Skip to main content Skip to Footer

BLOG


July 25, 2018
Thinking bigger with artificial intelligence: Part 2
By: Mitch Gross

In a previous blog post we talked about the growing importance of artificial intelligence to companies’ operations, and the need for organizations to think bigger about how they can use AI to drive the fundamental transformation of key business processes to unlock exponentially greater value. In this post, we address another big question executives ask us: How do we get started using AI?

First, it’s critical to recognize that implementing AI is a transformation initiative that requires some of the same basic elements as any large-scale change project. This includes proper change management, as well as tight collaboration between the IT organization, the line of business involved, and a third party that may be providing expertise and knowledge to help guide the initiative.

Beyond change management, there are several other keys to a successful AI implementation.

Define the North Star

You don’t necessarily need to have a fully fleshed-out AI strategy to get going, but you do need to define your North Star. This doesn’t mean identifying the specific technology you want to adopt or the exact processes you want to automate. Rather, it means determining the explicit outcomes you want to achieve, the challenges you need to address, the pivot you want your business to make. And a critical element of defining the North Star is to specify the new measures that will gauge your success and keep you focused on those outcomes.

Establish the right senior-level sponsorship and alignment

Many companies today still operate within silos—technology versus business functions, or within different countries, regions, or units. To get beyond these silos, senior leadership needs to establish AI as a priority and put in place the right governance model to get people working together toward a successful implementation.

Make sure you have the right data

Achieving the promise of AI is as much dependent on data as anything else. Just like humans, AI needs high-quality, relevant, trustable, and accessible data to make good, informed decisions. To ensure your data meets these standards, you need to manage and govern it as you would any other strategic, inherently valuable asset.


#AI making strides, but business leaders are still getting their hands around potential benefits. Six need-to-knows:

 
 

Address the talent dimension

You have to define the kind of people you’ll need for AI to flourish. According to Accenture’s Intelligent Operations research, the most important workforce requirement today is having creative and innovative talent, with a healthy entrepreneurial curiosity. Such individuals are well equipped to understand, from both a process and technology perspective, how to apply AI to solve specific business problems; and also to manage work that’s done by a combination of humans and machines. This talent can come from inside the organization, but will likely require training to become proficient in their new roles. That’s why it’s vital to have a plan that maps out your workforce of the future and identifies the new skills you’ll need to train for, hire or source from an ecosystem partner.

The key here is to understand that AI is meant to augment, not replace, today’s workforce. If you're not thoughtful about the way people work (and work with AI), how you communicate what’s happening, how you train people, and get their buy-in, AI won’t gain momentum or scale.

Identify and prioritize areas of opportunity and use cases

You should first look for processes where there’s scale, because you need scale to make a big impact. You also should look for centralized processes that are stable—ones that don’t change often and don’t have a lot of exceptions. The more localized and exception based a process is, the less likely it can be successfully automated. And target processes that add value and are lean. Don’t fall into the trap of automating a bad, broken, inefficient, or non-value-adding process just because you can. Prioritizing in this way may mean you’ll start on a smaller scale, but you'll be starting in areas that are most impactful and that align with your North Star. That helps establish some early wins and build momentum quickly.

Get the right combination of automation technologies

From a sophistication standpoint, automation falls on a continuum—from RPA to advanced analytics to cognitive automation to AI (see figure)—but that doesn’t necessarily mean your adoption must sequentially follow that continuum. Realizing the full value of intelligent automation comes from integrating the right elements of the continuum to maximize overall impact. That means matching the right solution to the right process, which is typically driven by the nature of the data (is it structured or unstructured?) and the nature of the process (is it more rules based or judgment based?). Some processes may benefit more from RPA than AI, while others could be best served by a solution comprising three or four different automation technologies.

Although AI has made great strides recently, companies are still in the very early days of trying to figure out what it is, how to operationalize it, and how to quantify the benefit and value to their organizations. A lot of decentralized experimentation is happening, albeit on a very small scale. In fact, companies very rarely have a broader, enterprise-wide strategy in place—and that eventually will be needed for AI to truly scale.

Also needed will be a change in mindset. Companies must stop focusing tactically on incremental cost savings or headcount reduction, stop thinking in silos or in the context of specific steps of a process, and avoid looking at AI as a technology or software to be deployed. Instead, they need to see AI for what it is: a driver of fundamental, end-to-end transformation of the workforce, business processes, and overall enterprise.

Contact us to learn how AI can power your journey to intelligent operations.

Popular Tags

    More blogs on this topic