Cycle on May 24, 2026

I will list the active threads for this topic to see what specific questions or directions have been queued up for this cycle.

I will search the web for details on Salesforce's Agentic Enterprise License Agreement (AELA) and how other enterprise incumbents like ServiceNow, Workday, and SAP are structuring their agentic AI pricing, digital labor bundles, and API/connector fees.

This search returned several excellent sources. I will fetch the Forrester blog, the SaaStr article, and The Register article to understand the details of Salesforce's AELA and how it relates to other software incumbents adapting their pricing models. Let's fetch these URLs.

This is a fantastic analysis. The Forrester article describes how the Agentic Enterprise License Agreement (AELA) lets customers get unlimited use of consumption-based products (Agentforce, Data Cloud, MuleSoft) for a fixed fee over 2 or 3 years. It's a flat-rate "digital labor" bundle designed to lock customers in. Let's fetch the SaaStr article and the Register article to dig deeper into the actual mechanics and what other enterprise incumbents are doing.

Let's try fetching the Register article instead, or do a broader search on how other incumbents are pricing. I will fetch the Register article first.

Ah, the URL for The Register was truncated in the search results or had some extra characters. Let's search again to see if we can find the correct URL or other articles. I will search for `"Salesforce willing to lose money on AI"`.

Let's search for "Salesforce willing to lose money on AI agent licenses when customers are locked in" to find the original article. Or let's fetch the forum/thread or the fiduciarytech copy. Let's do a search for the exact title from The Register.

Excellent, we have two very relevant articles from The Register: 1. `https://www.theregister.com/special-features/2025/12/15/salesforce-willing-to-lose-money-on-ai-to-lock-in-customers/2615737` 2. `https://www.theregister.com/special-features/2026/01/27/gartner-questions-if-salesforce-ai-will-stay-all-you-can-eat/4776384` Let's fetch both of them to gather concrete evidence of the pricing mechanics, customer lock-in, and competitor responses.

This article is highly informative. Now let's fetch the second Register article about Gartner's perspective on Salesforce's pricing, as well as checking the linked articles if possible, or doing a search on "ServiceNow" and "Workday" pricing responses. Let's fetch the Gartner one first.

This is very detailed. The Gartner analyst Hannah Decker is pointing out that: 1. Gartner believes these flat-rate, all-you-can-eat AELA deals will be converted into defined quantity contracts when they expire, creating a potential cost shock. 2. Vendors retain the right to change the multiplier or the rate of consumption of credits at any point during the term of the agreement, which means consumption can increase even if usage hasn't changed. 3. Salesforce price rises of 6 to 15% are feeding directly into renewal uplifts, with GenAI and agentic AI SKUs leading to steep cost increases. 4. Salesforce's Bill Patterson (EVP, Corporate Strategy) disputed this, saying renewals remain flexible and mentioning their "Digital Wallet" tool for visibility. Let's look into other enterprise software incumbents' structures. Specifically: - ServiceNow - Workday - SAP - Oracle How are they adapting their pricing and licensing in response to Salesforce's AELA? Are they introducing digital labor bundles or raising API/connector fees ("data tolls")? Let's run a web search.

Let's run a search for Workday's pricing of AI and agents, and SAP's pricing of AI and agents. Let's see if they are doing any flat-rate, consumption, or API-based pricing. Let's search `Workday AI agent licensing pricing` or `"Workday" "pricing" "AI" agent` and `SAP AI "digital labor" OR Joule pricing`.

Oh, wow! Look at these search results: 1. `https://www.pymnts.com/artificial-intelligence-2/2026/servicenow-sap-and-workday-make-ai-agents-pay-to-play/` — "ServiceNow, SAP and Workday Make AI Agents Pay to Play" 2. `https://erp.today/sap-shifts-to-ai-consumption-pricing-as-agents-threaten-saas-revenue-model/` — "SAP Shifts to AI Consumption Pricing as Agents Threaten Saas Revenue Model" 3. `https://redresscompliance.com/workday-illuminate-ai-pricing-guide` — "Workday Illuminate AI Pricing Guide" (standalone agents are 12 to 38 dollars per FSE per month. Per agent. Seven agents available in 2026.) 4. `https://joshbersin.com/2026/03/workday-and-sana-unveil-a-bold-new-strategy-for-ai/` — "Workday and Sana Unveil A Bold New Strategy For AI" 5. `https://www.workday.com/en-us/artificial-intelligence/ai-agents.html` — Workday AI agents available through Workday Flex Credits. Let's fetch the PYMNTS article, the ERP Today article, and the Redress Compliance Workday guide to get exact details on how ServiceNow, SAP, and Workday are pricing their AI agents and whether they are using digital labor bundles, consumption pricing, or credit systems. Let's do that now.

