of the marketing environment evolves, it is imperative for marketers to shift from a campaign-focused to a customer-focused mentality. Transforming from a campaign-based approach to a more customer-centric one expands the considerations needed to efficiently manage marketing. Rather than only asking which campaigns had the most converted last-clicks or the best ROI, the conversation evolves to include questions about customers and their behavior: How can media be used to increase the value of existing customers? What media attracts new customers? How does customer interaction with media differ with geography? To provide answers to these questions, digital marketing analytics must also evolve. CREATING A CUSTOMER INSIGHTS PLATFORMTo strengthen our clients digital strategy through deeper customer insights, we created a Connected CRM Customer Journey marketing dashboard (Figure 1) by combining digital conversion stream event data and customer segmentation scores. The dashboard provides a breakdown of conversion path position by event type for the selected customer subset (#1.1). Conversions paths are broken down to those with a single event (one touch) and multi-touch event paths. For multi-touch event paths, the first event in a conversion path is considered an “introducer,” the final event is considered a “closer,” and all events between the introducer and closer events are considered “promoter” events. Figure 1. Connected CRM Customer Journey Marketing DashboardMultiple segmentation scores for the customer associated with each conversion are joined to event-level data to denote the customer segment at both the time of the purchase and at 90 days after the purchase (#1.2). The incorporation of segmentation scores from two time points with the event-level detail enables analysis of customer migration between segmentation groups based on media interaction. For each conversion, customers are designated as new or returning (#1.3). Counts of unique new and existing customers, as well as overall unique customers, are included in the dashboard view. This breakout eases the filtering of the customer subset to include only first-time or returning purchasers, and also shows the #1.1#1.2#1.3#1.43HANNAH HILLER1.877.9MERKLE | MERKLEINC | © 2015 MERKLE INCcustomer split between the two groups in a given subset. The map displayed on the dashboard provides the capability to evaluate differences in media event interactions between countries and geographic regions, while the calendar permits filtering by date. Details for the filtered data subset are presented on the bottom of the dashboard where appropriate (#1.4). For example, the unfiltered version of the dashboard shows that the most commonly interacted with paid search campaigns were primarily branded campaigns, whereas the email campaigns that were interacted with most frequently were monthly deal alerts.CUSTOMER MARKETING INSIGHT: HOW CAN MEDIA BE USED TO INCREASE THE VALUE OF EXISTING CUSTOMERS?Rather than just providing a static snapshot of all conversions, the real power behind the Connected CRM Customer Journey marketing dashboard lies in the interconnectivity of its filters to investigate questions and uncover insights. This functionality empowers the marketer to effectively target a customer segment and influence consumer behavior.One goal of our client was to increase the value of existing customers. In order to influence the advancement of customers into more valuable segments, we first had to recognize how media affected customer-value migrations. That is, what media interactions are characteristic of customer migrations from low-value segments to high-value segments and vice versa? To determine this, it was critical to have customer value data from multiple dates linked in order to review trends of how customers transform over time. We began by restricting the dashboard to customers that were in the lowest value tier (Bronze) at the time of purchase and that had also transitioned to the highest value tier (Gold) within the next 90 days (Figure 2). In order to do this, we first selected the wedge for Bronze in the Original Customer Value Seg chart. Selecting data on one chart automatically cascades the filter through all the other charts, updating the entire the dashboard. Compared to all customers in the unfiltered version of the dashboard, we saw the 90-day segment of customers with the original score of Bronze were much more likely to be Bronze or Silver. To complete the creation of the desired customer subset, a second filter was placed on the Customer 90-Day Value Seg chart, which also cascaded through the entire dashboard, including any charts that had already been filtered (Figure 3). The count of unique customers, which had been 206,146 in the unfiltered dashboard, was now reduced to the 506 that fit the filtered criteria. These 506 customers placed 1,201 orders and had a total of 42,163 events in their conversion paths. Figure 2. Dashboard: Restricted Customer Value SegmentsFigure 3. Dashboard: Restricted Customer Value Segments, 2ndFilter4HANNAH HILLERA NEW VIEW ON MEDIA: A CUSTOMER-CENTRIC APPROACH TO DIGITAL MARKETING ANALYTICSWe found that this subset of customers had a relatively large number of interactions with a non-branded product campaign that appeared in its paid search events with greater frequency than in the paths of customers in the unfiltered version (Figure 4). With the assistance of the dashboard, this campaign was identified as interacting frequently with Bronze customers that went on to become Gold customers. With this knowledge, further analysis on the campaign confirmed it as a good customer value transformer.Depending on attribution method, contribution of an upper-funnel interaction may be overlooked, despite it being critical to valuable customer conversions. The campaign our analysis identified appeared with much greater frequency as an introducing event than a closing event for Bronze customers that were later seen to migrate to Gold. However, the performance evaluation method currently being used was