Mountaineers know a harsh truth. At a certain altitude the body begins to rebel. The pulse quickens. Decisions slow down. Strength drains faster. Anyone who continues moving at the same pace as on the plain loses the advantage. Anyone who accepts the new reality and changes the way they breathe goes farther.
Exactly the same thing happened to the software-as-a-service (SaaS) sector. Agentic artificial intelligence raised the altitude sharply and without warning. Old habits - selling seats, adding a “smart” copilot, and relying on predictable growth - suddenly proved insufficient. The market imposed acclimatization. Some companies completed it. Others are still struggling with the thinner air.
This is the story of that forced adaptation of SaaS companies to agentic artificial intelligence: how it began, how the market punished it, what it looks like today, and what it means for investors.
How the Acclimatization of SaaS to AI Began
A year ago the major names in software-as-a-service approached artificial intelligence cautiously. Salesforce added Einstein. Adobe launched Firefly. ServiceNow introduced Now Assist. Workday released Illuminate. The logic was convenient: give the customer a smarter tool inside the familiar product and keep the premium. Stocks still carried a growth premium.
Then the market changed its mind sharply. The fear that agentic artificial intelligence would destroy the seat-based licensing model became the dominant thesis. Salesforce lost roughly 40% from its highs. Adobe fell between 35% and 49%. ServiceNow at certain points dropped more than 50-60%. The median enterprise-value-to-annual-recurring-revenue (EV/ARR) multiple in the sector collapsed to 3.6x. The market was no longer paying for a promise. It was pricing the risk.
How SaaS Companies Adapted to Agentic Artificial Intelligence
Today the picture is clearer. Software-as-a-service companies did not die. They were forced to retool.
The leaders stopped talking about assistants and started building agents - systems that do not merely suggest but execute entire workflows. Salesforce Agentforce already generates more than $1.2 billion in annual recurring revenue with growth exceeding 200%. The company now measures success in agentic work units. ServiceNow is building AI Control Tower and AI Agent Fabric. Adobe is embedding creative agents directly into Photoshop, Premiere, and Illustrator. Workday moved to Flex Credits and domain-specific agents.
The real competitive advantage is no longer the model itself. It is the data. History of customer interactions, ticket logs, creative assets, records of people and money. Pure model laboratories do not have this layer. Platforms that succeed in turning it into a working agentic layer retain control.
How Users Work with AI Agents of SaaS Companies Today
What does the user’s real work look like in 2026? The person describes the desired outcome. The agent plans, retrieves data, executes actions, and returns a finished result.
With Salesforce Help Agent a large share of requests are completed without human intervention. When escalation is required, the agent hands over the full context. With Adobe a creator says “prepare these assets according to the brand and the campaign” and receives finished files. With ServiceNow IT tickets are often diagnosed and resolved autonomously before they ever reach a specialist.
Pricing of AI Agents in SaaS in 2026
Pricing follows the same logic. Pure licensing is retreating. Hybrid and outcome-based models are advancing.
Table 1: Comparison of AI Agent Pricing Models at Leading SaaS Companies (2026)
| Company | Main Pricing Models | What the Customer Pays | Strongest Use Case |
|---|---|---|---|
| Salesforce Agentforce | Conversation / Flex Credits / per user / per resolution | $2 per conversation or from $125 per user | Sales and customer service |
| ServiceNow | Tiers + consumption | Significant uplift on top of the base license | IT, HR, and enterprise processes |
| Adobe | Credits + Firefly packages | Embedded in Creative Cloud + additional credits | Creative and marketing teams |
| Workday | Flex Credits | Based on actual agent usage | HR, finance, and talent management |
Financial Results and Market Multiples of Leading SaaS Companies
The results show who adapted faster to agentic artificial intelligence.
Table 2: Key Financial Metrics and AI Revenue of Salesforce, Adobe, and ServiceNow (2026)
| Company | Revenue Growth (YoY) | AI Revenue and Dynamics | Margins and Cash Flow |
|---|---|---|---|
| Salesforce | ~13% | Agentforce $1.2 bn (+200%+) | Operating margin 34.8% (record) |
| Adobe | 11-13% | AI-first >$500 m (3x), Firefly ~$300 m | Improving margins and strong earnings growth |
| ServiceNow | 20-22% | Strong growth in Now Assist and the AI layer | High margins and strong remaining performance obligations |
Multiples, however, remain compressed. Salesforce trades at around 5x annual recurring revenue. Adobe is near 11x earnings - historically low levels. ServiceNow holds a higher premium. The median EV/ARR multiple in the sector stands at 3.6x.
What the Acclimatization of SaaS Means for the Entire Technology Sector
For the technology sector the split is already clear.
Artificial-intelligence models (OpenAI, Anthropic, and others) are becoming commodity infrastructure. They earn from inference volume. Software-as-a-service companies that turn themselves into the orchestration layer and the system of record for agents retain power. Horizontal platforms that remain mere features lose pricing power. Vertical AI agents are already taking territory in specific domains.
This is not the death of the SaaS model. It is a forced change of the rules - faster and more merciless than the shift to the cloud.
For the investor, conviction must be clearly tiered:
- High conviction - companies that own the data, successfully monetize the agentic layer, and improve margins.
- Medium conviction - companies still moving through the adaptation.
- Low conviction - companies that continue to sell primarily seats.
Conclusion: The Acclimatization Is Complete
The acclimatization is complete. The air is thinner and will remain so.
The companies that changed the way they breathe - that moved from selling access to selling outcomes, turned their data into a real advantage, and built working agents - are already moving with confidence. The rest are still losing strength at altitude.
The question has long ceased to be whether artificial intelligence will destroy software-as-a-service. The question is who has finished the adaptation and who is still carrying the old oxygen tank. In that difference lies both the real investment advantage and the real risk.
Frequently Asked Questions (FAQ) about SaaS and Agentic Artificial Intelligence
1. Will the seat-based model in SaaS disappear completely?
Not in the medium term. A hybrid model is more likely. Base access remains, but an ever-larger share of value and revenue will come from outcomes and consumption. Companies that fail to build the second layer will lose pricing power.
2. Why are SaaS multiples still so low in 2026?
The market continues to price the risk of lower revenue predictability. Until AI revenue becomes a sufficiently large and stable share of total revenue, the premium will return only slowly.
3. What is the biggest risk facing Salesforce, ServiceNow, and Adobe right now?
The speed at which large customers insist on paying only for successfully completed work (outcome-based pricing). If that transition accelerates, margins and revenue predictability will come under pressure.
4. Do vertical AI agents pose a serious threat to the large SaaS platforms?
In narrow domains (legal services, specialized support, certain HR processes) - yes. In the broad enterprise context the large platforms still hold the advantage of data and distribution. The battle is over who controls the orchestration.
5. Where is the most asymmetric investment opportunity in SaaS + AI at the moment?
In the companies that already show rapid growth in AI revenue, improving margins on that revenue, and still-compressed valuations. That is where the gap between market price and actual degree of adaptation is widest.
Important notice
This article is for information and education only. It is not financial advice, an investment recommendation or professional advice of any kind. The views expressed are the author’s personal opinions and are not recommendations to buy, sell or hold securities, assets or investments. Investing involves substantial risk, including the possible loss of all capital. Past performance is not a guarantee of future results. Each reader should conduct independent research, consult a qualified financial adviser and weigh financial circumstances, risk tolerance and investment objectives before taking any decision. The author and borsite.net accept no liability for any loss arising from the use of the information in this article.