Value
A successful pilot that never hits the P&L is not a win.
Teams celebrate a demo, a satisfaction score, or a week of faster drafts. The function still has the same headcount, the same cycle time, and the same cost per case. The value leaked between the pilot and the operating rhythm.
How I see it
A pilot is not a win until it hits the numbers
Value realization is the work after something works in a room. The new path has to become the default path. Old steps have to be retired. The metric has to sit on a report an operator already uses.
That requires a baseline taken before the change, a definition of the completed result, and a period where both paths can be compared. It also requires an owner who can stop people from using the old queue because it feels familiar.
Many AI programs die here. The software is fine. The operating model around it is not. Nobody updated the staffing plan, the SLA, or the quality review. The savings stay theoretical.
The useful artifact is a value sheet: what changed, what did not, what it costs to keep running, and what decision that evidence supports. Expand, hold, or stop.
How the engagement runs
What you actually buy.
A service is a sequence with an artifact at each step. It is not a transformation program you purchase as a bundle.
Baseline the current cost
Time, volume, error, and capacity as they are today — not a target invented for the deck.
An owner already manages the number.
A benefit story with no starting point.
Build a conservative case
Addressable share, implementation cost, and run cost. If it is not material, stop.
A number that survives a skeptical operating review.
An ROI slide that ignores the cost to operate.
Get the win into the operating numbers
The result has to show up in cost, revenue, or capacity someone already reports.
The win is managed after the project team leaves.
Adoption of a tool as the only metric.
Do not buy this for
These engagements fail for predictable reasons.
Measuring usage instead of outcomes
Weekly active users of a copilot can rise while cycle time stays flat. Usage is a means.
Taking credit before the old path is closed
If both paths run, you have added cost. Value starts when the expensive path shrinks.
Waiting for perfect attribution
You need a credible before-and-after on the workflow you changed, not a finance model of the entire company.
Before we start
Questions that decide whether to engage.
Which operating number will move if this works, and who already owns that number?
If the answer is innovation KPIs, the value will not be managed.Do you have a baseline from before the change?
A story about how slow it used to feel is not a baseline.Has the old path been retired or capped?
If volume can still leak to the old path, the win is optional.What does it cost to keep the new path running for a year?
Gross savings without run cost is not realized value.What decision will you make if the number does not move in 90 days?
If there is no decision, this is a showcase, not value work.
How we size it
Realized valueverified change in cost, revenue, or capacity − incremental run cost − the cost of the old path that is still in use
If the last term is not near zero, you have not realized the value. You have added a parallel process.
This is the wrong engagement if
Do not start this engagement yet.
- There is no baseline and no appetite to take one now.
- The pilot owner cannot change how the function is staffed or measured.
- Leadership wants a success story more than a hold-or-stop decision.
- The initiative never reached production, so there is nothing to realize.
A useful next step
Bring the initiative you already launched.
We will look at the baseline, the current path, and whether the win is real enough to expand, hold, or stop.
Discuss an AI opportunity