Cycle on May 26, 2026

I will start by listing the open threads of interest for this research topic and reading any existing notes to understand the prior findings and context. This will help establish what has already been uncovered and what specific questions are currently prioritized.

I will now read any existing notes on the topic to see what has already been documented and avoid duplicate work. This will give me a foundation of prior findings.

The open thread is: - **Title**: *The Shift from Seat-Based to Outcome-Based AI SaaS Pricing Models* - **Description**: *As agentic AI capabilities mature, how are enterprise software pricing models shifting from seat-based or user-based metrics toward outcome-based, workflow-based, or token-based consumption models? Track how procurement teams evaluate and negotiate these non-traditional pricing structures.* Let's do some search queries to see what's happening in B2B software pricing models in 2026, specifically focusing on the shift from seat-based to outcome-based or agent-based pricing. Let's look up how procurement teams evaluate outcome-based, workflow-based, or consumption-based AI pricing models, and how vendors like Salesforce, ServiceNow, HubSpot, or others are structuring these contracts, as well as how buyers are responding. I will formulate some web search queries. I'll search for things like: - `outcome-based pricing AI SaaS 2026` - `agentic AI pricing procurement 2026` - `enterprise software pricing seat-based outcome-based 2026` - `negotiating outcome-based AI contracts procurement` - `consumption based pricing AI procurement risk` Let's execute the first search query.

The search results look incredibly rich and directly aligned with our target thread: "The Shift from Seat-Based to Outcome-Based AI SaaS Pricing Models". Let's see: 1. `https://www.getmonetizely.com/blogs/the-2026-guide-to-saas-ai-and-agentic-pricing-models` - "The 2026 Guide to SaaS, AI, and Agentic Pricing Models" 2. `https://www.chargebee.com/pricing-labs/ai-saas-pricing-outcome-value-models/` - "Outcome-Based Pricing in the AI Era: How SaaS Companies Prove..." 3. `https://pricingsaas.com/reports/pricingsaas-trends-report-2026-q1/` - "PricingSaaS Trends Report, Q1 2026" (mentions shift to hybrid seat+credit models in early 2026) 4. `https://www.linkedin.com/pulse/outcome-based-pricing-next-evolution-saas-revenue-genai-era-r96jf` 5. `https://flexprice.io/blog/why-ai-companies-have-adopted-usage-based-pricing` 6. `https://www.withvayu.com/blog/outcome-based-pricing-strategies` 7. `https://www.accenture.com/us-en/blogs/software-platforms/ai-rewriting-saas-value-equation` - "Beyond the SaaSpocalypse: Proving SaaS Value in the AI Era" (mentions: "Seat-based pricing breaks when agents do the work. Leading companies are shifting toward hybrid...") Let's do a few more searches to see how enterprise procurement reacts to these models and what specific negotiations look like. Let's search for: - `negotiating outcome based pricing procurement 2026` - `procurement challenge usage based AI pricing` - `"outcome-based pricing" procurement contract SaaS` - `how procurement evaluates outcome-based software` Let's run a search.

Let's look at some of these websites to understand the details. I will fetch: 1. `https://www.getmonetizely.com/blogs/the-2026-guide-to-saas-ai-and-agentic-pricing-models` 2. `https://www.chargebee.com/pricing-labs/ai-saas-pricing-outcome-value-models/` 3. `https://pricingsaas.com/reports/pricingsaas-trends-report-2026-q1/` 4. `https://www.npifinancial.com/knowledge-center/ai-spend-is-outrunning-procurement-heres-what-sourcing-leaders-are-actually-doing-about-it/` Let's fetch the first one.

Ah, the fetch failed due to a timeout. Let's try fetching another or re-fetching. Let's try `https://pricingsaas.com/reports/pricingsaas-trends-report-2026-q1/` or `https://www.chargebee.com/pricing-labs/ai-saas-pricing-outcome-value-models/`. Let's fetch `https://www.chargebee.com/pricing-labs/ai-saas-pricing-outcome-value-models/` first.

It seems the firecrawl scraping tool might be experiencing some performance or timeout issues, or the service is slow. Let's try to fetch another URL or try searching for more specific terms to see if we can get cached pages or try a different domain. Let's look up `https://www.withvayu.com/blog/outcome-based-pricing-strategies` or `https://www.npifinancial.com/knowledge-center/ai-spend-is-outrunning-procurement-heres-what-sourcing-leaders-are-actually-doing-about-it/`. Let's try fetching the NPI Financial link.

