I came across Gartner’s 2026 Leadership Vision for Tech CEOs this week — recommended reading from Ashley Tott — and it stopped me in my tracks. Not because the individual data points are surprising. But when you read them together, they describe something much larger than an AI product cycle.

They describe a transition that could fundamentally rewrite how enterprise software creates, captures and defends value.

SaaS Was Built Around Human Effort

For the last two decades, most software businesses have effectively monetised human interaction.

Seats. Licences. Users. Clicks. Usage tiers. API calls.

Even the most successful SaaS companies fundamentally depended on people operating the software. The customer still carried the operational burden. Software helped humans perform work more efficiently, but humans remained central to execution.

AI changes that entirely.

What Gartner calls Outcome-as-Agentic-Solution (OaAS) represents a transition away from software as a productivity tool toward software as autonomous execution.

That distinction matters enormously.

Because once AI agents can think, transact, negotiate and execute independently, the value proposition shifts from:

“We help your employees do work.”

to:

“We complete the work itself.”

That is not a feature upgrade.

That is a complete rewiring of software economics.

Value-as-a-Service

The real destination may not even be SaaS or OAAS.

It may ultimately become Value-as-a-Service.

Businesses will increasingly pay for measurable commercial outcomes rather than access to software platforms.

Not CRM licences.
Not workflow tools.
Not reporting dashboards.

But actual delivered outcomes:

Qualified sales meetings.
Customer retention improvements.
Reduced warranty claims.
Faster underwriting decisions.
Working capital optimisation.
Revenue recovered.
Margins improved.

The pricing logic changes completely.

And the winners will not necessarily be the companies with the best interface — or even the best model.

We'll recognise the companies most deeply embedded in operational workflows with the clearest connection to commercial value creation.

That is a much harder moat to replicate.

Two Very Different Businesses

One of the most striking observations in the Gartner report is the emergence of AI-native economics.

The top 30 AI-native companies are commanding valuation multiples of 40x, compared with 6–8x for non-AI-native peers. Autonomous workflow-native providers are achieving $1.57m ARR per full-time employee — against a software industry median of $107k.

Cursor. ElevenLabs. Mercor.

These are not abstract projections.

These businesses exist now.

This creates a brutal strategic divide that most investors continue to underprice.

There are two categories of AI business emerging, and they are not equally attractive.

The first is the velocity-first model.

Fast growth.
Aggressive customer acquisition.
Thin margins.
Heavy subsidy dependence.
Weak switching costs.

Gross margins are hovering around 25%, and many are negative.

These businesses scale quickly but remain perpetually exposed to commoditisation. The moment a foundation model improves, or a platform adopts its features, the moat disappears.

The second is the economics-first model.

Workflow depth.
Operational integration.
Durable retention.
Embedded decision-making.
Outcome-linked pricing.

Gross margins closer to 60%, SaaS-adjacent.

Slower initial growth, but compounding defensibility.

Gartner’s data suggests this distinction will define which AI companies survive the decade.

By 2028, software vendors that do not achieve $1m in ARR per employee will face significant valuation declines.

By 2030, 60% of growth-stage tech companies that fail to invest heavily in both product and operational AI will be displaced by AI-native competitors.

That is an extraordinary mortality rate.

And it maps almost exactly to the velocity-first cohort.

The Hidden Risk Most Are Missing

AI features are being commoditised at an extraordinary speed.

What looks differentiated today can become table stakes within months.

Large platforms are absorbing functionality rapidly.
Foundation model capability continues to improve.
Infrastructure costs are falling.
Open-source models are accelerating diffusion.

The implication is stark:

Many AI companies are not building businesses.

They are building temporary features.

The Gartner position is unambiguous — AI-accelerated commoditisation will kill 90% of startups that bet solely on product differentiation.

For any company between Seed and Series A, the next twelve months need to be dedicated to moving at least one dimension of competitive moat from Temporary to Transitional.

Not revenue growth alone.

Momentum toward defensibility.

The moat increasingly sits outside the model itself.

It sits in:

• proprietary operational data
• embedded workflow ownership
• distribution advantages
• trust layers
• regulatory positioning
• customer integration depth
• network density
• outcome accountability

The Automotive Parallel Is More Urgent Than It Looks

This framework lands differently if you spend time in the automotive and mobility space.

The industry has historically sold products:
cars, finance, servicing, insurance.

Individual transactions.
Point-in-time relationships.
Fragmented data ownership.

But look at what is actually beginning to emerge.

Fleet optimisation platforms that predict maintenance before failure.

Remarketing systems that autonomously price and place vehicles based on real-time demand signals.

Finance providers that underwrite in seconds using behavioural and telematics data.

Customer retention tools that intervene before churn, not after it.

Each of these is moving in the same direction Gartner describes:

From tools to outcomes.
From software licences to commercial performance.

The question for automotive technology businesses right now is whether they are positioning themselves as workflow software providers — or as outcome owners.

Because the market will, over the next three to five years, reward the latter at a dramatically different multiple.

The real long-term value may not sit with the person who sells the vehicle.

It may sit with whoever owns the operational intelligence layer around the vehicle lifecycle.

And in time, that operational intelligence layer may become more valuable than the physical distribution layer itself.

That is a very different company from most of what currently exists in automotive retail technology.

What This Means for Investors

The challenge is that velocity-first AI businesses often look more exciting at an early stage.

Fast ARR growth.
Impressive customer logos.
Bold product narratives.

They attract capital precisely because they scale quickly.

But the economics-first businesses — the ones building genuine workflow depth, proprietary data assets and outcome-linked pricing — often grow more slowly in the early years.

They look less exciting on a standard SaaS metrics dashboard.

The investor question is not:

“How fast is this growing?”

It is:

“What outcome does this company ultimately own — and how defensible is that ownership?”

Gartner puts it clearly: high-growth tech CEOs are 1.67x more likely to have integrated AI into both product and operations simultaneously.

The four priorities — AI-native economics, autonomous business models, next-generation GTM and defensible solutions — do not operate independently.

They compound.

And they collectively point to a future in which the most valuable software companies increasingly resemble autonomous operational partners rather than traditional vendors.

For investors who understand that distinction early, the opportunity is significant.

For those still pricing AI businesses purely on ARR growth multiples, the risk is that they are funding the wrong half of the market.

The last twenty years of software rewarded access.

The next twenty may reward ownership of outcomes.

That is a very different market.

Have a great week!