The truly data-agile organization requires an architecture that provides access to needed data as quickly as possible, applications that access and analyze large amounts of data looking for trends, and user interfaces that allow everyone to query information without having to rely on others for analyses.
Organizations today strive to be data-driven – looking to be able to use data to make informed business decisions. A quick look at the internet reveals several definitions, but in summary, a data-driven organization is one with a culture of embedding data into every major decision. As every big business decision is a bet on human behavior, the understanding of what our clients, and in some cases more importantly, those who are not our clients, want, need, and how they respond and act is critical. Data and information are king.
So why are so many data-driven organizations still lagging behind their competition? There is ample data, it is readily available, and it is of the level of quality required by the users. And, in following the definition of being data-driven, organizations have the culture necessary to base decisions on empirical evidence rather than ‘gut’ feelings.
Businesses must now look beyond being Data Driven and look to become Data Agile.
Understanding data agility and what it means for organizations
The Data Agile organization must, as a prerequisite, be data-driven. It embraces the definition and has all of the necessary components in place but goes a few steps beyond.
The term ‘data agility’ is one that already exists. It goes beyond being data-driven by stating that data needs to be made available to the end user at the speed of decision making. Basically, the speed in which you can extract data from the vast amount available to you as well as the speed in which you can turn that information into meaningful action.
There are three basic areas that we can look to in order to make organizations more agile when it comes to the utilization of data: Architecture, Applications, and User Interfaces.
Architecture in data agility
The first of these falls into the architecture area. Many new architectures have come into play in order to reduce the time necessary to make that information available to the consumer. These include edge computing and the concept of the data mesh. Both look to decrease the time required to access necessary information. These architectures, as well as others, are focused on making data access that much more efficient.
See also: Everyone Wants to be Data-Driven, but Few Want to do the Driving
Applications in data agility
A number of applications and languages now focus almost specifically on data analytics. Their capabilities and ease of use go far beyond the statistical tools used in the 80s (when I started), and their ability to look for trends in the data and report on them is advancing on a day-to-day basis. These increases in both capability and functionality make getting the information to the decision maker quicker and more efficient. However, while decisions are being made by individuals empowered to do so, many must still request information from those that have knowledge of how to gather and analyze that data to find trends and correlations. The roles of Data Scientists continue to increase, as do their roles becoming more embedded in corporate culture.
While there are a number of Analytics and Reporting applications available, knowing how to use those applications also requires training. If you do not have that training, then you must request those reports from resources that can create and maintain them for you. Even small modifications associated with replacing some data points with others can take time: days or even weeks. In today’s fast-paced business environment, this can seem like years.
Unlocking more potential through advanced user interfaces
If we are looking to make data available on an up-to-the-moment basis, then we must be able to make the analysis of this information up-to-the-moment as well. Our empowered decision maker may look at the results of a report or delve into the information on their dashboard, but what if that analysis leads to other questions. How can they get to a deeper meaning without having to call in a programmer or data scientist, explain the problem, see initial results, and then wait longer as the final report is generated? The ability of certain decision makers to act in the moment can be critical when dealing with evacuation orders for a hurricane or the routing of first responders during a forest fire. Time lag in these, as well as other situations, could be critical.
Can we build applications using today’s technologies that are more intuitive in nature? My 2.5-year-old grandson recently picked up an iPad and, after helping him get to the right program, was able to figure out how to play one of his games relatively quickly. He was able to take this knowledge and play other games even faster. I shudder to give him my iPad now for fear that he will erase all of my contacts and emails. But the fact is that the iPad was built to be as intuitive as possible.
Analytical applications can take a cue from technologies used in the home and bring that capability to the office. Absolutely, applications are becoming more intuitive, but can we build in not only appropriate icons but AI and ML as well? And what about NLP – how much easier would it be if I could just ‘ask’ a Siri/Alexa type system interface to determine if there were any trends in a repository of data and graph the results for me or modify the attributes considered in a trending report without having to consult with a programmer. A truly data-agile organization understands that productivity is higher when there is no learning curve associated with the technology. Intuitive programming that allows empowered decision makers the ability to become more knowledgeable to make better decisions should be built into data science applications.
Being Truly Data-Agile
The truly data-agile organization takes all three of these attributes into consideration. An architecture focused on getting needed data as quickly as possible, applications that access and analyze large amounts of data looking for trends that we may miss, and user interfaces that allow even those with little understanding to query the information on their own without having to rely on others for – at least at the moment – more simple analyses. The three of these facets, in conjunction, will truly empower the business decision maker and allow them to move ahead of their competition.
