Decision Management
2019 is the year enterprises take a more disciplined approach to analytics to capitalize on this new data-overwhelming reality.
2019 is a year to expect increasing use of artificial intelligence across businesses to fuel further automation and business agility. As such, enterprises will be taking a much more aggressive approach to advanced analytics and expecting more value from their investments. According to a study by IDC, 85 percent of organizations cite digital transformation (DX) initiatives as the top reason for investing in advanced analytics. Common responses driving the move to analytics include business growth, increased automation and agility, and optimizing complex decisions.Furthermore, companies are sitting on mountains of relevant and useful data to help inform their business processes and decisions. To unlock the value of that data, they’ve invested in data modeling and analytics tools and hired skilled data scientists. Yet, many are still struggling to derive the hoped-for outcomes in driving incremental value to the business. Where are they going wrong?
In my experience, the biggest hurdle to maximizing the use of data to create greater business value is being able to operationalize analytics throughout the organization. This isn’t a technology problem – it’s an execution problem. Organizations must get the people, processes and technology aligned and moving together to achieve results.
The first step to successfully operationalizing analytics is to define the process and the desired outcome by modeling the business problem and understanding where analytics will add value. Effectively captured, this then requires placing the analytics into the hands of business stakeholder in an easy to use and explainable framework that empowers the operation – the manifestation of the captured process and objectives.
One question to consider is how can analytics be applied to uncover a better way to do something already done today? Or, how can analytics be incorporated into a business operation to make a decision in automated fashion? Finding the problem to solve requires getting input from business stakeholders, and not relying solely on the data scientists. Otherwise, you run the risk having analytics that no one actually applies and therefore provides zero value to the organization.
Once analytics are effectively deployed into such a framework, the next challenge is to scale, moving from a single integration point into widespread operational use. The most effective organizations will find ways to embed advanced analytics into production control systems, supply chain management, logistics, ERP and more. They understand that the more analytics becomes embedded into their operational systems, the more they can automate decisions and improve business outcomes. Teams can then use technology tools to quickly develop, test and deploy predictive and prescriptive analytic models that impact business outcomes.
Don’t discount the people part of the equation for success. When businesses operationalize analytics, more and more decisions are automated. In theory, this seems like more effective way to ensure speed and consistency. Yet in practice, it requires significant buy-in from executives and stakeholders from across the business. Machine learning models have reached new, sophisticated levels and are being used to solve complex problems. Trusting the system to make good decisions is challenging and many projects have failed because they didn’t have the support needed to be successful. Understanding the how and the why behind the decisions and having data and accuracy checks is critical to galvanizing support.
This is where explainable AI (xAI), which provides a business user an explanation of why the analytics came to the conclusion it reached, is needed. This is particularly true in a regulated industry, but also applies to investors, partners and business leaders who may need to understand the business decisions that impacted certain outcomes. Compliance will also become increasingly important, as corporations will need to demonstrate accountability and transparency in their business decisions.
At the same time, businesses must embrace the iterative process when operationalizing analytics. Any system will need to incorporate feedback and make changes to improve decisions and outcomes. This can be challenging for those unaccustomed to an iterative learning process but is critical to the success of the initiative. Analytics should be continually improving based on feedback or changing conditions in order to improve predictions. This should assuage any concerns that AI is a mysterious black box and enable the business user to feel more in control of the outcomes since these systems can adapt and improve with new input.
Creating a collaborative, scalable, transparent process for getting advanced analytics out of the lab and into the everyday business is the only way to truly operationalize and get maximum business value from analytic investments. Given the velocity of change in today’s business environment, speed to decision and optimization of processes is no longer a nice-to-have. 2019 is the year for businesses to make real inroads to operationalizing analytics throughout the organization, or risk being left behind.
This article was originally published in Industry Today.
