What Happens When You Add AI to a Bloated, Value-Eroding Company? You May Lose Trust in Legal Operations and Decision Quality in R&D

In 2019, I began a Lean Management development and transformation journey that would last roughly two years. I was learning Lean while simultaneously leading the organizational transformation itself.

One of the first questions my mentor Hakan taught me was remarkably simple:

"Does the organization really need this?"

Years later, that question feels more relevant than ever.

Today, companies are calculating how much work AI can perform, how many hours it can save, and how many roles can be consolidated as a result.

Perhaps we should run a different calculation first:

How much of that work should never have existed in the organization in the first place?

The impact of AI on employment is now being debated intensely. Companies are increasingly linking restructuring, workforce reductions, and productivity initiatives to AI adoption. Yet the broader evidence does not support a simple conclusion that AI is already replacing human labor at massive scale.

So perhaps the most important management question is not:

"How many people can AI replace?"

It is:

"How lean was our organization before AI arrived?"

Turning 30 Hours of Waste Into 8

Consider an illustrative case.

Every Monday, a company prepares a 42-page performance report for its management meeting.

Sales extracts data from the ERP system. Finance reconciles the numbers. Operations sends another spreadsheet. An analyst consolidates everything. Department heads add their comments. Someone builds the presentation.

Total effort: approximately 30 hours.

Then the company introduces AI.

Data collection, reconciliation, summarization, and presentation preparation become increasingly automated. The same report can now be produced in eight hours.

Excellent.

Nearly a 70% reduction in time.

Then someone asks:

How many of those 42 pages do we actually use?

Suppose an examination of the previous six months of meetings shows that management regularly looks at only seven pages—and that actual decisions are generated from perhaps four.

The rest have been produced for years.

Because they have always been produced.

What did AI accomplish?

It turned 30 hours of waste into eight hours of waste.

It created technological efficiency.

But management efficiency?

No.

Lean thinking forces us to distinguish between activity and value. Performing work faster does not make unnecessary work valuable.

If AI can make a process 70% cheaper, perhaps the first question should not be how quickly we can deploy it. The first question should be: Why are we doing this process at all?

A Smaller Organization Is Not Necessarily a Leaner Organization

AI can increasingly enable one person to perform work that previously required three.

But what if the same organization still carries unnecessary reports, duplicate controls, five-step approval chains, poor KPIs, repetitive data entry, interdepartmental waiting, and low-value meetings?

What actually changed?

The number of people.

Not the management system.

Speeding up a bloated organization with AI does not make it lean. It automates the waste.

There is an even more uncomfortable possibility.

Companies may be eliminating the people who perform the waste without eliminating the waste itself.

That is not organizational transformation.

It is a smaller organization carrying much of the same structural burden.

AI Accelerates Bad KPIs Too

Now consider an illustrative legal and collections department.

Management establishes performance targets:

Open a minimum number of enforcement cases every day.

Win a minimum number of cases.

Complete a minimum number of collections.

Bonuses are linked to these targets.

The system looks disciplined and measurable.

But over time, employees naturally begin optimizing what management measures.

Then AI arrives.

Files can be reviewed faster. Documents can be processed automatically. Standard drafts can be generated. Research accelerates. Human capacity is no longer the same bottleneck.

Management responds by raising the target.

Three files become ten.

The dashboard turns green.

But the market may be telling a very different story.

Customers may begin to perceive the company as rigid, overly procedural, and too quick to escalate problems legally.

A prospective customer who has not even entered into a commercial relationship with the company may already be asking:

"If something goes wrong, will I even be able to reason with these people?"

The legal department can hit every KPI while the company quietly consumes its trust capital.

Legally correct.

Successful against the KPI.

Fast from a process perspective.

Wrong for the company.

You can optimize the performance of one department while destroying value for the enterprise as a whole.

Because there is an important distinction:

Opening a case is output. Collecting the money is a result. Doing so with the right cost, time, risk, and commercial judgment is value.

AI does not resolve that distinction for management.

AI does not fix bad KPIs. It makes them easier to achieve.

And that creates one of the most dangerous management traps of the AI era:

We may increase the efficiency we can measure while destroying the value we do not measure.

More Products—or More Value?

The same problem appears in R&D and product development.

AI can generate dozens of product ideas, variants, concepts, technical alternatives, and market possibilities.

Managers naturally begin asking for more.

More ideas.

More prototypes.

More products.

But what if the company's problem was never a shortage of ideas?

An experienced R&D or commercial leader may look at one of a hundred technically viable alternatives and say:

"We can build it. But this product will not work in the market."

That judgment may incorporate years of accumulated knowledge about customer behavior, manufacturability, channel acceptance, margin, quality risks, cannibalization, and the difference between what customers say they want and what they actually buy.

Much of that knowledge is tacit.

And that is where human experience becomes more—not less—important.

AI generates options. Experience distinguishes which option fits the context.

AI is dramatically expanding our answer to:

"What can we do?"

The value of management is increasingly shifting toward a harder question:

"Which of these things should we do?"

As the cost of producing unnecessary things falls, the ability to choose not to produce them becomes more valuable.

