Every few years we hear about a “silver bullet” that will allow organizations to become more data driven. These include local area networks, ERP systems, cloud computing, and in-memory databases. Four years ago, Blockchain was going to solve all IT and data needs. Alas, it did not. So, excuse me for being a little skeptical that generative AI will make every organization data driven.
But we have found a recipe that helps put organizations on the right path to being more data driven. Reducing data latency is part of the not-so-secret sauce. And improving data latency will better position organizations to implement and test generative AI and large learning models. Data latency is not a silver bullet; but combined with other best practices, it might lead to one. We recommend making it an area of focus in organizations not already doing so.
As more organizations strive to make data-driven decisions more frequently, data latency remains an issue. Although data keeps arriving faster, in about 50 percent of organizations a lag or delay of more than a day occurs before analytics engines can access those data. A lag of a week or month or more makes it hard for an organization to become data driven. A minority of organizations expect their data to be useable in real time or within an hour; they have a big lead in the race to be data driven and likely could be among the few (3 percent) implementing generative AI in production (see the Research Insight “Lower Risk and Create Business Value by Governing Not Limiting the Use of Generative AI“).
Data-driven organizations have significant advantages over their non-data-driven counterparts, including:
• Much higher success rates for their BI initiatives
• Higher data-literacy rates
• More in-place data leadership
• Higher BI penetration (usage) rates
• Easier to find analytic content
Many organizations should make solving data-latency issues a priority so that data will be accessible sooner—maybe sooner than associates request it—to better ensure data-driven decision making. However, equally important to faster data is the need for processes—some of which have legacy components that can be decades old—to adapt or undergo wholesale updating to address lower data-latency rates. Such revisions not only require system updates but often also require significant change-management efforts to address not only the processes but the people that use them.
Lowering data-latency rates also significantly can help foster a data culture by making more relevant data available to inform a greater number of business decisions. Achieving such a culture across all the business functions may require many shifts and transitions to ultimately transform an organization into a data-driven one. With high data-latency remaining the norm in half the organizations, this represents a hurdle to overcome on the path to a data-driven culture.
You do not have permission to access this document. Make sure you are logged in and/or please contact Danielle with further questions.