The rapid and increasing pace of change in the business landscape has led many data leaders to assume they need to drive toward real-time data delivery in support of all business intelligence (BI) deployments. But not all business processes and decisions require real-time (or even daily) data, nor could they readily take advantage of it, if it was available. Instead, data leaders should consider applying the concept of operational tempo, which aims to manage the timing of processes and data flows so they properly and optimally support and enable business objectives.
Limitations or complexities in business processes, management approaches, and levels of data literacy are just some of the many factors that can limit the speed at which an organization can operate. Therefore, engineering all BI solutions for high-recency data does not represent a valid and cost-effective strategy.
Data recency measures how timely data must be to ensure that users make effective data-driven business decisions. Recency can range from real-time (truly current data) to quarterly (and possibly at even longer intervals, although such cases are rare). Most organizations categorize data recency for specific BI implementations as real-time, hourly, daily, weekly, monthly, and quarterly. Data delivery processes are engineered to ideally match data recency to the specific decision-making requirements of the business processes and people that the BI deployment supports.
Where it makes sense, organizations are investing in increasing data recency, driving toward real-time availability of data for BI and analytics. But monthly and daily data delivery still dominates within BI implementations because that frequency suffices for many types of decision making that support business processes not operating at real-time speed. Examples include financial analysis and financial-close processes, product development processes, and longer-term business development activities. These lower-recency requirements are not going away quickly—organizations likely will have substantial need for them in the long term.
Our data shows no clear correlation between higher recency of data and a greater degree of success with BI. In fact, the organizations most successful with BI acknowledge their deployments span a mix of data recency requirements, including everything from real-time to quarterly. As part of any strategic investment in data and analytics, data leaders need to analyze their data recency requirements, rather than assume real-time (or even daily) data is always necessary. Likewise, reviewing existing data-delivery architectures to identify cases where data recency does not align well with business requirements will help increase the value of in-place BI initiatives.
Data leaders will achieve success and create high levels of BI value when they architect and design data recency to match the timeliness requirements of specific business processes and strategic decisions, and take into consideration how those processes and decision-makers actually make use of higher-recency data. Understanding the facts about current data recency relative to business-process, user, and decision-making needs and abilities will help organizations continually optimize data pipelines to get the most value from their data and analytics programs.
You do not have permission to access this document. Make sure you are logged in and/or please contact Danielle with further questions.