most-recent-paid-click, which often erased the contribution of the upper-funnel keywords by transferring conversion credit to clicks later in the event path. These insights led to testing opportunities that incorporated upper-funnel contributions into performance measurements in order to raise the SERP position of identified customer-value transforming keywords to incrementally influence the migration of customers to higher value segments.CUSTOMER MARKETING INSIGHT: WHAT MEDIA ATTRACTS NEW CUSTOMERS?Attracting new customers was another goal of our clients digital marketing strategy, so we investigated which media and campaigns new customers interacted with at a greater frequency than returning customers. By filtering in turn for new customers in the Customer Type chart, then existing, we were able to compare how the two groups differed in their interaction with media (Figure 5). We found that the conversion paths of new customers had organic and paid search terms act as an introducer at a much higher rate than for existing customers. Further analysis led to the identification of several specific paid campaigns that introduced new customers at a significantly higher rate than other paid search campaigns. Increased funding of these campaigns allowed the client to attract a greater number of new customers. Using conventional campaign evaluation methods, these campaigns would not have had their funding increased, as their most recent paid click ROI judged them less efficient than others at driving revenue. By evaluating the performance of the campaigns based on their effectiveness at attracting new customers, the identified campaigns came out as the clear winners. By taking a customer-centric view to the marketing analysis, we were able to support the clients strategic initiative using campaigns that would not have been identified by using conventional campaign-based ROI metrics.Figure 4. Dashboard: Restricted Customer Count & Campaign ViewFigure 5. Dashboard: View of Customer-Media Interactions5HANNAH HILLER1.877.9MERKLE | MERKLEINC | © 2015 MERKLE INCCUSTOMER MARKETING INSIGHT: HOW DOES CUSTOMER INTERACTION WITH MEDIA DIFFER WITH GEOGRAPHY?Customer behavior differs by culture, geography, and language. Understanding how behavior varies permits spend and time to be focused on the most effective interactions in each region. Although there may be a similar channel spend percentage breakdown across countries, different channels are more important in conversion paths than others depending on location. Interpreting these geographic differences can inform more appropriate media spend allocations. In our analysis we saw that restricting data to customers in the United Kingdom revealed slightly over a quarter of single-event conversion paths were comprised solely of paid search events (Figure 6). Conversely, restricting to customers in Germany showed that hardly any single-event conversion paths had paid search events as the sole event (Figure 7). Insights like these led the team to a better understanding of how customers interact with media dependent on geography, which in turn enabled implementation of more geo-specific marketing strategies. More detailed analyses at state and postal code levels generated more geographically refined insights to be incorporated into campaign management. LEVERAGING A CUSTOMER INSIGHTS PLATFORM IN MARKETING ANALYTICSRather than starting with media and then looking at how it affected different customers, the Connected CRM Customer Journey marketing dashboard allows analysis to begin with the customer at the center, by first identifying the desired customer behavior (conversion to higher value segments, first-time purchasers, etc.), then examining the relevant media. A customer-centric marketing platform enables integrated customer data to be cross-sectioned numerous ways to achieve agile comprehension of customer journeys, migrations, and behavior. Customer-centric insights and strategies will continue to evolve along with data collection and management capabilities. Further integration between offline and online touchpoints will improve our visibility into the overall customer experience, as will better customer identification and recognition across devices. This marketing analytics dashboard has facilitated understanding of interactions between numerous customer interactions across media leading to more integrated and effective marketing decisions. By putting the customer at the center of our data analytics, deeper insights for clients can be driven and a point of competitive differentiation created.Figure 6. Dashboard: UK, View of Customer-Media InteractionFigure 7. Dashboard: Germany, View of Customer-Media Interaction6HANNAH HILLERA NEW VIEW ON MEDIA: A CUSTOMER-CENTRIC APPROACH TO DIGITAL MARKETING ANALYTICSHANNAH HILLERSenior Digital Marketing AnalystHannah works in Merkles Digital Agency Group where she provides analytic support to clients digital marketing activities. Her areas of focus include trend and variance analysis, cross channel attribution, report development and scenario planning.She has a MBA from the University of Pittsburgh and undergraduate degrees in Chemistry and German from Allegheny College.Merkle is a global data-driven, technology-enabled performance marketing agency and the largest independent agency in the US for CRM, digital, and search. For more than 25 years, Fortune 1000 companies and leading nonprofit organizations have partnered with Merkle to maximize the value of their customer portfolios. The agencys heritage in data, technology, and analytics forms the foundation for its unmatched skills in understanding consumer insights. When combined with its strength in performance media, Merkle creates customer experiences that drive improved marketing performance and shareholder value. With more than 2,700 employees, the privately held corporation is headquartered in Columbia, Maryland with 14 additional offices in the US and offices in London, Shanghai and Nanjing.For more information, contact Merkle at 1-877-9-Merkle or visit merkleinc.