ERP Transformation Was Yesterday. AI Agents Redefine How The C-Level Must Lead Change

             

Every C-level executive knows the playbook for ERP implementations. It is linear, structured, and finite: define the end state, prepare for go-live, enforce adoption. ERP has one clear destination, built once, then maintained.

AI agents change the rules. They arrive incrementally, taking over routine tasks before exp anding into entire workflows. Their share of work grows as capabilities improve and as people learn to trust them. This is dynamic and constantly evolving.

For C-level leaders, the implications are profound.

---

1. Principles Over Blueprints

The end state cannot be pre-designed. Instead, guardrails need to be set:

• CFOs must define where financial judgment must remain human, versus where agents can take over transactional reporting or reconciliations.
• CIOs must establish standards for data quality, ethics, and integration—ensuring agents grow within safe boundaries.

---

2. Trust as the Core KPI

ERP success is about adoption; AI success is trust.

• CFOs must not only track error rates, but whether controllers trust agent-generated forecasts enough to act on them.
• CMOs must measure whether teams rely on agent-drafted proposals, or still redo them from scratch.

---

3. Continuous Communication

ERP has a “go-live” day. AI requires ongoing narrative management.

• CHROs must communicate role evolution clearly: “Agents will now draft 60% of onboarding documents; HR professionals focus more on talent engagement.”
• COOs must set expectations that operations will continuously shift as agents assume more scheduling, monitoring, and reporting.

---

4. Identity and Workforce Evolution

ERP never questions your identity—people still work in finance, HR, or procurement. AI redefines role and identity itself.

• CHROs must build reskilling programs that help employees move from task execution to oversight, analysis, and relationship-driven work.
• CEOs must signal that this is not an efficiency story but a adaptation story.

---

5. Adaptive Governance

ERP requires rigid change control. AI requires portfolio-style governance: sandboxes, rapid scaling of wins, and structured exits for failed use cases.

• CIOs must own the framework in partnership with business leaders.
• CEOs must sponsor enterprise-wide councils that ensure balance between innovation and risk.

---

The CEO’s Imperative

ERP is about installing systems - AI agents are about building an organizational capability: continuous adaptation.

The leaders who succeed will not treat AI as a one-time implementation but as an evolving partnership between humans and machines.

This article was first posted on LinkedIn on September 9, 2025.

The hidden economics of ride share: A strategy map

We take ride share for granted — open the app, choose a ride, swipe, done. But behind that simple interface is a complex and fast-evolving battle of economics and experience.

The recent Mary Meeker chart of the Waymo market share in San Francisco was linearly extrapolated into the future by Peter Gostev:

This begs a number of questions:

- Cost per mile: What does it take to deliver the ride?
- Perceived value: What does the rider get ?
- Price: What are consumers paying today for the value?

What It Really Costs to Move One Person One Mile

Most people don’t know what it really costs to run a ride. But under the hood, there’s a different story playing out — one where variable costs, fixed assets, and scale economics define who will win the future of mobility.

The current ride share leaders such as Uber and Lyft are variable-cost platforms where the drivers own the car and cover fuel, depreciation, and insurance, and the platform providers take ~20–30% of the fare. For the drivers, there is little to no scale efficiency per ride.

The Waymo cost base has high fixed upfront cost that are amortized over massive ride volumes. The cost per mile falls dramatically as usage increases.

Cost curves tell you who can win on price.

Which ride is actually worth your money?

Not all rides are created equal. Some offer luxury that doesn’t feel that luxurious. Others cost more but remove the driver entirely. Some are just cheap — and feel that way. So how do we make sense of all the options in rideshare?

You can map this out using a simple lens: value vs. price.

• Price = what you pay per mile
• Value = what you get for it (safety, predictability, speed, comfort, novelty)

When plotted, this reveals how well each product serves its customer relative to its price.

UberX and Lyft remain the sweet spot for mass-market utility. Other models such as MILES short term rental of the myriad of offerings by the incumbents serve a wide range of niche demands - and these niches can be quite large on their own. Waymo today justifies its premium in some markets — especially for safety- or tech-conscious riders.

To wrap up: The industry cost curve tells us who can scale profitably. Value-to-price tells us who earns customer love.

Put together, and we get the strategic battlefield for the future of mobility.