Over the past 20 years, analysts have taken advantage of the power of transactional data. The values can be analyzed and reported in many new ways. Once consumers open their wallets and make purchases, certain details of those transactions turn out to be very trustworthy indicators of future orders. Human behavior is quickly becoming predictable in new ways as neuroscience and database technology integrations provide behavioral indicators.
The game-changing element is that the metrics are now driven by human behavioral data, rather than only transactional data. Scoring these new indicators has been less tangible until now. But now we can test beyond just how orders are placed, and figure out why orders are placed.
Let’s look at three elements concerning human behavioral data for our 2017 applications.
Grouping records categorically has been successful within our transactional methodologies for identifying the preceding lead generation. We have used segmentation for decades to report on how orders are placed: Now we can dig deeper.
For example, most brands usually have two or three significant groups of records that tend to have similar order process variables. Transactional elements allow us to identify different records in order to segment these macro groups. We use transactional data tied to a single customer or prospect identification number to segment by fundamental order metrics.
Let’s consider business segments vs. consumer segments: What drives this macro grouping is that the fundamental data elements necessary for inclusion in each group are very different from the other. Business purchasing, with all its complexities regarding who influences the buying decision, involves many overlapping data elements including sites, site penetration and key person(s). With consumer data, there is a single person at a single billing address. The very apparent differences in the two records leads to frequent separation for both strategic and reporting purposes.
Combinations of transactional fields like recency, monetary value, frequency, channel of purchase and product category exist on a more granular level. For some brands, all of these elements are scored for marketing decisions. But for others, only appropriate key performance indicators (KPIs) are used. This further segmentation and classification allows even greater flexibility and targeting than ever before.
Web browsing behavior is the most accessible of the new online opportunities to score and leverage human behavior for smarter marketing decisions. By tagging websites and storing browsing data, we are now able to interpret resulting data in ways that allow experienced marketing analysts to develop KPIs based on the behavior that leads to orders.
Here is an example (see chart above): A significant amount of browsing data from websites reveals that customers who shopped in the middle of the day were more likely to visit additional pages, create a shopping cart and place an order, vs. those who visited the site in the middle of the night. We can also learn that when this browsing activity happens on a desktop computer vs. other devices, along with a Windows operating system, more carts and orders are likely. A behavioral KPI can then be developed using the data’s attributes, using records as predictors of how likely a prospect is to respond to an email campaign.
What is new for 2017? Testing data models that may be able to predict why customers choose one product or service over another.
Elements like time of day and the content of pages viewed indicate what is really driving the sale. Even a particular computer brand can help us target: Apple loyalists vs. PC fans — we know they behave differently!