Cycle on May 26, 2026

I am going to list the open threads for the research topic `019e471f-94c7-7037-bf14-6994c512ce55` to see what specific questions or investigations are currently active.

I will perform a web search to gather information on Salesforce's Agentforce pricing models, specifically looking at how they are structured (e.g., the $2 per conversation model), how HubSpot and Atlassian are structuring their AI agent pricing, and how these companies are defending their traditional per-seat revenue models against erosion.

The search results reveal a very interesting development: Salesforce recently (looks like May 15, 2025/2026, let's verify the exact date and details) revamped its Agentforce pricing from a simple "per conversation" model (which faced customer confusion/backlash) to a more flexible model including "Flex Credits", action-based consumption, and even per-user options. Let's fetch several of these articles to understand: 1. The transition from $2/conversation to "Flex Credits" and "action-based" pricing. 2. The exact breakdown of the models (Conversations vs. Flex Credits vs. Per-seat). 3. How this acts as a defense of their per-seat revenue model, and the dynamics of customer adoption/backlash. I'll start by fetching the official Salesforce news release from May 15, 2025/2026, the Constellation Research insight, and the SaaStr post. Let's do a few fetches.

The Salesforce press release is very helpful. It was released on May 15, 2025 (less than a year ago, given today is May 26, 2026). This is a highly relevant, fresh piece of evidence. Let's fetch the other articles to see the analysis and backlash/confusion around the initial model, and how they are trying to protect per-seat revenue. I'll fetch: 1. `https://www.getmonetizely.com/blogs/the-doomed-evolution-of-salesforces-agentforce-pricing` (The Doomed Evolution of Salesforce's Agentforce Pricing) 2. `https://www.saastr.com/salesforce-now-has-3-pricing-models-for-agentforce-and-maybe-right-now-thats-the-way-to-do-it/` (SaaStr: Salesforce Now Has 3+ Pricing Models for Agentforce) 3. `https://www.constellationr.com/insights/news/salesforce-revamps-agentforce-pricing-flex-credits-what-you-need-know` (Constellation Research: Salesforce revamps Agentforce pricing with Flex Credits) Let's fetch the first one.

This is an incredibly detailed and rich article. It outlines the exact timeline of Salesforce's Agentforce pricing: 1. Fall 2024: Launch at $2 per conversation. Faced massive customer confusion, "sticker shock", lack of predictability, and backlash. Only ~8,000 of 150,000+ customers adopted it by May 2025. 2. May 15, 2025: Overhauled with three models: - **Flex Credits**: $0.10 per action (sold in packs of 100k for $500). Customers pay for discrete AI tasks (e.g., updating a record, sending an email) rather than conversations. - **Flex Agreement**: Allows companies to convert unused user licenses into Flex Credits and vice versa, creating a hybrid enterprise license. - **Per-User Licensing Options**: Unmetered agent usage for a fixed per-user-per-month price ($125/user/month for Sales/Service Cloud add-ons, or $550/user/month for premium "Agentforce 1" editions bundling core platform and credits). 3. Late 2025 - Early 2026: The "Agentic ELA" (Agentic Enterprise License Agreement) which helps Salesforce lock in customers by bundling AI into premium tiers, offering "all-you-can-eat" options to compete with Microsoft and prevent customers from using third-party overlay AI. Let's do some research on HubSpot and Atlassian to see how they structure their pricing for agentic AI. Let's run a search for HubSpot AI agent pricing and Atlassian Rovo pricing.

HubSpot has recently (April 2026!) flipped its AI pricing on its head with outcome-based Breeze agents! Let's fetch the CX Today article `https://www.cxtoday.com/crm/hubspot-outcome-based-ai-pricing-breeze/` and the SiliconANGLE article `https://siliconangle.com/2026/04/02/hubspot-flips-ai-pricing-head-outcome-based-breeze-agents/` (or the SaaStr one, let's look at SaaStr: `https://www.saastr.com/hubspot-switching-ai-pricing-from-per-use-to-per-resolution-but-does-it-really-matter/`). Let's fetch `https://www.cxtoday.com/crm/hubspot-outcome-based-ai-pricing-breeze/` and `https://siliconangle.com/2026/04/02/hubspot-flips-ai-pricing-head-outcome-based-breeze-agents/`.