This is an amazing, highly relevant source! Let's analyze the findings from this NPI Financial article published in May 2026: - The survey was of 89 IT sourcing leaders from large enterprises. - **Biggest AI Spend Challenge (56%)**: Unpredictable usage and scaling costs. - **The Shift**: Per-seat pricing to consumption- and token-based models. - **Tokenomics and the SaaSpocalypse**: "The move from per-seat pricing, which we all knew how to negotiate, to consumption- and token-based models, which most IT procurement playbooks were not built for." - **Procurement Tactics / Best Practices**: 1. **Pilot-first, with shorter terms**: "The pattern emerging is one-year renewals on AI tooling — sometimes shorter — specifically because the underlying technology and pricing are moving too fast to commit to three-year deals. Procurement is trading discount depth for optionality..." 2. **Super caps and price-change clauses**: "Several organizations are negotiating contractual ceilings on consumption-based pricing and explicit clauses governing how vendors can redefine credits or change token multipliers mid-term." 3. **Treating tokens/credits as an internal budget**: "Allocate them to business units. Make people request more when they run out." 4. **Forward deployed engineers (FDE) as a hidden line item**: The professional services attached to AI deployments are scarce, expensive, and often not modeled in TCO. 5. **Data, training, and quiet legal exposure**: Making sure contract language reflects policies around proprietary data in training sets. 6. **ROI measurement remains unsolved**: "ROI on AI right now is a portfolio conversation, not a per-tool conversation..." Let's try to fetch other resources to see how vendors are responding and what other pricing models (e.g., outcome-based pricing) look like. Let's try fetching `https://pricingsaas.com/reports/pricingsaas-trends-report-2026-q1/` again or search for other reports. Let's run a search for: - `site:pricingsaas.com 2026` - `site:linkedin.com/pulse "outcome-based pricing" 2026` - `"outcome-based pricing" SaaS 2026` - `HubSpot outcome based pricing 2026` - `ServiceNow outcome based pricing 2026` Let's execute a search.

The `PricingSaaS Trends Report, Q1 2026` and `PricingSaaS Trends Report, Q2 2026` look very promising. Let's try to fetch both. First, let's try fetching `https://pricingsaas.com/reports/pricingsaas-trends-report-2026-q1/`. Let's see if we can get it.

Ah, the Q1 2026 page is mostly a landing page with sponsor information and an overview. Let's try fetching the Q2 2026 trends report, if it has a similar structure, or let's search for other pages on PricingSaaS that might contain actual details about the trends, or search for other articles about SaaS pricing trends in 2026. Wait, let's search for: - `site:pricingsaas.com "hybrid seat"` - `site:pricingsaas.com "outcome-based"` - `site:pricingsaas.com "monetize"` - `site:pricingsaas.com "HubSpot" 2026` - `site:pricingsaas.com "Figma" 2026` Let's run a search.

Let's fetch: 1. `https://pricingsaas.com/reports/ai-credit-based-pricing/` 2. `https://newsletter.pricingsaas.com/p/the-plg-vs-slg-debate-is-over` 3. `https://newsletter.pricingsaas.com/` Let's fetch `https://pricingsaas.com/reports/ai-credit-based-pricing/` first.

This page is also a directory page, but it contains a very interesting description under "PricingSaaS Trends Report, Q2 2026": *"Shift to AI-driven pricing dominated first three months of 2026: SaaS is moving from seats to hybrid seat+credit models, with 1 in 5 companies making AI pricing changes. Vendors are cutting free AI credits, pricing agents as SKUs, and tying monetization to usage and outcomes."* Oh, look at that! - "SaaS is moving from seats to hybrid seat+credit models, with 1 in 5 companies making AI pricing changes." - "Vendors are cutting free AI credits, pricing agents as SKUs, and tying monetization to usage and outcomes." - There is also a report: "MCP Packaging Strategies: How 99 SaaS vendors package and price Model Context Protocol — as a free feature, a paid-tier feature, or a dedicated add-on — with breakdowns by tier and category." (MCP is Model Context Protocol, which is very hot in early 2026!). Let's fetch `https://newsletter.pricingsaas.com/p/the-plg-vs-slg-debate-is-over`. Let's see what's in there.