This is incredible! This is a goldmine of information that perfectly answers the thread's questions about other enterprise software incumbents' pricing and licensing structures in response to agentic AI, specifically around: 1. **ServiceNow Action Fabric**: Introduced in May 2026 at Knowledge 2026. A metered integration layer that external AI agents must pass through to access data and execute workflows. They charge on an action-based pricing model (effectively a "tax" or "toll" on third-party AI agents interacting with ServiceNow data). 2. **SAP's API Policy**: Updated in April 2026 to prohibit third-party AI agents from interacting with its systems outside of SAP-endorsed architectures. They bar the use of SAP APIs for autonomous or generative AI systems that "plan, select, or execute sequences of API calls" without SAP's official sanction, pushing users to their own Joule Agents. This has provoked major lock-in concerns and pushback from the DSAG user group. 3. **Workday's Approach**: Workday CEO Aneel Bhusri said charging for agent access offers considerable financial upside. They have also launched Sana AI Agents and Workday Flex Credits (and standalone agents at $12 to $38 per FSE per month). 4. **DataDog**: Capping daily/monthly requests for their Model Context Protocol (MCP) server for third-party agents. Let's fetch the SAP API policy article from The Register or search for it to get the exact quotes. Let's do a search for `"new SAP API policy" "The Register" 2026` or `SAP API policy "lock-in"`.

This is an absolute goldmine! The search results confirm: 1. Gartner published a report on April 24, 2026: "SAP's New API Policy Risks Unbudgeted AI Costs for Customers" (increases lock-in, creates unbudgeted costs). 2. The Register published "AI clause in new SAP API policy provokes lock-in concern" on April 29, 2026. 3. The Register published "SAP user group slams 'uncertainty' in ERP giant's API policy" on April 30, 2026. 4. Fivetran published "SAP's latest API policy raises the stakes for your AI strategy" around late April/early May 2026. 5. Kai Waehner wrote a blog "Data Ownership in the Age of Agentic AI: Why SAP's API Policy Forces a Data Integration Reckoning for Every Enterprise" on May 2, 2026. Let's fetch the Register articles to understand the exact wording and quotes about SAP's API policy. Let's do `https://www.theregister.com/software/2026/04/29/ai-clause-in-new-sap-api-policy-provokes-lock-in-concern/5225767` and `https://www.theregister.com/software/2026/04/30/sap-user-group-slams-uncertainty-in-erp-giants-api-policy/5220996`. Let's fetch them now.

Let's fetch the second Register article about DSAG (the German-speaking SAP User Group) slamming the "uncertainty" in SAP's API policy. This will give us excellent context. Let's do that now.

This is incredibly detailed and highly relevant to the thread `019e53e0-467d-7295-bd52-cbb63f1689db` ("Incumbent Pricing Responses: Tracking AELAs and Data Tolls Across SAP, ServiceNow, and Workday"). Let's look at what we've discovered: 1. **Salesforce AELA (Agentic Enterprise License Agreement)**: Introduced in late 2025 (October/December). It represents a flat-rate, seat-based, "unlimited-use" arrangement for AI agents (Agentforce, Data Cloud, MuleSoft) over a 2 or 3-year term. - Chief Revenue Officer Miguel Milano admitted that Salesforce is willing to "lose money" on some of these deals in the short term because it locks customers in, giving Salesforce "another 20 years to monetize" them. - Gartner (via analyst Hannah Decker in Jan 2026) warned that these flat-rate, all-you-can-eat deals will likely be converted into "defined quantity contracts" at renewal, creating a potential cost shock. - Gartner also warned that vendors retain the right to change the multiplier or consumption rate of credits under these agreements, which adds risk. 2. **ServiceNow Action Fabric**: Announced at Knowledge 2026 (May 2026). - A metered integration layer that external AI agents must pass through to access data and execute workflows inside the ServiceNow platform. - Charged on an action-based pricing model. JPMorgan analyst Mark Murphy called this effectively a "tax" or "tollgate" on customers using outside AI agents to interact with data they already store in ServiceNow. - Launch partner: Anthropic's Claude (with Claude Cowork). 3. **SAP's New API Policy**: Updated in April 2026 (v4/2026). - Prohibits the use of SAP APIs to integrate with (semi-)autonomous or generative AI systems that "plan, select, or execute sequences of API calls" except through SAP-endorsed architectures. - Pushes customers to use SAP's own Joule Agents. - DSAG (German-speaking SAP User Group) slammed the policy for creating "huge uncertainty" and potentially blocking third-party AI innovations. Gartner warned it "Risks Unbudgeted AI Costs for Customers" and increases lock-in. 4. **Workday Illuminate & Sana AI Agents**: - Workday launched Sana as its AI studio/infrastructure in March 2026. - Standalone agents are priced at $12 to $38 per FSE (Full-Time Equivalent) per month, per agent (7 agents available in 2026). - Available via "Workday Flex Credits" for flexible consumption. - CEO Aneel Bhusri has noted that charging for agent access offers considerable financial upside. Let's do a search on how enterprise buyers are reacting to these changes, or what other general procurement trends exist in 2026. Let's look up `enterprise procurement B2B software AI table stakes 2026` to see what else we should update or write about. Let's do a search.

