Companies have invested hundreds of billions of dollars in AI. But if you ask most executives about AI right now, the conversation quickly turns to one question: Where is the return?
I hear it in nearly every conversation I have with customers. Leaders are investing in and experimenting with AI. However, when it comes time to show the impact of those investments, especially to CFOs and boards, the answer is often less clear than expected.
That’s not because AI isn’t delivering value. It’s because many organizations are still looking for value in the wrong places—and expecting it to show up too quickly.
Across industries, I see the same pattern. Organizations default to measuring productivity and labor cost savings. When those signals are modest or slow to appear, momentum fades. Initiatives stall. What started with energy and intent gets labeled a pilot and never quite scales.
What’s becoming clear, both in our work with customers and in our own transformation at Microsoft, is that the issue isn’t the technology. It’s how we define and measure value.
AI’s impact rarely shows up first as efficiency. It shows up in greater insight, more predictive power, in-task skill building, and the ability to evaluate more scenarios before acting. Those gains don’t fit neatly into traditional metrics, and they don’t map cleanly into cost reduction.
That gap is where many organizations get stuck. Leaders need to stop measuring AI activity and instead understand how AI is changing the way the business performs. Too many leaders measure the ROI of AI through the lens of cost savings. Instead, should focus on how AI can make their business better.
The most important question isn’t where to deploy AI. It’s what outcome matters most to your business, and whether AI can help move it.
That means grounding every AI effort in a clear priority. It could be increasing revenue per salesperson, improving customer retention, accelerating product development, or reducing risk. In some cases, cost matters. In many, it’s not the primary driver.
Organizations that focus too much on the tools often end up with pockets of activity but little business impact. Some leaders describe this as “pilot purgatory.” In contrast, companies that make real progress start from a different place. They define the outcome first, then work backward to where AI can make a meaningful difference.
We’re seeing this across industries. Some teams are using AI to shift from reactive audits to proactively identifying risk. Others are focused on catching system vulnerabilities earlier, before they escalate. Even in areas like sales, teams are rethinking how they prepare for customer conversations, while engineering groups are designing products that anticipate customer needs rather than respond to them. We are starting to call this AI’s “capability add,” an ability to create new strategic value from work processes, in addition to optimizing the efficiency, speed, and quality associated with the original process outcomes.
The discipline is simple but not widely practiced. Leaders must be explicit about the value you are trying to create, and measure against that from the beginning.
The trap of measuring AI usage
Even when leaders align AI to business goals, a second challenge shows up quickly: timing.
The outcomes that matter most, like revenue growth or margin improvement, take time to appear. But leaders can’t wait months to understand whether they are on the right track.
This is where many efforts to measure the impact of AI fall short. Adoption becomes the default signal: how many people are using the tool, how often, and where. These are useful indicators, but they are only activity metrics. They tell you something is happening, not whether it’s working.
What matters more is whether AI is changing how work gets done—and if those changes are moving the business forward.
That means asking a different set of questions. Are people using AI in ways that align with the outcome you defined? Is it changing how they spend their time? Are those changes improving the quality or speed of their work? And are those improvements starting to show up in business performance?
This dynamic plays out among many sales teams. The goal isn’t tool usage. It’s higher revenue per seller. AI can reduce administrative work and improve how teams engage with customers. But the real signal worth measuring is whether that leads to more time with customers, stronger pipelines, and better win rates. Revenue will follow, but it takes time. These leading indicators show up much sooner.
When you connect adoption to outcomes, measurement becomes far more useful—not just for reporting results, but for making decisions along the way.
From measurement to momentum
This crucial shift in measurement ultimately comes down to leadership.
First, leaders need to define what success actually looks like. Not in broad terms, but in specific outcomes. What are we trying to move? How does this translate into specific success metrics? And how will we know if we are making progress?
In many organizations, that clarity is still missing. Teams are experimenting with AI, but without a shared understanding of the outcome they are driving. Without that, measurement becomes fragmented and progress is hard to see.
Second, leaders need to align on what counts as value. In nearly every conversation I have with customers, there is still pressure to show ROI primarily through labor cost savings. It is a familiar model, and in some cases it will matter. But in most, it is not where the most meaningful gains are emerging.
The people closest to the work often see the impact first—in speed, quality, decision-making, and customer engagement. The challenge is translating those gains into signals that leadership can recognize and act on. This requires a shift. Leaders need to make those signals visible and legitimate, not dismiss them because they don’t fit traditional models.
Third, leaders need better visibility into how work is actually changing. In manufacturing, for example, nearly every step can be measured. In knowledge work, far less is visible. We can often track tool usage, but it is much harder to understand how work is changing. How people spend their time, how decisions improve, how outcomes evolve. Agentic capabilities and new telemetry will make this easier over time, but the importance of defining success metrics and mapping early indicators to those that show up later will remain critical. Done well, this is what turns measurement into momentum. It allows organizations to move from isolated experiments to repeatable performance, building a clearer view of what’s working and scaling it across the business.
I believe AI isn’t failing to deliver value—organizations are struggling to see it because they’re measuring it the wrong way. The leaders pulling ahead are clear about the outcomes that matter, disciplined in how they track early signals of progress, and intentional about turning those signals into repeatable performance.
As AI reshapes how work gets done, the advantage won’t come from adoption alone, but from how well organizations connect that adoption to real business impact—and how quickly they learn and scale what works.
Hence then, the article about what leaders get wrong about the roi of ai was published today ( ) and is available on Time ( Middle East ) The editorial team at PressBee has edited and verified it, and it may have been modified, fully republished, or quoted. You can read and follow the updates of this news or article from its original source.
Read More Details
Finally We wish PressBee provided you with enough information of ( What Leaders Get Wrong About the ROI of AI )
Also on site :
- Kevin Hart Roast: Comedian Ends Katt Williams Beef as The Rock, Teyana Taylor and More Cracks Jokes About Hart’s Height and Filmography
- 'Euphoria' Shows Sydney Sweeney Masturbating and Topless on OnlyFans
- Chelsea Handler and Shane Gillis Exchange Heated Jabs at Kevin Hart Roast