Companies already have most of the information they need to improve the customer experience. Getting the necessary data out of the systems where it resides and making it useful is the big challenge with “big data”. Application-specific data-storage solutions evolve over time with little thought to enterprise-level access or usability. Scattered entities, be they different departments, affiliated companies under a corporate umbrella, or product groups, store their customer data in various formats. It is extremely difficult to assemble this hodgepodge of information into a single customer view.
But here is a secret – much of that work may already be done. Companies have been collating customer-specific information from repositories across the enterprise for decades. A good portion of the data necessary to improve customer experiences exists in the documents corporations have been producing all along. Documents almost always include data from CRM systems, billing, marketing, and more. The information residing in documents can be just what an organization needs to jumpstart their initiatives to improve the customer experience.
There are two problems though:
1. Most documents are only a snapshot in time. They may contain details about customer activity over the last month or quarter, but compiling a historical record of customer interactions within an organization requires analyzing thousands or sometimes millions of individual documents.
2. Documents are frequently archived in print formats such as AFP, Postscript, PCL, or metacode. These file formats, besides being dissimilar from each other, lack the structure required for effective use of data analytic software.
Going from Documents to Data
Companies have been using specialized software, known collectively as “document re-engineering” or “post-composition reformatting” tools for years to enhance documents before distribution. Adding barcodes, re-positioning address blocks, or colorizing data elements are some of the tasks organizations commonly achieve with document re-engineering software. Organizations can use the same tools to extract valuable data from multiple types of customer communications, allowing corporations to compose a 360-degree view of customer relationships.
As an example, an insurance company seeking to leverage customer-specific information about policy types and limits, claims activity, correspondence, and billing history could face time-consuming and expensive development to pull data from four different sets of databases. Securing approval and I.T. resources to work on such a project could be difficult. Conversely, software designed to extract the relevant information from archived documents can trim development time drastically. The insurance company can begin using individual customer history information sooner to achieve corporate goals for customer relationship improvements and distinguish themselves from competitors.
While extracting big data from archived documents can be immensely valuable for special undertakings (see our project with IBM), companies can employ the same methods to influence daily customer interactions. Organizations can use knowledge about customer activity derived from past communications to compose more relevant future messaging.
The information necessary to cause a change in customer relationships already exists. Unlocking the data essential to improve future customer interactions just takes the right tools.