Apart from the perpetual shortage of analytics talent in the labour market, proper data management and stewardship is arguably one of the strongest inhibitors of any organization’s continued ascent up the analytical value chain.
Legacy practices have resulted in enterprise data environments that are typically oriented around an operational view of the customer. Each source system that feeds the data warehouse has its own set of rules and assumptions. Depending on the vintage of the source system, the default customer view may be any combination of a card, an account, a household, etc. Outlier events such as lost/replaced cards and the like are often overlooked and suddenly the same customer becomes many different customers within the same source dataset. Inconsistent notions of customer across source systems make the selection of integration points across source datasets a mutable exercise.
Absent of any overarching guiding principles, these issues will proliferate in the data warehouse and the end result is a fractured view of the customer. And no matter how you slice it, if you’re using your data as a source of customer intelligence, a fractured view of the customer will inevitably result in a poor understanding of the customer and ultimately yield a disjointed (read: bad) customer experience. An operational view may be fine for operational and/or financial needs, but it is not adequate for effective customer intelligence. A view of the data that is oriented around the customer POV is required.
The role of the analyst has always been best positioned to forestall irregularities in customer data lineage but has traditionally only been peripherally involved in the organization of the enterprise data. As outlined in the previous chapter of this blog series, this is changing. The more data savvy organizations have already begun to split data roles and responsibilities, handing the reins of conceptual design and business rules definition to the business (Analytics). The results are game changing.
Customer Journey View – The Basics
The modern customer journey is made up of many touch points across multiple channels. Customers don’t just browse your website. Or shop your stores. These are not isolated events. They browse other sites, then browse yours, then visit your store, then check your app while walking the aisles, then make a purchase, followed by a comment about the experience on Twitter, and so on and in any combination. These singular journeys are deconstructed across multiple source systems and data feeds that should be pulled back together in your data warehouse. A common customer identifier that allows you to track the same customer across all touch points is just the beginning.
Typically, most organizations take a myopic view of their customers’ journeys from a data and analytics perspective. Digital focuses on the digital properties, usually each in isolation. Loyalty focuses on the loyalty data. You get the picture. The end result is that each group sees only a partial view of the complete customer journey. Rather than converging to a consistent, unified view, each data domain begins to naturally diverge in form and function, taking on the underlying character of its source data. Clicks become visits, customers become cards, etc. A single voice is needed to drive the overall conceptual design (in consultation with the individual business owners of the data) to ensure consistency across all dimensions (e.g. product, location, etc.), not just customer. This is the key difference between a single customer view and a customer journey view. The latter is a prerequisite for higher order (advanced) analytical capabilities.
All customer touch points will be unified in a customer journey data environment:
- Transactions – currency exchanged (e.g. sales, returns, loyalty currency earned/redeemed, donations, etc.)
- Interactions – inbound (call center switch & speech, web, mobile) and outbound (contact history across all channels)
- Intent – web search
- Sentiment – survey/research, social
- Contextual – weather, geo-location, sensor, etc.
- Demographics – age, gender, income, life stage, etc.
Each of the above provides a window into customer behaviour and (where available and/or applicable) needs to be considered when evaluating the true causal drivers of customer outcomes. Along these lines, the complete customer journey is indispensible to understanding customer needs and anticipating important events.
Customer Journey View – The Benefits
Improved customer intelligence: Analytical output across the enterprise is built off a complete (not partial) view of the customer. In general, as more of the aforementioned customer touch points are integrated into your analytics capabilities, the better your customer intelligence.
Improved productivity: In addition to improvements in the quality and utility of your customer intelligence, a fit-for-purpose data environment minimizes the burden of data preparation (collection, cleansing, linking, etc.) and frees up analytical resources for higher order capabilities.
Automation: A unified data framework facilitates automation and scaling of capabilities across the business.
Speed and volume of knowledge creation: Analytical output developed on a common data framework facilitates knowledge sharing, which in turn accelerates knowledge building across the enterprise.
Michael Poyser, VP Analytics at Aimia
Mark Babij, Associate VP, Customer Insight & Analytics, Canadian Tire
Joon Park, Senior Marketing Manager, BMO Insurance