Efficiency Gains or new Value: What Can Your AI Strategy Deliver?

AI: How to Move from Optimization to Transformation

You already use artificial intelligence (AI) to speed up certain tasks, reduce costs or free up time. That’s a good start. However, if you don’t take your AI strategy further, you could miss out.

The real issue is asking only “Where can we use AI?” rather than “Based on the opportunities it creates, what would we like to become?”

An organization can adopt multiple tools and usage cases without tangibly transforming its work processes. AI can enhance what you already do. It can also pave the way for completely revamping your processes and, most importantly, considering what more you can do.

Don’t equate speed with transformation

Let’s take a standard example: accounts payable processing. An invoice is received and opened. Information is entered into a system, verified and compared to the purchase order before it is authorized and forwarded for payment.

In order to speed up the process, you could integrate AI into certain steps. Time spent on data entry would be reduced. Certain verification steps would be faster. Productivity would increase. However, the process would remain essentially the same.

A more ambitious approach could involve rethinking the entire process. Automation could allow an invoice to be received, analyzed, validated and forwarded automatically, and human intervention would be required only where there is uncertainty or an anomaly.

This would make a big difference. The first result is that AI improves the current approach. The second result is that it leads you to rethink the workflow.

This is where the transformation stage begins.

How far would you like to go?

This depends on what you’d like to accomplish. Initially, AI can be used to improve what is already in place. Then it can pave the way for more far-reaching changes.

On a basic level, it enhances what we do. This includes working faster, reducing errors, making better use of data, personalizing interactions and supporting decision-making.

It can also enhance what you offer. A service becomes more personalized, a company anticipates some of its clients’ needs or a public organization enhances the user experience.

Indeed, technology can help you to undertake a more indepth transformation of your service offer or how you fulfill your mission. This could open up a new market, new capabilities and different partnerships.

Not all organizations need to go that far, and not all projects have to be transformative. Consciously choosing the direction you want to take rather than allowing the available tools to dictate your path is important.

What would you do with the added gains?

Let’s revisit the example of accounts payable. Where automation significantly reduces the workload, an organization can convert this saved time into cost savings. It can also be reinvested elsewhere.

For example, the finance team could focus more on analysis or finally create a dashboard that managers requested a long time ago.

The same applies to manufacturing companies. Where AI optimizes cutting an expensive material and minimizes waste, the recovered material could be used to manufacture another product, boost production or expand into another market.

The savings made can therefore free up resources to take things further. Bear in mind that each improvement doesn’t necessarily have to result in a major transformation.

When doing better is no longer enough

There are blind spots when you focus solely on optimization.

First, where competitors have access to the same tools, working faster or at a lower cost usually gives an organization a competitive edge for a short time only.

Leaders should also avoid constantly improving work processes that may no longer meet the company’s needs.

Lastly, if you consider AI based on cost alone, you risk focusing your efforts on what needs to be reduced rather than what can be improved, reimagined or developed.

What your data can reveal

Adopting generative AI is not the only path to potential gains. Analysis, prediction, recommendation and optimization can reveal insights that would likely not be evident otherwise.

The Walmart example speaks for itself. When the company analyzed its sales figures, it found that, in the days leading up to a hurricane, customers at certain stores in southern U.S. states purchased large quantities of Pop-Tarts, in addition to emergency supplies. Instead of simply noting this trend, the retailer placed the Pop-Tarts close to survival items, which led to even higher sales.

Carry out a ratio test

In order to view the progress of your strategy, you should take a look at the AI projects you have launched. Are they primarily used to reduce costs, shorten a timeline or automate a manual task? Do some projects enhance your customer or user experience? Have others created opportunities that you hadn’t previously explored?

There is no ideal ratio. An organization in the early stages of the process will likely have more optimization projects, which is to be expected.

However, if almost all your initiatives remain focused on internal productivity, you may need to reassess your strategy.

AI reveals its true potential when efficiency is no longer an end in itself and it becomes the starting point for a more ambitious transformation.