Decision intelligence: Thriving in the age of AI
To make magic with artificial intelligence, start with a clear business objective
September 20, 2024
Implementing artificial intelligence (AI) without confirming a clear business objective is like performing a magic trick without communicating what you’re trying to do.
“Pick a card, any card,” a magician will say while shuffling a deck. “Remember your card and then put it back in the deck.” The magician shuffles again and pulls out a single card. “Is this your card?” he asks. Of course it is, much to the audience’s delight.
Now imagine that same card trick without the interaction. The magician shuffles a deck of cards — silently. The magician shows the deck to the audience — again silently. The magician tries to get someone to pick a card…silently? It just doesn’t work.
The same is true for implementing AI without agreement on a clear business objective. Bringing in new technology for technology’s sake doesn’t work.
To implement AI, an organisation must build (or rent) a massive AI infrastructure, hire and train skilled AI workers, then develop data-rich AI models and train the system. Without defined business objectives, many of these resources can be wasted.
In fact, we see it far too often. Many companies have invested in AI, and many have also failed to achieve their desired returns. Consider that:
- 60% of all AI centre of excellence projects are scrapped within the first year because they fail to hit their return-on-investment targets.1
- 90% of all remote operations center projects with AI still unexpectedly require heavy human intervention.2
- 70% of “Big Brother” AI projects are not only ethically questionable, but they also fail to deliver their intended cycle-time improvements.3
To thrive in the age of AI, you need a clear purpose. Only then can you drive what we call continuous decision intelligence.
A virtuous chain reaction
An organisation with decision intelligence can empower its employees with actionable data to help them solve business problems. Leading AI implementers make sure they have only the datasets they need to build the models they want. Not all data is good data, and more data isn’t necessarily better.
As organisations with business objectives in mind have discovered, an effective AI implementation creates a virtuous chain reaction of continuous decision intelligence. Here’s an example of how this can work:
- Company X, using AI data, makes good decisions and gets great business results.
- Good returns enable Company X to invest in further AI innovations.
- These AI innovations lead Company X to make even more data-driven decisions that continue to foster strong business results, improving the customer experience and gaining a competitive advantage.
- And on it goes!
See it in action
Here are some examples from DXC Technology’s AI customers who are delivering on their defined objectives:
- A life-sciences company based in China uses intelligent key performance indicators powered by AI and equipped with embedded governance. The AI solution first captures and synthesizes edge-to-core data from multiple sources, then converts that data into analytical insights that enable data-driven decisions. As a result, the company is meeting its objective to improve clinical trial processes and innovate faster, continuing its commitment to develop new drugs for patients with unmet needs.
- An Australian provider of integrated services wanted to use data more strategically to inform and support better business decisions. A cloud analytics and AI platform provides a modern data warehouse and reporting solution to improve data access and alignment, and provides a foundation to scale up capacity that’s critical to the company’s growth. Better insights drive better decisions across the organisation.
- A German provider of aircraft maintenance, repair and overhaul services needed to predict potential defects that might occur during maintenance in order to accelerate and improve accuracy of its sales quotes. The AI-driven solution works with datasets far too large for workers to handle on their own and is key to estimating the tools, time and labor required to resolve defects — ultimately helping to improve customer service and drive smarter supply chains.
What these examples have in common is a clear sense of business objectives and a desired destination, which led to required results.
As with any good card trick, the secret to getting the results you need is in the details.