Let's check the LinkedIn post "2026 Tech Sourcing Predictions: AI Pricing Shifts Vendor Leverage" and the Docket.io blog "The 2026 AI Procurement Checklist for B2B SaaS". These seem extremely relevant to our topic of how enterprise buyers are changing their evaluation criteria for B2B software as AI becomes table stakes. Let's do a quick search to get more details on those.

Let's fetch the LinkedIn post by Per Melker on "2026 Tech Sourcing Predictions: AI Pricing Shifts Vendor Leverage" and the URL `https://www.linkedin.com/posts/per-melker-a67386_2026-tech-sourcing-predictions-by-jon-winsett-activity-7413936272592756737-nzIW` to see what Jon Winsett (founder of NPI, a prominent tech procurement advisory firm) wrote. Let's do that.

Ah, LinkedIn is blocked. Let's do a web search for `"2026 Tech Sourcing Predictions" "Jon Winsett"` or `Jon Winsett NPI 2026` to find the content of his predictions on other websites.

Let's fetch the AI TechPark interview with Jon Winsett: `https://ai-techpark.com/aitech-interview-with-jon-winsett/` to understand NPI's perspective on IT procurement and AI pricing in 2026. Let's do that now.

The URL is forbidden. Let's try to search for the NPI blog post "How to Beat ServiceNow at the Negotiation Table in 2026" or other NPI articles on ServiceNow, SAP, or Salesforce pricing. Let's do a search for `"How to Beat ServiceNow at the Negotiation Table in 2026"`.

Excellent. Let's fetch the NPI article "How to Beat ServiceNow at the Negotiation Table in 2026": `https://www.npifinancial.com/knowledge-center/how-to-beat-servicenow-at-the-negotiation-table-in-2026/` to see what they say about ServiceNow's pricing tactics, forced AI migrations, etc. Let's do that.