This is a goldmine of information! HubSpot changed its Breeze AI agents pricing to an outcome-based model effective April 14, 2026. Let's see: - Customer Agent: $0.50 per resolved conversation (no resolution, no charge). This is a direct competitor to Intercom's $0.99 per resolution and Zendesk's outcome-based pricing. - Prospecting Agent: $1 per qualified lead (no qualified lead, no invoice). - This is marketed as "We believe AI should be priced on the value it delivers, not the compute it consumes." Let's fetch the SiliconANGLE article too to see if there are more details. This will help us compare Salesforce's Flex Credits / action-based model with HubSpot's outcome-based model ($0.50 per resolution / $1 per qualified lead).

Now we have two very strong pieces of evidence on Salesforce and HubSpot. Let's do some research on Atlassian's pricing for its AI agentic features, specifically "Atlassian Rovo" which was launched in late 2024/2025. Let's search for Atlassian Rovo pricing model to see if they are sticking to per-seat, or if they have some other structure.

This is a highly fascinating finding! Atlassian is using a hybrid model: "Rovo pricing combines fixed user licensing with variable AI Credit consumption." Let's fetch: 1. `https://support.atlassian.com/rovo/docs/rovo-usage-limits/` (Rovo usage allowance - Atlassian Support) 2. `https://www.atlassian.com/licensing/rovo` (Rovo Plans and Trial - Atlassian) 3. `https://www.eesel.ai/blog/atlassian-intelligence-and-rovo-pricing-explained` (Atlassian Intelligence and Rovo pricing explained) 4. `https://community.atlassian.com/forums/Rovo-questions/How-can-I-find-out-how-much-Rovo-will-cost-from-the-1st-March/qaq-p/3187868` (How can I find out how much Rovo will cost from the 1st March?) Let's fetch `https://www.atlassian.com/licensing/rovo` first, then `https://www.eesel.ai/blog/atlassian-intelligence-and-rovo-pricing-explained` or `https://support.atlassian.com/rovo/docs/rovo-usage-limits/`.

Let's fetch `https://support.atlassian.com/rovo/docs/rovo-usage-limits/` to understand exactly how Rovo credits are structured and how they are bundled into Standard, Premium, and Enterprise plans. This will show us how Atlassian is trying to defend its seat-based model by bundling AI into paid subscriptions with credit quotas rather than offering it completely unmetered or completely separate.

