This is first in a two-part blog series exploring the differences in capabilities and requirements between real-time, embedded analytics-based business systems and historical enterprise systems, as well as the impact this decision has for organizations trying to compete in today’s intensely competitive business environment.
Historical enterprise systems wrestle with the conundrum of the 60%. CFO Research conducted a study of senior managers in 2014 and discovered that 60% of CFOs felt their enterprise systems are truly self-service. However, when other senior managers were interviewed, 60% of respondents confessed that they could not use the data from their enterprise system to make effective decisions. The same 60% said they relied on subject matter experts to provide insight necessary to make a decision.
Organizations spend millions to implement enterprise systems that take too much time and too many resources to create a solution that fails to provide the insights necessary to direction daily operations or set the course of a new strategy – all of which demands an answer to the question – “why?”
The answer is that “context matters.” Data without context to a specific business problem or exception condition provides little value. Historical enterprise systems fail to provide context because they fail to incorporate embedded real-time analytics to provide the necessary context to support the “moment.” This could be a sourcing decision, a sales discount, a commitment to larger order within lead time, a change in a transportation option, an investment in an expensive machine setup, making a sensitive credit decision, and/or validating an employee’s certification to perform a job task for work that is running behind. The point is that data needs a context / framework with boundaries of metrics and tolerances to provide the awareness and guidance to make a decision.
Real-time business is about the ability to anticipate, while historical systems can only provide the ability to see what and why something happened. Historical enterprise systems can answer basic analytical questions of “Who did what?” or “Where did that come from?” But this comes at the cost of redundant data and expensive infrastructure. It also holds the business hostage to the backlog demands and skills of IT professionals.
The work of real-time embedded analytics requires a completely different hammer. The prerequisites are demanding; a) the ability to support large data volumes; b) the data must reside at the lowest level of granularity (in business terms this would be Purchase and Sales and General Ledger line item level) and c) there must be tools that allow business people to discover insights without the support of IT and the anxiety that they will bring the system to a crawl.
The second blog in this series will focus on the importance of real-time analytics to business platforms that can enable real-time decisions based on critical context.