Companies are spending a lot of time trying to figure out where AI fits into their business. What can we automate? What can AI take over? Which roles will it change?
Those are fair questions. But there’s another question that may be much more useful:
What are your most valuable employees spending time on that doesn’t actually require their expertise?
We recently saw this with a client whose clinical director was responsible for reviewing assessments before they moved forward. The director’s expertise was valuable, but part of that review involved catching basic errors, identifying missing information, and sending assessments back for corrections.
Our co-founder, Nesanel Moeller, put it simply: “You’re wasting the director’s hours.”
That didn’t mean the director shouldn’t review the assessments. It meant the director shouldn’t be the first person—or system—to catch problems that don’t require a clinical director to identify.
And that’s exactly where AI can become incredibly useful.
The Hidden Cost of Low-Value Work
This problem isn’t unique to clinical directors. It happens across businesses every day.
Senior managers chase down missing information. Experienced employees review documents for basic errors. Executives spend hours searching through emails and meetings for updates. Specialists manually summarize information before they can actually analyze it.
Individually, these tasks may not seem significant. But when they consume the time of your most experienced employees, they become very expensive.
Microsoft’s 2023 Work Trend Index found that the average employee using Microsoft 365 spent 57% of their time communicating and only 43% creating. The same research found that 62% of respondents struggled with spending too much time searching for information, while 68% said they didn’t have enough uninterrupted focus time during the workday.
That’s a lot of valuable time being consumed before the highest-value work even begins.
AI Doesn’t Have to Replace the Expert
A lot of conversations around AI still focus on replacement: Can AI replace this role? Can we eliminate this position? Can we reduce headcount?
But that framing can cause businesses to overlook some of the most practical AI opportunities available today.
Instead of replacing the expert, use AI to make sure the expert only spends time on work that requires an expert.
In the clinical director example, AI could perform an initial assessment review. It could check whether required information is missing, identify obvious inconsistencies, flag something that needs clarification, or catch basic errors that should be corrected before review.
If there’s an issue, the assessment can go back for corrections before it reaches the clinical director. The director still makes the decisions that require clinical judgment. They just don’t spend as much time doing the work that comes before those decisions.
Think of AI as the First Pass
This is a useful way to think about AI across many areas of a business. AI doesn’t necessarily need to own the entire process. It can simply take the first pass.
For example:
-
An AI agent can review a document for missing information before a manager sees it.
-
AI can summarize a long customer history before an account manager prepares for a call.
-
AI can categorize and prioritize incoming requests before an employee reviews them.
-
AI can compare information against a checklist and flag exceptions.
-
AI can prepare a first draft of a report before an expert adds analysis and recommendations.
The human remains responsible for the judgment. AI removes some of the work required to get to the point where that judgment is needed.
Research is beginning to show how significant that kind of assistance can be. A study by researchers at Stanford and MIT, published by the National Bureau of Economic Research, examined more than 5,000 customer support agents using a generative AI assistant. On average, access to AI increased productivity by 14%, measured by customer issues resolved per hour. The gains were even larger among newer and lower-skilled workers.
Importantly, the AI wasn’t replacing the customer support agents. It was helping them perform the work more effectively.
That distinction matters.
Measure AI ROI in Capacity, Not Just Headcount
If an AI implementation doesn’t eliminate a position, does that mean it failed to generate ROI?
Of course not.
Imagine a senior employee gets five hours back every week. That time could go toward serving more customers, coaching employees, solving more complex problems, improving customer relationships, or finally tackling initiatives that have been sitting on the back burner.
That recovered capacity has value.
In fact, Microsoft’s Work Trend Index found that when managers were asked what they would value most from AI in the workplace, they were twice as likely to choose increasing employee productivity as reducing headcount.
McKinsey has also estimated that generative AI and other existing technologies have the technical potential to automate activities that currently absorb 60% to 70% of employees’ time.
That doesn’t mean 60% to 70% of jobs disappear. Jobs are made up of many different activities. The opportunity is to identify which of those activities technology can handle so employees can spend more time on the parts where humans create the most value.
Start With Your Most Expensive Repetitive Work
You don’t need to start an AI initiative by looking across your entire company. Start smaller.
Look at the people whose time is particularly valuable and ask:
What are they doing repeatedly that doesn’t require their level of expertise?
Are senior employees checking work for basic errors? Are managers repeatedly gathering information before making a decision? Are specialists manually preparing data before they can analyze it? Are executives searching for information that could have been summarized automatically? Are experienced employees answering the same internal questions over and over again?
Those are often much more useful AI opportunities than starting with a broad goal like, “We need to implement AI.”
Your Experts Should Be Doing Expert Work
The goal of AI shouldn’t automatically be to remove people from your business. Sometimes the much better opportunity is removing the wrong work from their day.
Your clinical director should be applying clinical judgment. Your sales leaders should be coaching and closing. Your managers should be managing. Your specialists should be solving the problems that require the experience you hired them for.
AI can increasingly handle some of what happens around that work: gathering, checking, summarizing, categorizing, preparing, and flagging.
The question isn’t only, “What can AI do for us?”
A better question might be:
“What are our best people doing that AI could help them stop doing?”
That’s where some of the most practical AI opportunities may be hiding.