This is incredible! Let's analyze the Atlassian Rovo model: 1. Rovo is bundled directly into existing paid subscriptions (Standard, Premium, Enterprise) of Jira, Confluence, Jira Service Management, etc. 2. It uses a **hybrid model**: - Each user seat gets a monthly Rovo credit allowance (e.g., 25 credits for Jira Standard, 70 for Premium, 150 for Enterprise; or 250, 700, 1500 for Teamwork Collection). - These credits are **pooled at the organization level**. For example, 100 users on Confluence Premium + 50 users on Jira Premium = a pool of 10,500 credits per month. - Credits do not roll over. - Specific advanced AI features consume credits: - 10 credits per request to **Rovo Chat**. - 10 credits per request to **Rovo Agents**. - 100 credits per request for **Deep Research**. - Basic features like Rovo Search, summaries, and definitions do *not* consume credits. 3. Extra usage: Atlassian is **not currently billing** for extra usage above the allowance, but they will require an explicit opt-in and 90 days' notice before charging. 4. For Rovo Dev (specifically designed for developers, code reviews, planning), they charge **$20 per developer per month**, which includes 2,000 Rovo Dev credits per month. Extra usage is billed at $0.01 per credit. To avoid disruption, a default extra usage of 2,000 credits ($20) is auto-enabled, but admins can cap it. This is a beautiful example of how an incumbent (Atlassian) is structuring its agentic pricing: - Instead of moving away from per-seat pricing, Atlassian is **deepening its per-seat value** by bundling a base allowance of "pooled AI credits" directly into existing per-seat tiers (Standard, Premium, Enterprise). - This encourages customers to upgrade to higher seat-based tiers (Premium or Enterprise) because it increases their shared organization-level pool of AI credits (e.g., Jira Premium gives 70 credits/user/month vs. 25 for Standard; Enterprise gives 150/user/month). - For highly specialized workflows (like Rovo Dev), they create a separate seat-based add-on ($20/developer/month) that also has a credit quota (2,000 credits/month) and usage-based overage ($0.01/credit). This protects their margins against heavy LLM compute costs while maintaining a predictable seat-based billing structure. Let's look at the thread that we listed earlier: "Investigate how legacy SaaS incumbents (e.g., Salesforce with Agentforce, HubSpot, Atlassian) are structuring their agentic pricing models (such as credit-based or consumption-based models) to defend against the erosion of their traditional per-seat revenue. What are the early adoption patterns, contract renegotiation dynamics, and churn risks for these incumbent offerings?" This thread is a perfect fit for a comprehensive finding that compares these three distinct incumbent strategies for agentic pricing and how they defend per-seat revenue! Let's do a quick search on "HubSpot Breeze vs Salesforce Agentforce pricing" and "Zendesk outcome based pricing" to see if there is any discussion of seat erosion or how these companies are trying to prevent it. I'll search for: `"per-seat" erosion AI agents Salesforce HubSpot Atlassian` to see what industry commentators say about seat erosion.

These are incredible leads! Let's fetch the taskade.com article `https://www.taskade.com/blog/great-saas-unbundling` and the Digital Applied blog `https://www.digitalapplied.com/blog/saaspocalypse-ai-agents-software-industry-analysis` if they are high-quality, or we can look at the LinkedIn post by Silvia Gomez Del Pozo: `https://www.linkedin.com/posts/silvia-gomez-del-pozo_i-saved-the-most-iconic-model-for-last-seat-based-activity-7446564221779304448-ZyHU` (if it's searchable) or the substack article `https://sergeycyw.substack.com/p/crm-itsm-the-seat-based-era-is-ending`. Let's fetch `https://www.taskade.com/blog/great-saas-unbundling`.

This is a fantastic article! It has a wealth of concrete data, including: - "The February 2026 Reckoning": $285 billion in market capitalization evaporated from SaaS stocks in what analysts dubbed the "SaaSpocalypse." Per-seat companies were hit hardest. - Monday.com replaced 100 SDRs with AI agents (response time down from 24 hours to 3 minutes, 90%+ reduction in team costs, 90% seat compression across the supporting SaaS stack). - Atlassian reported its first-ever enterprise seat decline. - The shift in pricing models: usage-based, outcome-based, credit-based. - How legacy per-seat vendors are trying to adapt. Let's do a quick search on "Atlassian seat decline 2026" or "Atlassian earnings seat compression" to confirm the details about Atlassian's seat decline and see if there are any specific quotes or reports about it. This will make our finding incredibly robust and concrete. I'll search for: `Atlassian "seat decline" OR "seat compression" 2026` or `Atlassian earnings report "seats" decline` or `Atlassian first-ever seat decline`.