Wow, this is an incredibly rich source! Let's break down the key findings from Rob Litterst's March 24, 2026 post, "The PLG vs. SLG debate is over": - **Hybrid-led growth**: The collapse of the line between PLG (Product-Led Growth) and SLG (Sales-Led Growth). - **Freemium is bifurcating**: Some companies rolling back free plans (Deputy, Plaid), others doubling down but monetizing via add-ons (TravelPerk). - **Trials are getting shorter**: Reducing trial length from 30 days to 14 days, and AI-native tools trending to 7 days. This is because AI onboarding allows users to hit value in minutes. Voiceflow cut its trial from 14 to 7 days while increasing AI token allocations by 150% (shorter window, richer product, faster conversion). - **Credits are the new expansion currency**: Credit-based pricing adoption grew **126% year-over-year** across the index. - Monday.com added 500 AI credits to every plan (Basic through Enterprise). - HubSpot shifted from per-conversation credits to a flat 500-credit pool. - Figma moved from AI as a standalone add-on to free-tier credits. - Credits serve as self-serve consumption with built-in sales triggers. - **AI Monetization Lifecycle**: Consistent three-stage lifecycle: 1. Standalone add-on. 2. Embedded in premium tiers. 3. Core feature with usage limits. - **Expansion playbook rewritten**: Old model (user hits seat limit -> sales rep calls -> tier upgrade) is fracturing into: - Add-on layering (e.g., Salesforce Agentforce at $125–$550/user/month). - Credit-based expansion. - Capacity-for-price swaps (e.g., Shorthand raised prices 119% while making members unlimited). - **Quote from Mark Walker (CEO of Nue.io)**: *"Very few companies reached massive scale on pure PLG. Eventually, either customers’ CFOs want control, or you start dealing with enterprise tiers that require security reviews and formal contracting. AI is accelerating this."* Let's do a search on how enterprise buyers are managing these credits. The NPI Financial article mentioned that companies are treating tokens and credits as an **internal budget** allocated to business units. If a business unit runs out, they have to request more, which helps prevent invoice shock. Let's find more details about outcome-based pricing specifically. When agents do the work, seat-based pricing breaks. How do companies price "agents"? Let's search for: - `"Agentforce" pricing model 2026` - `Salesforce Agentforce pricing per conversation` - `HubSpot "Breeze" pricing model 2026` - `ServiceNow agent pricing 2026` - `Intercom agent pricing "Fin" 2026` Let's execute this search.

Wow, this is an incredible, fast-moving development in early 2026! Let's look at the search results: 1. `https://ekfrazo.com/resources/blogs/resources-blogs-salesforce-agentforce-pricing-2026/` - "Salesforce Agentforce Pricing 2026: Tiers, Flex Credits & Real Costs" 2. `https://aquivalabs.com/blog/agentforce-pricing-gets-a-long-overdue-fix-flex-credits-are-now-live/` - "Salesforce Agentforce Pricing Update (Q3 2025) - Aquiva Labs: List price for Agentforce 1 Editions is $550 per user/month. Agentforce editions include everything in the add-ons, bundled cloud-specific..." 3. `https://martech.org/salesforce-introduces-new-agentforce-pricing-models/` - "Salesforce Agentforce pricing models continue to evolve" 4. `https://www.linkedin.com/posts/martykihn_hyperscalers-agentic-ai-activity-7330564610229534720-jIcM` - "Salesforce changes Agentforce pricing to $0.10 per action invoked" Let's fetch `https://ekfrazo.com/resources/blogs/resources-blogs-salesforce-agentforce-pricing-2026/` or `https://aquivalabs.com/blog/agentforce-pricing-gets-a-long-overdue-fix-flex-credits-are-now-live/`. Let's try to fetch `https://aquivalabs.com/blog/agentforce-pricing-gets-a-long-overdue-fix-flex-credits-are-now-live/`. Let's see if we can read it.

