The Great Enterprise Software Repricing Has Started
Supply chain will be the first place we see who wins
A few days ago, SAP lost roughly $130 billion in market value from its peak in February 2025. The stock dropped to its lowest level in 17 months, and the decline wasn’t treated as “just another bad week.” It was interpreted as a signal — a public-market reaction to a growing fear across enterprise software:
AI may compress the value of traditional software and the services that come with it.
This isn’t an obituary for SAP. In fact, the most measured voices are clear: the worry isn’t SAP’s existential future — it’s the pricing power and services economics that historically came with large enterprise platforms.
And that distinction matters.
Because what we’re watching is not a collapse of enterprise software.
We’re watching a repricing of how value is created.
And nowhere will this become more real — faster — than in supply chain.
What the market is really saying (in plain language)
The Reuters coverage captures the anxiety very directly: if AI makes software modules easier to build and replicate, then the risk is that average selling price of services and billable hours falls.
That’s the headline beneath the headline.
For decades, enterprise software value was defended by friction:
- It took years to build complex modules
- Integrations took months
- Customization required specialists
- Rollouts took quarters or years
- Services and consulting became durable revenue engines
In that world, complexity wasn’t just a byproduct — it was a moat.
But AI changes the economics in a way that makes investors nervous:
If the “cost to build” drops and the “speed to copy” rises, then traditional moats get thinner. Not overnight — but inevitably.
Which forces a hard question:
If software becomes easier to create, what will enterprises pay a premium for?
The answer is already showing up.
Enterprises are not going to pay a premium for “more features.”
They will pay a premium for more certainty.
Why supply chain will feel the shift before everyone else
Supply chain is not a department where “good enough software” survives.
Because supply chain is where software directly touches reality:
- freight is moving or it isn’t
- inventory is available or it isn’t
- product is compliant or it isn’t
- the customer receives on time or they don’t
- a delay becomes an escalation in hours, not quarters
This is why supply chain leaders don’t romanticize technology.
They care about outcomes, not dashboards.
And they live with the consequences when outcomes fail.
So, when the enterprise software industry goes through a reset, supply chain is the first arena where “AI value” gets tested brutally:
Does it improve execution reliability?
Does it reduce exception workload?
Does it keep operations running through chaos?
Does it prevent failures instead of reporting them?
That’s the new standard.
Difference between the old vs. the new guards: bolt-on AI vs. AI-native execution
This is the important distinction that gets lost in the noise.
Most large platforms are pursuing bolt-on AI.
That means:
- AI copilots inside existing workflows
- AI summaries layered onto complex screens
- AI insights that still require humans to orchestrate execution
- AI features added to products designed for a pre-AI era
This approach helps, and it will absolutely improve many processes.
But it has a ceiling.
Because enterprises aren’t drowning due to lack of information.
They are drowning due to lack of execution bandwidth.
Bolt-on AI often improves decision support.
It rarely removes enough work from the system.
The new guard, on the other hand, is being built AI-native.
AI-native doesn’t mean “we use AI more.” It means the platform is designed from day one around a different objective:
Make the work disappear.
Less manual follow-up
Less coordination overhead
Less swivel-chair exception management
Less dependence on heroic humans
AI-native systems are built around closed loops.
They don’t stop at “insight.”
They proceed to:
Insight → Recommendation → Action → Confirmation → Learning
That loop is the product.
A simple example: the difference between “visibility” and “execution”
Let’s keep this real.
A traditional system might tell you:
“Shipment is delayed and temperature is trending high.”
That is visibility.
The question is: what happens next?
Does someone open a ticket?
Does someone call the carrier?
Does someone email the consignee?
Does someone escalate to security?
Does someone reroute?
Does someone verify compliance evidence for release?
In most enterprises today, “what happens next” is still human-heavy.
And that is where the new guard is attacking.
The new guard: companies built for the new reality
This is not a “startup vs incumbent” story.
It’s a “new operating model vs old operating model” story.
Here are a few companies shaping the AI-native era in supply chain and logistics, each in their own lane:
Decklar: AI-native Decision Intelligence for Supply Chain
At Decklar, our philosophy is straightforward.
Visibility is only valuable when it converts into business continuity.