This is an incredible set of results! Let's see: - There is a major debate about "seat compression" in the market. - In February 2026, Atlassian's stock was under pressure due to "seat compression" worries and a broader software selloff. - However, in late April 2026 (their Q3 2026 earnings call, which would be for the quarter ending March 31, 2026, reported on April 30, 2026), Atlassian reported strong results and their stock rose 30% after Q3 earnings. - Let's read the Q3 2026 earnings transcript or highlights to see exactly what they said about seat compression. - Let's fetch `https://finance.yahoo.com/markets/stocks/articles/atlassian-corp-team-q3-2026-072148354.html` and `https://www.investing.com/news/transcripts/earnings-call-transcript-atlassian-beats-q3-2026-forecasts-stock-dips-93CH-4651640` or `https://www.aol.com/finance/atlassian-team-q3-2026-earnings-223226154.html`. Let's fetch `https://www.investing.com/news/transcripts/earnings-call-transcript-atlassian-beats-q3-2026-forecasts-stock-dips-93CH-4651640` or `https://finance.yahoo.com/markets/stocks/articles/atlassian-corp-team-q3-2026-072148354.html`. Let's use `web_fetch` on the Yahoo Finance or Investing.com transcript. Let's try `https://www.aol.com/finance/atlassian-team-q3-2026-earnings-223226154.html` first.

This is incredibly interesting! Let's see: In February 2026, the SaaS market faced a major "SaaSpocalypse" with $285 billion in market cap evaporating due to "seat compression" fears. However, in Atlassian's Q3 2026 earnings call (reported April 30, 2026), they *strongly* pushed back against the seat compression narrative: - "AI Rovo credit usage is growing more than 20% month over month." - "Customers using Rovo are also growing their ARR at roughly two times the rate of customers who are not using Rovo." - "Growth attributed to cross-sell in Collections, especially Teamwork Collection, and continued seat expansion in core Jira standalone offerings; no evidence cited of seat compression." - "The company is leveraging both seat-based and consumption-based pricing models, with Teamwork Collection seen as a leading vector for AI feature adoption." - "Collections Pricing Dynamics — Teamwork Collection customers receive 10x more credits on Rovo versus the standalone subscription and use more than twice as many credits per user as standalone subscribers." This is a beautiful, nuanced picture! While the *fear* of seat compression (prompted by things like Monday.com replacing 100 SDRs with AI agents) is driving SaaS valuations down, incumbents like Atlassian are successfully fighting seat compression by: 1. **Bundling AI Credits to Drive Seat Expansion**: They bundle Rovo AI credits into premium collections (like Teamwork Collection, which gives 10x more credits than standalone Jira/Confluence). This incentivizes customers to buy the higher-tier/higher-priced collections (which increases ARR), and actually drives seat expansion in core Jira. 2. **Platform Context Lock-In**: Because Atlassian's "Teamwork Graph" has all the company's data, context, and history, customers are choosing Atlassian's native AI (Rovo) over third-party overlay AI. This drives a "2x ARR growth rate" for Rovo-using customers compared to non-Rovo customers. Let's look at HubSpot's strategy too: HubSpot is adopting an **Outcome-Based Pricing Model** (effective April 14, 2026): - $0.50 per resolved conversation (Breeze Customer Agent) - $1.00 per qualified lead (Breeze Prospecting Agent) - This removes the "blank check" risk for buyers (they only pay when it works) and allows HubSpot to capture value even as human seat counts compress. If an AI agent replaces human SDRs (Monday.com case), HubSpot can still monetize via the $1 per qualified lead model, capturing a share of the "digital labor" value. And Salesforce's strategy: Salesforce overhauled Agentforce pricing in May 2025 to offer **three models**: 1. **Flex Credits** ($0.10 per action, sold in $500 packs of 100k): Pay-for-use for discrete AI tasks (updating a record, sending an email). 2. **Flex Agreement**: Allows enterprises to convert unused user licenses (seats) into Flex Credits and vice versa. This is a direct defense against seat compression—if a customer automates human roles and needs fewer seats, Salesforce doesn't lose the revenue; the budget is simply transferred to Agentforce Flex Credits. 3. **Per-User Unlimited Licenses** ($125/user/month add-on, or $550/user/month premium Agentforce 1 Editions): Classic seat-based SaaS pricing for unlimited internal AI use, providing cost predictability and a price ceiling. Let's organize this into a beautifully detailed finding that addresses the open thread perfectly! Let's double-check if there are any other notes we should read or write. First, let's read the current notes for the topic to see if we should update an existing note or write a new one. I'll read the notes using `read_notes` tool.