This is absolute gold. This details the exact evolution of pricing for Salesforce Agentforce, which is the most prominent agentic platform in the enterprise. Let's summarize the key points here: 1. **The Evolution of Agentforce Pricing**: - Initially, Salesforce launched Agentforce at **$2 per conversation** (a session-based or conversation-based model). - However, this faced backlash for being pricey and unpredictable, particularly for smaller/mid-sized businesses who couldn't forecast their spend. - In response (mid-2025/2026), Salesforce introduced **Flex Credits**, which is a *pay-per-action* consumption model. - Under Flex Credits, businesses pay a minimum of **$500 per 100,000 credits** (~$0.005 per credit). - A standard action consumes 20 credits, which equates to **$0.10 per action invoked**. - An "action" is defined as up to 10,000 tokens (roughly 8,000 words or 22 pages). Pure dialogue without execution does not incur costs—only when the agent actually performs work (updating a record, summarizing a case, answering an inquiry, executing a prompt/flow). - This works alongside **Agentforce Add-ons ($125 per user/month)** for unlimited internal usage by licensed employees, and **Agentforce 1 Editions (starting at $550 per user/month)** which bundle unlimited internal usage, 1M Flex Credits, and 2.5M Data Cloud credits. 2. **The "Flex Agreement" (License-to-Compute Swaps)**: - This is a *major* innovation for procurement. The **Flex Agreement** allows organizations to manage both human and digital labor. - Procurement can **convert human user licenses into Flex Credits and vice versa**. - This directly addresses the "seat-based pricing breaks when agents do the work" problem. If a company deploys an AI agent that replaces or reduces the need for human seats, they don't lose that budget or get locked into unused seats; they can swap those human seat licenses for Flex Credits (compute) to run the agents! - This is a massive shift in how contracts are structured. 3. **Buying Models**: - **Pre-purchase**: Buy upfront for the entire term (locks in set usage at a fixed price, best discount). - **Pay-as-you-go**: No upfront commitment, pay monthly in arrears. - **Pre-commit**: Commit to a baseline level without upfront payment, combining predictability with usage billing. Let's do some research on other major players. What is HubSpot doing? What is ServiceNow doing? What is Intercom doing? Let's do a search for: - `HubSpot "Breeze" pricing model 2026` - `ServiceNow "Action Fabric" pricing model` - `Intercom Fin pricing per resolution 2026` - `outcome-based pricing ServiceNow 2026` Let's search.

Oh, this is incredible and highly specific! Let's look at the dates: **April 14, 2026** is the date HubSpot transitioned its AI pricing! This is literally a brand-new, current trend happening right now in 2026! Let's look at the details: - **HubSpot Breeze AI Agents**: - Transitioned from a "per-use" model ($1.00 per conversation) to a **"pay-per-result" or "pay-per-resolution" outcome-based model** starting **April 14, 2026**. - **AI Customer Agent**: Pricing moves from $1.00 per conversation to **$0.50 per resolved conversation**. - **AI Prospecting Agent**: Moves to **$1.00 per lead recommended** for outreach (or $1 per lead recommended). - This is a direct shift to outcome-based pricing! Let's fetch: 1. `https://www.cmswire.com/customer-experience/hubspot-shifts-breeze-ai-agents-to-pay-per-result-pricing/` 2. `https://www.saastr.com/hubspot-switching-ai-pricing-from-per-use-to-per-resolution-but-does-it-really-matter/` 3. `https://theaieconomy.substack.com/p/hubspot-breeze-outcome-based-pricing-model` Let's fetch the first one.

This is an incredible, highly specific and timely source! Let's see: - On **April 2, 2026**, HubSpot announced its transition of Breeze AI agents to **outcome-based pricing**, effective **April 14, 2026**. - **Breeze Customer Agent**: Pricing drops from $1.00 per conversation to **$0.50 per resolved conversation**. - **Breeze Prospecting Agent**: Moves from a recurring monthly charge per enrolled contact to **$1 per lead recommended for outreach**. - Jon Dick, Chief Customer Officer at HubSpot, states: *"Outcome-based pricing removes that risk. You pay when it works, full stop. Customers can move faster, experiment more, and trust that their spend is tied to real results."* - This aligns with the Deloitte AI investment ROI report which notes that proving ROI is a struggle due to fragmented data, unclear success metrics, and experimentation-heavy deployments. - HubSpot's Breeze platform includes 20+ Agents and Assistants, Breeze Studio, and Breeze Marketplace. Let's do a search on other major software vendors transitioning to outcome-based or pay-per-resolution pricing. Let's search: - `Intercom Fin pricing per resolution 2026` - `ServiceNow Action Fabric price` - `Cognition Devin pricing model 2026` - `outcome-based pricing customer service SaaS 2026` Let's execute this.

This is incredibly rich! Let's see: 1. **Intercom Fin AI Agent**: - Pricing is **$0.99 per outcome/resolution** (or deflection). - "Outcome-based pricing, which aligns with the value users receive from Fin. In this model, users pay 99¢ per resolution, meaning they are only charged when a conversation is resolved." - Intercom President Archana Agrawal explained how **Fin hit $100M ARR** using outcome-based pricing, AI agents, and a **$1M performance guarantee**. - "Intercom's CEO bet $1M on their AI agent hitting 65% resolution rates." - "When companies shift from 'per-seat' to 'per-resolution' pricing, they often keep their old onboarding. But seat-based PLG optimizes for 'a user logging in', whereas outcome-based pricing requires redesigning onboarding and activation." Let's fetch `https://gtmnow.com/how-intercom-built-the-highest-performing-ai-agent-on-the-market-using-outcome-based-pricing-with-archana-agrawal-president-at-intercom/` to read the details on how Intercom did this, what procurement challenges they faced, and how they structured their $1M performance guarantee. This is gold for our research.