This is incredibly rich! Let's analyze what we have. We have two incredibly detailed and distinct storylines that perfectly answer the topic brief and the specific thread: ### Finding 1: Incumbent Pricing and Access "Tolls" in the Agentic Era: ServiceNow's Action Fabric and SAP's AI API Ban * **Slug**: `incumbent-pricing-responses-agents-data-tolls` (or we can reuse the thread's focus but create a fresh slug as it's a new finding). Let's use `incumbent-pricing-responses-agents-data-tolls`. * **What's happening**: - As third-party AI agents threaten to erode the traditional per-seat licensing model (since a single autonomous agent can perform the work of dozens of seats), major enterprise software incumbents (ServiceNow, SAP, Workday) are shifting their pricing and access structures to protect their revenues and lock in customers. - **ServiceNow** unveiled **Action Fabric** at its Knowledge 2026 conference in May 2026. This is a metered integration layer that external/third-party AI agents must pass through to access data and execute workflows. It charges customers based on how many operations/actions an AI agent completes. JPMorgan analyst Mark Murphy called this effectively a "tax" or "tollgate" on customers using outside AI agents to interact with data they already store in ServiceNow. - **SAP** took an even more restrictive route. In April 2026, it updated its API policy (v4/2026) to explicitly prohibit third-party AI agents from interacting with its systems outside of SAP-endorsed architectures. The policy states that SAP "prohibits API use for: (a) interaction or integration with (semi-) autonomous or generative AI systems that plan, select, or execute sequences of API calls." This forces customers to route agent traffic through SAP's own Joule Agents, sparking massive lock-in concerns and criticism from the German-speaking user group DSAG. - **Workday** has also implemented **Workday Flex Credits** for AI agents, pricing standalone agents at $12 to $38 per FSE (Full-Time Equivalent) per month, and CEO Aneel Bhusri has noted that charging for agent access offers considerable financial upside. * **Quotes**: - From PYMNTS: *"ServiceNow, SAP and Workday are drawing new lines around the customer data stored inside their platforms as external artificial intelligence (AI) agents begin to erode the per-seat pricing model..."* - From PYMNTS: *"JPMorgan analyst Mark Murphy described the charge as effectively a tax on customers using outside AI agents to interact with data they already store in ServiceNow’s apps."* - From The Register (SAP API): *"except through and within the limits of SAP-endorsed architectures, data services, or service-specific pathways expressly identified and intended for such purposes, SAP prohibits API use for: (a) interaction or integration with (semi-) autonomous or generative AI systems that plan, select, or execute sequences of API calls..."* - From The Register (DSAG): *"Jens Hungershausen, chairman of the German-speaking user group DSAG board, told The Register... 'The unclear formulation of the policy creates uncertainty on the customer side. If you're uncertain, you probably won't do anything about it, and that's a risk that innovation is not taking place.'"* ### Finding 2: The Agentic Enterprise License Agreement (AELA) and the Return of "All-You-Can-Eat" Seat Pricing—With a Catch * **Slug**: `salesforce-aela-pricing-lock-in-risk` * **What's happening**: - To counter customer resistance to unpredictable consumption-based AI costs, Salesforce introduced the **Agentic Enterprise License Agreement (AELA)** in late 2025. It creates a flat, seat-based, "unlimited" arrangement over 2 or 3 years for Agentforce, Data Cloud, and MuleSoft. - Chief Revenue Officer Miguel Milano admitted that Salesforce is perfectly fine with losing money on these deals in the short term, because the extensive deployment of agents locks customers in for decades: *"If that is not profitable for me, it means that the customer is the happiest customer in the world. And then I have another 20 years to monetize that customer."* - However, Gartner analyst Hannah Decker warned in January 2026 that these "all-you-can-eat" buffet deals are a trap: *"Gartner believes that these are going to be converted into defined quantity contracts at the end of the agreement,"* leading to major cost shocks at renewal. Furthermore, vendors retain the right to change consumption multipliers/rates mid-contract, and Salesforce's 6% to 15% price hikes are feeding directly into renewal uplifts. * **Quotes**: - From Forrester: *"Salesforce’s new Agentic Enterprise License Agreement (AELA) gives customers unlimited use of consumption‑based products such as Agentforce, Data 360/Data Cloud, and MuleSoft for a fixed fee over two or three years."* - From The Register (December 2025): *"We take the risk because we want our customers to be successful. There's nothing that I would love more than a customer that I price... at $5 million incremental AELA, and the customer deploys so much that all of a sudden, that deal is not profitable for me... And then I have another 20 years to monetize that customer."* - From The Register (January 2026): *"Hannah Decker, Gartner director analyst for IT sourcing and procurement, said... 'Gartner believes that these are going to be converted into defined quantity contracts at the end of the agreement.' ... 'If they're moving to a defined quantity contract, there needs to be limits on price increases at renewal.'"* ### Finding 3: AI Overages and Forced Upgrades: The Hidden Cost Drivers of 2026 Renewals * **Slug**: `ai-overages-forced-upgrades-negotiation-leverage` * **What's happening**: - IT procurement advisory firm NPI released data in April 2026 highlighting that generative AI is driving massive cost increases and compliance risks for enterprise renewals, particularly with platforms like ServiceNow. - ServiceNow's generative AI (Now Assist) is gated behind Pro Plus and Enterprise Plus tiers, representing a 30% to 60% increase in per-user costs. - Furthermore, because AI usage grew 9x in H1 2025, customers are rapidly blowing through the annual "Assist" allotments bundled into these tiers. Overage costs are notoriously non-transparent, creating multi-million dollar "AI credit exposure" that hits buyers unexpectedly at renewal. - NPI advises procurement teams to require order form language specifying Assists per license, per-pack overage costs, and contractual consumption caps with notification obligations before overages accrue. * **Quotes**: - From NPI: *"Now Assist (ServiceNow’s generative AI suite) introduces a new overage risk. AI usage grew 9x in H1 2025 alone, quickly outpacing the annual Assist allotments bundled into Pro Plus and Enterprise Plus licenses. Overage costs are not transparently priced. Organizations may accumulate millions of dollars in AI credit exposure without visibility until the next renewal cycle."* - From NPI: *"Upgrading to these tiers represents a 30 to 60% increase in per-user cost. ServiceNow account teams are applying sustained pressure to move customers up, often framing the upgrade as a business-critical evolution."* Let's organize these three findings. Let's first search if there are other threads or notes we should review or update. We have: - `buyability-framework-linkedin` (score 0.70) - `enterprise-buying-journey-stages` (score 0.69) - `ai-table-stakes-new-baseline` (score 0.65) - `buyers-use-ai-dont-trust-it` (score 0.65) - `saas-inflation-budget-volatility-2026` (score 0.64) The findings we've uncovered perfectly address the thread `019e53e0-467d-7295-bd52-cbb63f1689db` ("Incumbent Pricing Responses: Tracking AELAs and Data Tolls Across SAP, ServiceNow, and Workday"). Let's resolve that thread! Wait, let's write the notes first. I will write three high-quality research notes using the `write_note` tool. Let's prepare the bodies of the notes. Every substantive finding must: - 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 - then say what it means. Let's write the first note about **Incumbent Pricing Responses: ServiceNow's Action Fabric and SAP's AI API Ban**. Slug: `incumbent-pricing-responses-agents-data-tolls` Topic: `019e4704-d70e-72d3-beaa-91954e215b8c` Let's write the note. I will state my intent first. I will write a note with detailed findings and quotes about ServiceNow's Action Fabric metered layer and SAP's restrictive API policy.I will write a detailed finding note tracking how enterprise software incumbents like ServiceNow and SAP are creating new pricing structures and technical "tolls" (such as Action Fabric and API bans) to monetize or restrict external AI agents that threaten their traditional seat-based models.