Which means the platform must do more than show data.
It must drive execution:
- detect what matters (not just alerts)
- prioritize exceptions that actually carry risk
- guide the next best action
- orchestrate workflows across the network
- prove closure, not just activity
The supply chain doesn’t need more screens.
It needs a reliable operational brain that turns real-time signals into consistent outcomes.
That’s the shift from “visibility platforms” to decision intelligence platforms.
Lyric: AI-native planning built for the speed of change
Traditional planning tools are powerful, but they often operate in cycles.
Lyric represents a modern shift: planning that behaves more like a living system.
Always adjusting, always learning, always reflecting reality.
This matters because in 2026, planning is no longer a calendar event.
It’s a continuous response loop.
HappyRobot: AI-native operational coordination
A huge portion of supply chain execution still happens through:
- status calls
- follow-ups
- check calls
- appointment confirmations
- exception escalations
In other words: human coordination.
HappyRobot is part of the new wave building AI-native automation around communication and execution so operations don’t need to scale headcount linearly just to keep freight moving.
Project44 and FourKites: modern visibility with network-scale intelligence
Overhaul: AI-native shipment risk, security, and recovery workflows
High-value and high-risk shipments don’t fail politely.
They fail with theft, tampering, loss, and claims.
Overhaul modernized the playbook around prevention, escalation, and recovery—workflows that have direct operational and financial impact.
Flexport: software-driven logistics operations
Flexport helped show that logistics isn’t just an industry—it’s a product surface.
The “operating system” mentality is what matters here.
It made it obvious that execution is where advantage lives.
These companies aren’t all doing the same thing.
But they share one trait:
They are building for the world where execution must be faster, more autonomous, and less dependent on humans stitching systems together.
What it really means to compete with the old guard
Here’s the truth I’ve learned after years in supply chain and enterprise deployments:
You don’t beat incumbents by saying, “We’re better.”
You beat incumbents by being safer to adopt.
Because legacy platforms win on something much deeper than product:
- procurement trust
- security comfort
- integration gravity
- referenceability
- “career safety” for decision makers
So, competing with the old guard requires a different kind of maturity.
The new guard must win on:
1) Reliability as a feature
If the platform is business-critical, uptime is not technical.
It’s operational survival.
2) Speed to measurable value
No enterprise wants to wait a year to prove ROI.
3) Integration without disruption
The future stack is not “rip and replace.”
It’s “wrap and upgrade.”
4) Outcomes that show up in operations, not slides
Fewer escalations.
Fewer manual hours.
Fewer missed cutoffs.
Better service levels.
Better compliance.
That’s how credibility is earned.
The hidden message in SAP’s repricing: the world is moving from “software value” to “execution value”
If AI compresses the cost of development, then “software itself” becomes less scarce.
Which means differentiation moves upward.
The winners won’t be the companies with the most modules.
They’ll be the ones with:
- the cleanest data truth layer
- the strongest closed-loop execution
- the highest reliability
- the fastest time-to-value
- the most defensible outcomes
That is why markets are uneasy.
And it’s why some enterprise software valuations are being challenged.
Not because those companies aren’t good.
But because the definition of “good” is changing.
The future enterprise stack: systems of record + systems of action
SAP, Oracle, Blue Yonder, Manhattan—these platforms remain foundational.
They represent the system of record.
But the future belongs to the layer above them:
Systems of action.
AI-native platforms that:
- interpret signals in real time
- decide what matters
- automate the response
- drive resolution
- learn continuously
Thus, platforms like Decklar are not “more software.”
But they are supply chain decision intelligence.
Because for the modern supply chain, the real product isn’t visibility.
The real product is certainty.
Closing thought: AI won’t kill enterprise software. It will expose who actually delivers outcomes.
AI is forcing a new level of honesty in enterprise technology.
It’s no longer enough to say:
“We support your workflows.”
The new bar is:
“We reduce your workload, prevent failures, and make execution reliable.”
That’s what the new guard is building.
And that’s why the market is repricing the old assumptions — including the ones behind companies as strong as SAP.
In supply chain, the winners will not be defined by who has the biggest platform.
They will be defined by who keeps the business running when everything goes sideways.
That is the era we are entering.