This is incredibly rich! Let's see: - Intercom's Fin AI Agent: - Billed at **$0.99 per resolution** (outcome-based pricing model). - Scaled from **$1M to $100M+ ARR** using this model. - Backed by up to a **$1M performance guarantee** if resolution targets aren't met. - "Guarantees change buyer psychology more than pricing ever could. The $0.99 price gets attention, but it's the $1M performance guarantee that builds trust." - "Outcome-based pricing is a good forcing function. Charging $0.99 per resolved issue exposed every weak link." - "AI raised the bar for human work... Humans moved from execution to system design: training the AI, handling edge cases, and improving performance." - "Forward-deployed engineers are a learning engine. Intercom avoided heavy customization by design. Every customer interaction fed back into the core product." Let's do some more research on how procurement teams evaluate these contracts. We have: 1. **Salesforce Agentforce**: - Initially $2/conversation. - Now **Flex Credits** ($0.10/action, minimum $500 for 100,000 credits). - **Flex Agreements**: License-to-compute swaps (converting seat licenses to credits and vice versa). This is a massive shift allowing procurement to swap human seats for AI compute. 2. **HubSpot Breeze**: - Switched in **April 2026** to outcome-based pricing. - **Customer Agent**: $0.50 per resolved conversation (down from $1.00 per conversation). - **Prospecting Agent**: $1.00 per lead recommended. 3. **Intercom Fin**: - **$0.99 per resolution** (with no seat costs, no setup fees, and up to a **$1M performance guarantee**). - Has scaled to $100M+ ARR under this model. Let's do a search on other companies. What about ServiceNow? What about SAP or other enterprise giants? Let's search for: - `ServiceNow "Action Fabric" pricing 2026` - `ServiceNow AI agent pricing model 2026` - `SAP AI API ban pricing 2026` - `ServiceNow "Now Assist" pricing model` Let's run a search.

Let's look at the search results: 1. `https://atonementlicensing.com/blog/servicenow-now-assist/` - "ServiceNow Now Assist Pricing 2026: AI Licensing &..." - Description: "ServiceNow is bundling Now Assist AI capabilities into its platform tiers, driving effective price increases of 20–40% at renewal. Enterprises that understand..." 2. `https://www.reddit.com/r/servicenow/comments/1sn8vpc/gave_now_assist_3_months_and_i_am_not_sure_what/` - "Gave Now Assist 3 months and I am not sure what we are paying for" Let's fetch `https://atonementlicensing.com/blog/servicenow-now-assist/` first to see how ServiceNow is packaging its AI and what procurement teams are facing. This is highly relevant to the "AI Tax", "Forced Upgrades", and how procurement evaluates these contracts.