This note is successfully written. Now let's write the second note about **Salesforce's Agentic Enterprise License Agreement (AELA) and the Return of "All-You-Can-Eat" Seat Pricing—With a Catch**. Slug: `salesforce-aela-pricing-lock-in-risk` Let's do this. I will state my intent first. I will write a note detailing Salesforce's AELA strategy, why the CRO is willing to lose money on it to ensure long-term monetization, and why Gartner warns it is a potential renewal trap.

This note is also successfully written. Now let's write the third note about **AI Overages and Forced Upgrades: The Hidden Cost Drivers of 2026 Renewals**. Slug: `ai-overages-forced-upgrades-negotiation-leverage` Let's do this. I will state my intent first. I will write a note detailing how ServiceNow and other enterprise vendors are using generative AI (like Now Assist) to force tier upgrades (Pro Plus/Enterprise Plus) and how unbudgeted AI overages are creating massive financial exposures for buyers.

This is outstanding! All three detailed findings notes have been successfully persisted. Let's review the active threads to resolve the open thread. The active thread is: `019e53e0-467d-7295-bd52-cbb63f1689db` — "Incumbent Pricing Responses: Tracking AELAs and Data Tolls Across SAP, ServiceNow, and Workday" I will resolve this thread using the `resolve_thread` tool and provide a summary of our findings. Let's do that.

The thread is successfully resolved. Let's double-check if there are any other threads or rules. - "Call finish_cycle to end the cycle. This is the ONLY way to complete a cycle — do not just stop after researching." - "If this cycle genuinely surfaced nothing new worth persisting, still call finish_cycle and pass no_findings_reason explaining why." (We did surface and write great findings, so we don't need to pass `no_findings_reason`). Let's call `finish_cycle`. I will state my intent first. I will call `finish_cycle` to complete this research cycle.

I have completed the research cycle, updated the topic notes, and resolved the open thread. This cycle analyzed how major enterprise software incumbents (Salesforce, ServiceNow, SAP, and Workday) are adapting their pricing and licensing in response to agentic AI. I documented Salesforce's Agentic Enterprise License Agreement (AELA) "all-you-can-eat" flat-rate model, ServiceNow's newly unveiled "Action Fabric" metered action-based tollgate, SAP's highly restrictive April 2026 API policy update blocking third-party agents, and Workday's standalone agent pricing. All findings have been persisted as detailed research notes and the open thread has been resolved.