Let's Pull a Rabbit Out of the Hat

The answer is not to slow down AI.

Quite the opposite.

The answer is to use AI in the right sequence.

A simple management sequence consistent with Lean's focus on value and waste is enough:

Eliminate first.
If you cannot eliminate it, simplify it.
Then automate it.
Only then augment it with AI.

Before applying AI to a process, ask four questions:

Does the organization really need this?

Does the customer receive more value as a result?

Does this KPI measure activity—or enterprise value?

Is AI improving the work, or merely multiplying it?

The purpose is not simply to consume fewer resources.

It is to create the value customers and the enterprise actually need with less waste.

The First 90 Days

Days 0–30 | Question the Need Before the Work

Identify the 20 white-collar processes consuming the most organizational time.

Include reports, meetings, approvals, controls, proposals, recurring analyses, and repetitive administrative work.

Do not begin with technology.

Begin with:

"Why are we doing this?"

Days 31–60 | Eliminate and Simplify

For each process, make a deliberate decision:

Eliminate.
Simplify.
Automate.

At the same time, review the KPIs surrounding those processes.

Test whether metrics designed to reward activity are inadvertently damaging customer value, commercial relationships, decision quality, or enterprise value.

Days 61–90 | Put AI on Clean Processes

Scale AI only after unnecessary work and process steps have been removed.

Then stop measuring performance primarily through:

"How many did we produce?"

Measure customer value, quality, cycle time, total cost, commercial outcomes, and decision quality instead.

Conclusion

The answer to AI is not to slow technology down in order to protect people from it.

AI is extraordinarily powerful.

And precisely because it is powerful, management must be more careful about the operating model onto which it is deployed.

AI does not turn a poorly designed organization into a well-designed one.

It may simply allow the poorly designed organization to operate faster.

As unnecessary output becomes cheaper to produce, the discipline not to produce it becomes more valuable.

The human competitive advantage in the AI era will not be the ability to produce more than machines.

It will be the ability to understand what is worth producing, what should be stopped, and which decision protects the total value of the enterprise.

Editorial and Intellectual Property Note

This article is an editorial analysis of artificial intelligence, Lean Management, organizational design, performance management, and management productivity. It does not constitute an audit, valuation, legal opinion, investment recommendation, financial, technical, or management consulting advice, or a definitive performance assessment of any specific company, person, or institution.

The 42-page management report, legal/collections, and R&D examples are illustrative scenarios created to explain management problems. The personal section concerning the Lean Management development and transformation journey beginning in 2019 is based on the author's own management experience.

Publicly available data and sources were evaluated as of August 12, 2026. The original expression, conceptual structure, classifications, examples, application logic, and overall presentation of this article belong to Orkun Akçasarı. The Eliminate → Simplify → Automate → Augment sequence is not presented as an original or proprietary model; it is used here as a management sequence consistent with Lean principles concerning value and waste.

No monopoly is claimed over abstract ideas or methods beyond what applicable law recognizes. Protection applies to the original expression, arrangement, and overall presentation. Rights are reserved against unauthorized reproduction, adaptation, republication under another name, or commercial use in training, consulting, software, artificial intelligence systems, reports, presentations, or similar applications. Short quotations should identify the author, article title, publication date, and active publication URL. Specialist legal advice should be obtained where specific disputes, registration, or licensing issues arise.

© 2026 Orkun Akçasarı. All rights reserved.

References

Lean Enterprise Institute. "What is Lean?" Accessed August 12, 2026.
https://www.lean.org/explore-lean/what-is-lean/

Lean Enterprise Institute. "Waste." Accessed August 12, 2026.
https://www.lean.org/lexicon-terms/waste/

Lean Enterprise Institute. "Value." Accessed August 12, 2026.
https://www.lean.org/lexicon-terms/value/

Lean Enterprise Institute. "Lean Thinking and Practice." Accessed August 12, 2026.
https://www.lean.org/lexicon-terms/lean-thinking-and-practice/

Katrina Appell, Lean Enterprise Institute. "Why It's Better to Focus on Value, Not Waste." January 20, 2021. Accessed August 12, 2026.
https://www.lean.org/the-lean-post/articles/why-its-better-to-focus-on-value-not-waste/

MarketWatch. "AI Is Killing Worker Confidence, but There's No Sign It's Muscling People Out of Their Jobs on a Massive Scale." August 12, 2026. Accessed August 12, 2026.
https://www.marketwatch.com/story/workers-are-worried-but-theres-no-sign-ai-is-muscling-people-out-of-their-jobs-on-a-massive-scale-8eb92638

LinkedIn Corporate Communications. "New LinkedIn Data: 90% of C-Suite Leaders Are Continuously Building Skills to Lead in the AI Era." June 2, 2026. Accessed August 12, 2026.
https://news.linkedin.com/2026/new-linkedin-data--90--of-c-suite-leaders-are-continuously-build

Business Insider. "Oracle Has Drawn Up Plans for a New Round of Layoffs This Month, Sources Say." August 12, 2026. Accessed August 12, 2026.
https://www.businessinsider.com/oracle-is-planning-another-round-of-job-cuts-this-month-2026-8