Let's see if we should write a new note specifically detailing the incumbent pricing strategies or update an existing note. The existing notes include: - `ai-pricing-models-outcome-consumption-2026` (The Death of Flat-Seat SaaS: The Shift to Outcome-Based and Pooled Consumption Pricing) - `ai-agent-pricing-churn-auditing-2026` (The AI Agent GTM Shift: Outcome-Based Pricing, Portable Prompt Churn, and the Rise of Agent Auditing) Wait, the topic brief and the thread specifically ask us to investigate how *legacy SaaS incumbents* (e.g., Salesforce with Agentforce, HubSpot with Breeze, Atlassian with Rovo) are structuring their agentic pricing models (such as credit-based or consumption-based models) to defend against the erosion of their traditional per-seat revenue, early adoption patterns, contract renegotiation dynamics, and churn risks. This is a highly focused question about *incumbents defending seat revenue* and the specific mechanisms they use (Salesforce's Flex Credits and Flex Agreements, HubSpot's Breeze outcome-based pricing, Atlassian's pooled Rovo credits). Let's write a fresh, comprehensive note with a title like: "Incumbent Agentic Pricing: How Salesforce, HubSpot, and Atlassian Defend Per-Seat Revenue Models" (slug: `incumbent-agentic-pricing-defense-models`). This maps directly to the thread and provides a deep, evidence-grounded analysis of this specific, crucial shift in 2026. Let's structure the note beautifully, including: 1. **The Context of the 2026 "Seat Compression" Crisis**: The "SaaSpocalypse" of February 2026 where $285B in market cap was wiped out on fears of seat erosion (e.g., Monday.com replacing 100 SDRs with AI agents). 2. **The Salesforce Defense Playbook**: - Initial $2/conversation model (2024) failed due to lack of predictability, causing sticker shock and low adoption (~8k out of 150k+ customers as of May 2025). - The May 15, 2025 Overhaul: **Flex Credits** ($0.10/action), **Flex Agreements** (allowing seat-to-credit conversion to prevent budget loss), and **Per-User Unlimited Licenses** ($125/user/month add-ons or $550/user/month premium "Agentforce 1" editions) to offer a price ceiling. 3. **The HubSpot Outcome-Based Pivot**: - Effective April 14, 2026, HubSpot flipped Breeze agent pricing to pure outcomes: **$0.50 per resolved conversation** (Customer Agent) and **$1.00 per qualified lead** (Prospecting Agent). - This removes "buyer risk", aligns cost with value, and allows HubSpot to capture a share of the "digital labor" value even as human seat counts compress. 4. **The Atlassian Hybrid Pooled Credits Strategy**: - Atlassian Rovo (GA late 2025/2026) bundles AI credits directly into existing per-seat tiers (Standard, Premium, Enterprise). - Credits are **pooled at the organization level** (e.g., Jira Premium gives 70 credits/user/month, Enterprise gives 150), which incentivizes upgrading to higher seat-based tiers rather than compressing them. - For specialized workflows (Rovo Dev), they charge $20/developer/month seat fee + 2,000 credit allowance, with $0.01 per additional credit overage. - Q3 2026 earnings (April 30, 2026) showed 20% MoM growth in Rovo credit usage and a *pushback* against seat compression fears, with Rovo-active customers growing ARR at twice the rate of non-Rovo customers. 5. **Key Takeaways for GTM Playbooks**: How to structure pricing to offer predictability, manage margins against compute costs, and align with buyer value perception. Let's do a quick review of the sources and quotes we will use to ground this: - **Salesforce News (May 15, 2025)**: - *"Flex Credits ensure you only pay for the exact actions Agentforce performs... Each of these actions consumes just 20 Flex Credits ($0.10 per action)..."* - *"Salesforce’s new Flex Agreement allows organizations to manage both human and digital labor and strategically shift their investments between user licenses or digital labor..."* - *"new Agentforce user licenses and add-ons... offering unmetered employee-facing agent usage in a simple per user per month (PUPM) pricing model."* - **The Doomed Evolution of Salesforce's Agentforce Pricing (getmonetizely.com, Jan 21, 2026)**: - *"What began as a simple-sounding $2 per conversation fee quickly met with customer confusion and backlash, forcing Salesforce to overhaul its approach. By early 2026, Salesforce had shifted to a far more complex scheme combining usage-based credits and per-user licensing..."* - *"In LinkedIn discussions, many predicted that 2026 would bring a backlash against opaque credit-based pricing as customers struggle with budgeting for 'AI credits'..."* - **HubSpot Joins the Outcome-Based Pricing Revolution (CX Today, Apr 2, 2026)**: - *"Effective April 14, HubSpot is moving its Breeze Customer Agent and Breeze Prospecting Agent to an outcome-based pricing model. The Customer Agent will be charged at $0.50 per resolved conversation; the Prospecting Agent at $1 per qualified lead."* - *"We believe AI should be priced on the value it delivers, not the compute it consumes."* - **Atlassian Rovo support and FAQ docs / Q3 2026 Earnings Call (Apr 30, 2026)**: - Atlassian Support: *"Rovo credits... Are provided as a fixed allowance per seat per month, based on your edition... Are pooled at the organization level... Reset monthly."* - Atlassian TEAM Q3 2026 Earnings: *"AI Rovo credit usage is growing more than 20% month over month," and customers using Rovo "are growing their ARR at roughly two times the rate of customers who are not using Rovo."* - **Great SaaS Unbundling: AI Agents vs Per-Seat (Taskade, 2026)**: - *"In February 2026, the market delivered its verdict. $285 billion in market capitalization evaporated from SaaS stocks in what analysts dubbed the SaaSpocalypse. Per-seat companies were hit hardest."* - *"Monday.com replaced 100 SDRs with AI agents... SaaS seats needed: 500+ (across stack) -> ~50 (90% compression)."* This is an exceptionally strong, comprehensive, and up-to-date finding. I will write this note now. Let's make sure the formatting is exactly as requested (no YAML frontmatter, `# ` heading is the title, slug parameter is passed).