This is an incredibly detailed and valuable source! Let's analyze what's happening with ServiceNow's Now Assist in 2026: - **ServiceNow Now Assist**: - ServiceNow is bundling Now Assist generative AI capabilities into its platform tiers, driving effective price increases of **20–40% at renewal**. - The per-user premium for Now Assist vs. the base platform is reported to be **$150–$250 per user/year** (or £150–£250 depending on region). - **Tier Bundling (The "Pro Plus" and "Enterprise Plus" Trap)**: The most common model is that Now Assist is included in a new "Pro Plus" or "Enterprise Plus" tier at a higher per-user price. Staying on the old "Pro" or "Enterprise" tier means losing AI capabilities in the new product roadmap. - **Per-User Add-on**: Priced at $150–$250/user/year, resulting in a 20–35% cost uplift across the fulfillment user population. - **Credit Consumption**: An emerging model where heavy LLM inference tasks are priced on a credit/consumption basis. - **The cost uplift in practice**: A 1,000-fulfillment-user ServiceNow deployment currently on ITSM Pro at £240/user/year faces a Pro Plus upgrade to £310–£340/user/year—a 30–42% increase, or £70,000–£100,000 annually. For larger deployments (5,000+ users), the annual uplift exceeds £500,000. - **Negotiation Levers**: 1. **Phased Adoption Commitments**: Committing to Now Assist for a subset in Year 1, with agreed pricing for full deployment in Years 2 and 3. This typically gets a 15–25% discount. 2. **Pilot ROI Requirements**: Requiring ServiceNow to demonstrate outcomes (productivity improvements, incident resolution time reduction) through a paid pilot before enterprise deployment. 3. **Module Selectivity**: Negotiating Pro Plus only on modules where productivity ROI is clear (e.g., ITSM), while staying on Pro for others (e.g., HRSD or Field Service). 4. **Contract Protections**: Provisions preventing reclassification of current-tier features to "Plus" tiers, caps on annual escalation, and renewal pricing protection. Let's look back at the open thread: *As agentic AI capabilities mature, how are enterprise software pricing models shifting from seat-based or user-based metrics toward outcome-based, workflow-based, or token-based consumption models? Track how procurement teams evaluate and negotiate these non-traditional pricing structures.* We have an extremely cohesive and robust set of findings that perfectly answer this thread and provide actionable insights for founders selling to enterprises today! Let's double check if there are any other specific angles we can uncover about how procurement is tackling this. In the NPI Financial article, there are several key procurement tactics: - **Pilot-first, shorter terms**: Shorter contracts (1 year or less) to avoid long-term lock-in since the tech and pricing are moving so fast. Optionality is prioritized over discount depth. - **Super caps and price-change clauses**: Contractual ceilings on consumption-based pricing, and explicit clauses governing how vendors can redefine credits or change token multipliers mid-term. - **Tokens/credits managed as internal budgets**: Internal allocation to business units to prevent invoice shock. - **Forward deployed engineers (FDE) as hidden line items**: Professional services attached to AI deployments are scarce, expensive, and must be modeled into TCO. - **Data, training, and legal exposure**: Ensuring contract language prevents proprietary data from entering vendor training sets. - **ROI measurement is unsolved**: It's a portfolio conversation, not a per-tool conversation. And from the PricingSaaS / Good Better Best newsletter and reports: - **Hybrid-Led Growth**: Combining PLG and SLG. - **Credits are the new expansion currency**: Credit adoption up 126% YoY across the index. Monday.com, HubSpot, Figma are all using credits as self-serve consumption with built-in sales triggers. - **AI Monetization Lifecycle**: Standalone add-on -> embedded in premium tiers -> core feature with usage limits. - **Shorter trials**: AI onboarding reduces trials from 30 days to 14 or 7 days (Voiceflow cut from 14 to 7 days while increasing token allocation 150%). And from the Vendor Pricing Models: 1. **Salesforce Agentforce**: - Initially $2 per conversation. - Now **Flex Credits** ($0.10/action, $500 minimum for 100k credits). - **Flex Agreements**: Human license to digital compute swaps! Procurement can swap human user licenses for Flex Credits and vice versa. 2. **HubSpot Breeze**: - Switched on **April 14, 2026** to outcome-based pricing. - AI Customer Agent at $0.50 per resolved conversation (down from $1.00 per conversation). - AI Prospecting Agent at $1.00 per lead recommended. 3. **Intercom Fin**: - Outcome-based pricing at **$0.99 per resolution**. - Backed by a **$1M performance guarantee** to build trust and change buyer psychology. Let's write up some incredibly detailed, high-quality notes that address this thread. Wait, let's see if we can read the existing notes again to see if we should update or write new ones. The list of existing notes: - `salesforce-aela-pricing-lock-in-risk` (discusses Salesforce's Agentic Enterprise License Agreement) - `saas-inflation-budget-volatility-2026` - `review-platforms-ai-citation-substrate` - `platform-consolidation-2026` - `pilot-trap-enterprise-ai` - `it-sourcing-infrastructure-constraints-2026` - `incumbent-pricing-responses-agents-data-tolls` (discusses ServiceNow's Action Fabric and SAP's AI API ban) - `enterprise-buying-journey-stages` - `data-tolls-connector-fees-lockin` - `buyers-use-ai-dont-trust-it` - `buyability-framework-linkedin` - `ai-vendor-lock-in-zapier` - `ai-tax-and-sprawl-2026` - `ai-table-stakes-new-baseline` - `ai-overages-forced-upgrades-negotiation-leverage` - `ai-build-vs-buy-myth-reality` - `agentic-enterprise-license-agreements-aela` - `aeo-axo-frameworks-2026` Wait, we have: - `incumbent-pricing-responses-agents-data-tolls` - `agentic-enterprise-license-agreements-aela` - `ai-overages-forced-upgrades-negotiation-leverage` Let's look at the open thread again: `The Shift from Seat-Based to Outcome-Based AI SaaS Pricing Models` "As agentic AI capabilities mature, how are enterprise software pricing models shifting from seat-based or user-based metrics toward outcome-based, workflow-based, or token-based consumption models? Track how procurement teams evaluate and negotiate these non-traditional pricing structures." This is a specific, dedicated thread. We can write a note with the slug `outcome-based-ai-pricing-procurement` or update an existing one if it matches. Wait, let's see if there is an existing note that covers this. Let's look at the previews: - `agentic-enterprise-license-agreements-aela`: "As enterprise buyers push back on the unpredictability and budget volatility of consumption-based AI pricing (per token, per conversation, or per credit), market leaders are shifting back to..." - `incumbent-pricing-responses-agents-data-tolls`: "As third-party autonomous AI agents threaten to erode the traditional per-seat licensing model (where a single agent can execute workflows that previously required dozens of human seats), enterprise..." - `ai-overages-forced-upgrades-negotiation-leverage`: "As enterprise software vendors look to monetize generative AI and offset slowing growth in core seat licensing, they are employing aggressive packaging and tier-upgrade tactics. For IT procurement..." Let's write a comprehensive new note with the slug `outcome-based-ai-pricing-procurement` that details: 1. The Vendor Transition: - HubSpot's transition on April 14, 2026 to outcome-based pricing for Breeze ($0.50 per resolved conversation, $1 per lead recommended) from flat-rate or per-use ($1.00 per conversation) to align spend with outcomes. - Intercom's Fin AI Agent ($0.99 per resolution) which has scaled from $1M to $100M+ ARR using this model, backed by a $1M performance guarantee to overcome buyer skepticism. - Salesforce's Agentforce transition from $2 per conversation to Flex Credits ($0.10 per action invoked, minimum $500 for 100k credits) and the introduction of "Flex Agreements" to allow license-to-compute swaps. - ServiceNow's Now Assist "Pro Plus" and "Enterprise Plus" bundling, driving 20–40% cost uplifts at renewal, which procurement is negotiating via phased adoption and module selectivity. 2. The Procurement Playbook for Non-Traditional AI Pricing: - Super caps and price-change clauses to prevent invoice shock. - Shorter contract terms (1-year or less) prioritizing optionality over discount depth. - Tokens/credits managed as internal budgets allocated to business units. - Forward deployed engineers (FDE) modeled as a hidden TCO line item. - ROI measured at a portfolio level, not per-tool, or requiring paid pilots with documented ROI. 3. The "Hybrid-Led Growth" Model: - Collapse of the line between PLG and SLG. - Credits as the new expansion currency (126% YoY growth across the index). - Shorter trials (reducing from 30 days to 14 or 7 days) because AI onboardings deliver value in minutes. Let's double-check the rules for writing notes: - No YAML frontmatter — the title is the first `# ` heading and the slug is the slug parameter you pass. - Name and link the actual thing (the story, article, paper, or page that was posted/published). - Carry 1-2 REPRESENTATIVE VERBATIM QUOTES, copied exactly, each kept with the link it came from (an HN comment permalink, the article URL, etc.). - Say what it means. Let's review the exact quotes we have: From NPI Financial: - Link: `https://www.npifinancial.com/knowledge-center/ai-spend-is-outrunning-procurement-heres-what-sourcing-leaders-are-actually-doing-about-it/` - Quote: *"The move from per-seat pricing, which we all knew how to negotiate, to consumption- and token-based models, which most IT procurement playbooks were not built for."* - Quote: *"The pattern emerging is one-year renewals on AI tooling — sometimes shorter — specifically because the underlying technology and pricing are moving too fast to commit to three-year deals. Procurement is trading discount depth for optionality, and on AI right now, that’s the right trade."* From HubSpot (CMS Wire): - Link: `https://www.cmswire.com/customer-experience/hubspot-shifts-breeze-ai-agents-to-pay-per-result-pricing/` - Quote: *"Outcome-based pricing removes that risk. You pay when it works, full stop. Customers can move faster, experiment more, and trust that their spend is tied to real results."* - Quote: *"Breeze Customer Agent pricing drops from $1.00 per conversation to $0.50 per resolved conversation. Breeze Prospecting Agent moves from a recurring monthly charge per enrolled contact to $1 per lead recommended for outreach."* From Intercom (GTMnow): - Link: `https://gtmnow.com/how-intercom-built-the-highest-performing-ai-agent-on-the-market-using-outcome-based-pricing-with-archana-agrawal-president-at-intercom/` - Quote: *"Guarantees change buyer psychology more than pricing ever could. The $0.99 price gets attention, but it’s the $1M performance guarantee that builds trust."* - Quote: *"Outcome-based pricing is a good forcing function. Charging $0.99 per resolved issue exposed every weak link."* From PricingSaaS (Newsletter): - Link: `https://newsletter.pricingsaas.com/p/the-plg-vs-slg-debate-is-over` - Quote: *"Credit-based pricing adoption grew 126% year-over-year across our index. Monday.com added 500 AI credits to every plan simultaneously — Basic through Enterprise. HubSpot shifted from per-conversation credits to a flat 500-credit pool."* - Quote: *"Trials are getting shorter. Trial reductions outnumbered extensions 3:1 across 2025. The median SaaS trial is moving from 30 days toward 14, and AI-native tools are already trending toward 7. Why? AI-powered onboarding means users hit value in minutes, not weeks."* From Salesforce Agentforce (Aquiva Labs): - Link: `https://aquivalabs.com/blog/agentforce-pricing-gets-a-long-overdue-fix-flex-credits-are-now-live/` - Quote: *"The new model, called Flex Credits, offers the same features but it has introduced the credit system this time. So, instead of paying for each action, you’ll pay it as a one-off."* - Quote: *"The new flex agreement allows organizations to manage both human and digital labor. Thus, businesses can convert the user licenses into Flex credits and vice versa. This way, companies can shift the budget between human user licenses and Agentforce AI agents."* From ServiceNow Now Assist (Atonement Licensing): - Link: `https://atonementlicensing.com/blog/servicenow-now-assist/` - Quote: *"ServiceNow is bundling Now Assist AI capabilities into its platform tiers, driving effective price increases of 20–40% at renewal. Enterprises that understand how Now Assist is priced — and when to push back — negotiate substantially better outcomes."* - Quote: *"The most commercially significant dynamic in ServiceNow renewals is the 'tier upgrade conversation.'... The per-user price difference between Pro and Pro Plus is typically 25–40% depending on agreement size and negotiation."* This is absolutely incredible. The depth and specificity of this data is spectacular. It directly answers the open thread with 2026 data. Let's write this note! I will construct the markdown body for the note `outcome-based-ai-pricing-procurement` and then call `write_note`. Let's do that now.

