The pandemic has sparked a significant increase in AI adoption, with 52% of companies accelerating their plans in response to the crisis. But even as businesses scaled their AI rollouts in 2021, a report from O’Reilly’s revealed key obstacles to tangible AI adoption and application within the enterprise.
The biggest challenge companies faced was a lack of skilled talent, with difficulties in identifying appropriate business use cases close behind. The bottom line?
Many businesses just don’t have the infrastructure to make their visions for AI a reality.
At the same time, critical business challenges from supply chain to inflation are making the forecasting and cost efficiencies promised by AI more urgent than ever.
As companies chase these benefits, a new technology is beginning to move the needle on AI adoption. Decision intelligence is a commercial application of AI that’s helping businesses overcome common barriers to putting AI models into practice.
Decision Intelligence (DI) platforms are helping workers across business verticals make AI-informed decisions in their day-to-day work.
With Gartner predicting that over a third of large organizations will be using DI by 2023, getting to grips with this technology will be critical to ensuring the majority don’t get left behind.
But what do we really mean by Decision Intelligence? And how can businesses use it to cross the threshold to applied AI?
Understanding Decision Intelligence
Decision making is at the heart of every business so it makes sense to put it at the center of commercial AI adoption. DI helps ensure consistent, accurate and rapid decision making by harnessing the predictive power of machine learning (ML) and embedding this into the pursuit of an outcome. This means that a Decision Intelligence solution is built with an objective in mind. This might sound like common sense, but the current approach is too often ‘bottom up’: what data do we have, and what can we do with it? Decision Intelligence starts with the end in mind – Decision Intelligence applications are built for a specific purpose.
Decision Intelligence gives an AI strategy purpose – something even the businesses most advanced in their application of the technology struggle with…
In a context where 76% of commercial machine learning models never make it into production, let alone make a commercial impact, the need and potential for this technology is striking. Decision Intelligence can be leveraged by just about every business in every sector to optimize across an entire value chain. It can route delivery vehicles more efficiently, advise on an optimum markdown price, or even recommend associated products to website users.
Decision Intelligence or DI stops teams throughout an organization from making a ‘best guess’; it means they can stop hoping, and start knowing.
Since it is outcome focused, Decision Intelligence gives an AI strategy purpose – something even the businesses most advanced in their application of the technology struggle with – and ensures activities are focused on adding business value. It will, ultimately, be the technology that facilitates mass commercial AI adoption, and helps every department and every individual within an organization leverage AI daily.
The Path to Adoption For Businesses
Businesses looking to implement Decision Intelligence need three core things to make it happen; an AI-ready dataset, a centralized intelligence unique to their business, and technology that surfaces that intelligence to non-technical teams.
That last point is important.
As businesses struggle to hire the talent needed to drive AI adoption, Decision Intelligence can deliver actionable AI insights to workers who lack data-specific skills. It means that commercial users can interact directly with a model, informing and helping to iterate it. It can bridge the gap between the theory of data science and the practice of decision making; fundamental for a discipline that is focused on driving business outcomes.
But, perhaps the biggest difference between traditional AI applications and Decision Intelligence is that the former aren’t tailored to a specific business. If businesses want to leverage AI for competitive advantage, then they can’t use generic or shared models. Each business is unique: it has its own customers, products, data and logic. One generic intelligence for multiple businesses simply can’t deliver.
To date, the industry has focused on standardizing AI with point solutions to optimize websites, plan routes and forecast demand. But it’s the creation of that AI that should be standardized, not the AI itself.
DI Adoption Beyond Big Business
Gartner predicts over a third of large organizations will be using DI by 2023. But as we think about the bigger picture for DI adoption, it’s important to consider the huge economy of small-to-medium sized businesses, as well as sectors that lag behind in AI-readiness.
It’s clear that vital industries like healthcare and education would benefit from DI, especially in an increasingly unpredictable world. But even as AI adoption advances, big companies in the tech and finance sectors have far and away the biggest share of the pie. This is not surprising, as these sectors already have the data literacy and skills needed to start machine learning projects.
But how do we bridge the gap?
The transition could look similar to what we saw with CRMs in the early 2000s.
At that point, the majority of companies were building CRMs in-house. Today, you’d be hard pressed to find a business that didn’t buy its CRM off the shelf. Decision Intelligence is likely to follow a similar trajectory, with businesses moving to licensing their DI platforms instead of building them from scratch, decreasing time to value and cost of ownership.
But whatever the road ahead, organizations are going to have to become data literate as AI is more widely adopted, and the gulf between those that can and those that can’t increases. With much of Decision Intelligence’s potential still uncharted, the more businesses that implement it, the more we’ll discover its power in the face of today’s biggest challenges — from climate readiness, to resilience in the face of continued global business uncertainty.