When confronted with a big technological shift, it is easy to focus on the obvious changes. New tools. New workflows. New AI agents that make the future feel suddenly closer. But some of the most important changes are subtle. They hide in plain sight, inside concepts we already think we understand.
Take online search. We type into a box, scan a few snippets, click what looks promising, and move on. The whole experience is designed around a human who can browse, compare, interpret, and decide, and the web has adapted to that model for decades.
But AI agents do not search like humans. They do not care whether the page layout is elegant or how many links sit above the fold. They want reliable, structured, low-latency, action-ready information that fits inside a workflow. 200 milliseconds of latency feels right to a human; for an AI working in sub-millisecond increments, it is a showstopper.
Even though "search" looks the same from the outside, the engine an agent reaches for is a different product than the one a human reaches for.
The same thing is about to happen to value management. The name may stay. The discipline will not. AI does not only change how work gets done. It also changes how quickly companies can move from insight to action. That makes the strategic question more important, not less.
Before companies can manage value, they need to know what kind of value they are trying to create. Cost? Working capital? Margin? Revenue? Risk? Better use of scarce capacity?
If agents act faster than organizations can decide, the problem is not a lack of automation. It is a lack of value discipline. That is why value management has to change. It cannot remain a periodic exercise around business cases, dashboards, and end-of-program reviews. In the age of AI, it has to become a continuous operating discipline.
The episodic version we have all gotten used to
Today, most companies still treat value management as episodic. A business case at the start. A benefit estimate before a transformation. A steering dashboard during the program. A realization review at the end. Aligned with the yearly planning cycle.
But as AI agents move from assisting work to executing parts of it, the cycle between insight and action compresses. Opportunities surface faster, recommendations land faster, changes get drafted, scoped, and sometimes executed faster.
That acceleration sounds exciting. But it raises the stakes. Executing faster is like driving faster: it is only fun as long as the steering and navigation can keep up.
If value management stays on its annual rhythm while everything around it speeds up, the result is not faster transformation. It is a louder version of the same problem: a lot of activity, little validated impact.
What value management has to mean
For us at SAP Signavio, value management is the practice of continuously identifying, qualifying, prioritizing, realizing, and validating business value from transformation opportunities, grounded in how the business actually runs. That last part matters.
It lives in operational reality. If it does not connect ambition to real process behavior, system data, business ownership, and financial impact, it remains a slide.
It also has to keep two things separate that companies routinely conflate. A value opportunity is not realized value. A business case is not. A dashboard is not. An AI-generated recommendation is not. Realized value is business-validated impact after operational change.
When organizations blur this distinction, early opportunity numbers get treated as commitments. The CFO sees them, the program is funded, and a year later the gap between promise and reality becomes someone's career problem. Treat those numbers as promises and you damage trust; distrust every number until it is fully proven and you miss opportunities.
This is rarely bad intent. Most of us have been in situations where a project needed a number to move forward, so a number appeared. But a number that gets a project approved is not the same as a discipline that helps a business realize value.
From episodic to continuous value management
One analogy I find helpful: the sales pipeline.
In sales, nobody confuses a lead with revenue. A lead may be interesting, a qualified opportunity promising, a late-stage deal likely. Until the customer signs, it is potential.
Transformation value works the same way. A process insight, a benchmark gap, a system recommendation, or an AI-discovered improvement is a lead, possibly a hot one. But it is not yet value; it needs qualification.
Is the opportunity real? How large is it for this customer, with its constraints, systems, and operating model? What is visible in the system, and what still needs human judgment? Who owns the change? This middle layer is where credibility is created or lost.
The pipeline metaphor also explains why a one-time business case is a poor steering instrument. It kicks off a project; it rarely starts a continuous practice. Assumptions fade, operating changes turn out harder than expected. The system goes live, the dashboard turns green, and the actual behavior of the business does not change enough. The value does not vanish in a single moment. It leaks.
That was already a problem before agents. With agents it becomes more pressing: the organization can find more opportunities, produce more recommendations, and trigger more downstream activity at higher speed, while value still leaks out just as quietly as before.
That is why, in the age of agentic AI, value management has to become a continuous operating discipline.
Value is financial, evidence is operational
Value management only becomes real when operational evidence connects to a business outcome someone will validate and own.
That outcome does not always start as a clean financial line item. Some value is direct: working capital, margin, revenue, cost. Some is enabling: process transparency, a shared language, better decision quality, a clearer view of how work flows. Enabling value should not stay vague; it improves the odds that the business makes better choices and realizes measurable impact.
Processes are not the unit of value, financial or financially defensible business impact is. But bottlenecks, variants, rework, exceptions, and policy violations are where value gets trapped, and process data makes them visible.
Many companies already measure plenty, with more dashboards and KPIs than anyone cares to check on a given day. AI agents will produce more measurements still. Without the right practice, ownership, and operating rhythm, the organization receives more information without making better decisions.
AI makes this unavoidable
All of this already matters without AI. Agentic AI makes it urgent. In the past, companies might identify a handful of transformation opportunities per year, slow enough that value management could remain manual, episodic, and detached from daily operations.
Agents will make it easier to move from analysis to recommendation to action, and the improvement cycle will compress. That is exactly why value discipline matters more: faster automation without validation creates faster disappointment.
This is where the search analogy returns. The search product an agent uses had to be rebuilt around the agent's actual constraints, even though "search" still meant the same thing on the surface. Value management is on the same path: the currency stays the same, financial impact grounded in process reality, but the product underneath has to change.
From value reporting to value operating
The clearest way to put it: value management has been a reporting practice, and it has to become an operating practice.
A reporting practice looks backward. It tells you what was promised, what was spent, and what number you can defend in the steering committee. An operating practice runs in the present tense: it surfaces what the business is seeing now, names the owner and the change, and defines how realization will be confirmed. The cadence is closer to a sales pipeline than a typical transformation portfolio. It also creates the discipline to say: this did not work, this assumption was wrong, this is how we adjust.
We now have line of sight to faster, denser, more continuous transformation. The companies that get this right will not be the ones with the most agents. They will be the ones who can take a continuous stream of process-grounded opportunities and turn them into validated impact at the same tempo as their agents.
The agents are not the moat. The discipline that grounds them is. Process insight is the lead; realized value is the signature, when the business changes how work gets done and the impact can be validated. That is the gap value management has to close.
This is where SAP Signavio plays a central role. Value does not live in plans, slide decks, or steering dashboards. It lives in the operational reality of your business: how orders flow, how exceptions are handled, how cash moves, how rework accumulates. Making that reality visible, and connecting it to financial impact and ownership, is what our Suite is for, and what value management has to be built on for the next decade.
The name does not have to change. The practice underneath it does. The point is not to add AI to yesterday's value management, but to rebuild the practice with SAP Signavio so humans and agents move at the same speed toward outcomes the business can validate.
With thanks to Lukas N.P. Egger for his input, review, and the discussions that helped sharpen this piece.