I will now write a comprehensive note detailing the shift from seat-based to outcome-based AI SaaS pricing models, highlighting specific 2026 vendor structures (HubSpot, Salesforce, Intercom, ServiceNow) and how enterprise procurement teams negotiate these non-traditional pricing agreements.

I have written the note detailing the shift from seat-based to outcome-based AI SaaS pricing models. Let's resolve the open thread since the note completely addresses it with very rich, concrete evidence from early 2026.

I have successfully written the note and resolved the corresponding thread. I will now end the research cycle by calling `finish_cycle`.

I have completed the research cycle on B2B software evaluation criteria and the shift from seat-based to outcome-based pricing models in 2026. ### Summary of Key Findings from this Cycle: 1. **The Vendor Shift to Outcome & Credit-Based Pricing (April 2026)**: * **HubSpot Breeze**: Shifted on April 14, 2026, from $1.00 per conversation to **$0.50 per resolved conversation** (AI Customer Agent) and **$1.00 per lead recommended** (AI Prospecting Agent). * **Intercom Fin**: Scaled to $100M+ ARR using a **$0.99 per resolution** model, backed by an up to **$1M performance guarantee** to build buyer trust. * **Salesforce Agentforce**: Introduced **Flex Credits ($0.10 per action invoked)** to address unpredictability, and launched **"Flex Agreements" (seat-to-compute swaps)** so procurement can convert unused human seat licenses into AI compute credits. * **ServiceNow Now Assist**: Bundling AI into premium "Pro Plus" and "Enterprise Plus" tiers, driving **20% to 40% price increases** at renewal ($150–$250 per user/year premium). 2. **The 2026 Procurement Playbook (NPI Financial Survey)**: * **56% of procurement leaders** state that unpredictable usage and scaling costs are their biggest AI spend challenge. * Sourcing teams are shifting to **one-year or shorter contracts** to preserve optionality over discount depth. * Procurement is negotiating **super caps** and mid-term price-change protection clauses to prevent credit multiplier inflation. * Teams are managing credits as **internal budgets allocated to business units** to prevent invoice shock. * Sourcing is modeling **Forward Deployed Engineers (FDE)** as a hidden TCO line item. 3. **The Rise of "Hybrid-Led Growth" (PricingSaaS)**: * Credit-based pricing adoption grew **126% YoY** (e.g., Monday.com, HubSpot, Figma). * **Shorter trials** (median moving from 30 days to 14 or 7 days) because AI onboardings deliver value in minutes. Voiceflow cut its trial to 7 days while increasing AI token allocations by 150% to accelerate conversions. All findings have been persisted under the topic notes, and the prioritized thread has been successfully resolved!