The SaaS slowdown: why subscription growth is stalling in 2025-2026 - BFB

The SaaS slowdown: why subscription growth is stalling in 2025-2026

by

For more than a decade, SaaS felt unstoppable. Subscription software became the default business model for startups and enterprises alike. Investors rewarded recurring revenue with premium valuations, CFOs accepted rising software costs as the price of digitisation, and companies enthusiastically added new tools every year.

But something changed between 2023 and 2025: the foundation beneath the SaaS model began to shift. Growth rates slowed. Renewal cycles tightened. CFOs started auditing every tool. And for the first time in years, the number of SaaS applications in use at the average company began to decline.

Below, we explore the key drivers behind the SaaS slowdown, and why these cracks matter heading into the broader structural concerns highlighted in the 2026 reality check across AI, SaaS and crypto.

The CFO-led era of Software consolidation

For most of the 2010s, SaaS spending grew freely inside organisations. Teams purchased tools independently, sometimes in duplicate, often without a central budget owner noticing. That era is over. CFOs have shifted from passive oversight to active optimisation, driven by:

  • Higher interest rates
  • Tighter capital availability
  • Rising cloud infrastructure costs
  • Increased pressure for profitability
  • Investor scrutiny of operational efficiency

In this environment, every recurring contract is questioned:

  • Do we actually use this tool?
  • Can another tool do 80% of the same job?
  • Is this cost justified per employee?
  • Does AI make this tool redundant?

The result is a predictable but material shift: SaaS portfolios are shrinking.

Enterprises are sitting on millions in unused licenses

SaaS bloat has finally caught up with companies. Recent data across large enterprise portfolios shows:

  • Up to 50% of paid seats go unused
  • The average large company wastes $18 million per year on idle SaaS licenses.
  • Duplicate tools often exist without anyone realising.
  • Renewals auto-fire without a proper utilisation

During years of cheap money, this wasn’t a priority. But when budgets tighten, these numbers become impossible to ignore.

CFOs now routinely enforce:

  • License right-sizing
  • Vendor consolidation
  • Usage-based renegotiation
  • Tool ownership accountability

The free-spending era that artificially inflated SaaS growth metrics is ending.

AI is cannibalising the SaaS stack

This is arguably the biggest disruption: General-purpose AI tools are replacing expensive vertical SaaS products. Tasks that were once required:

  • Project management apps
  • Reporting tools
  • Note-taking software
  • Analysis dashboards
  • Content creation suites

…can increasingly be handled by:

  • LLM copilots
  • AI-enabled office suites
  • Automated workflow agents
  • Universal summarisation/analysis tools

Why pay $30–$300 per user per month when a single AI system can handle 60–80% of the workload?

Startups that built their value on UI polish rather than deeply defensible data workflows are the most exposed. Their differentiation evaporates once AI-native alternatives exist. This is forcing a fundamental reckoning: SaaS companies must justify their pricing more rigorously than ever before.

The growth engine is slowing, and investors know it

SaaS IPO volume is at its lowest level in over a decade. Late-stage valuations remain well above what public markets are willing to pay. Expansion revenue (the lifeblood of SaaS) is plateauing. Investors have shifted their expectations from “Land and expand forever” to “Show me real cash flow.”

Even strong SaaS performers are seeing:

  • Lower revenue multiples
  • Slower ARR growth
  • Downward pressure on NDR (net dollar retention)
  • Rising churn in discretionary categories

2025–2026 will not be a generational crash for SaaS, but a reality check on its underlying economics.

What survives the slowdown

Not every SaaS company is at risk. The slowdown is separating essential infrastructure from commodity features. The SaaS products most likely to thrive are those that:

  • Integrate deeply into core workflows
  • Own critical business data
  • Deliver hard, measurable ROI
  • Use AI as a force multiplier, not a competitor
  • Provide compliance, governance, or industry-specific controls AI cannot replace

Companies that cannot demonstrate operational necessity will struggle as the market moves from abundance to scrutiny.

Conclusion: SaaS isn’t dying, but the story is changing

The SaaS slowdown is not a collapse; it is a correction. A long-overdue shift from sprawl to discipline. From novelty to Necessity. From growth-at-all-costs to actual profitability.

As explored in the broader AI, SaaS and crypto 2026 reality check, the entire technology sector is entering a period where hype alone cannot carry valuations. SaaS is simply the first and most visible segment to feel the pressure.

The companies that win in this new cycle will not be the loudest, but the ones that can prove their value.