This note is successfully written! Now, let's look at the active thread we listed earlier: `019e5e2f-fe35-73dd-ad6b-c52ea821e5f7`: "Investigate how legacy SaaS incumbents (e.g., Salesforce with Agentforce, HubSpot, Atlassian) are structuring their agentic pricing models (such as credit-based or consumption-based models) to defend against the erosion of their traditional per-seat revenue. What are the early adoption patterns, contract renegotiation dynamics, and churn risks for these incumbent offerings?" Our newly created note `incumbent-agentic-pricing-defense-models` directly and comprehensively addresses this thread! Let's resolve this thread with a clear explanation of how our findings address it. I will call `resolve_thread`.

The thread is now resolved! Let's see if there are any other threads or if we should open any new threads. Let's think: Are there any other interesting open questions that surfaced during this cycle that we should record as new threads for future cycles? Yes! A very interesting question is: **"The Churn Dynamics and ROI Attribution of Outcome-Based AI Pricing: How HubSpot, Zendesk, and Intercom Measure and Validate 'Resolutions' and 'Qualified Leads' to Enterprise Buyers."** As Matt Price (CEO of Crescendo) pointed out in the CX Today article, resolution rate alone does not tell the full story, and "a lot of AI is actually adding friction to the journey, and those customers end up being unhappier." How are buyers auditing these outcomes? Is there a risk of high churn when buyers realize "resolutions" might be low-quality or lead to negative customer satisfaction? Let's record this as a new thread to guide future research cycles on this topic. I will call `record_thread` to open this question.

Ah, the tool description says: "kind is one of 'question', 'investigation', 'contradiction'." However, the error says "invalid thread kind 'investigation'". Let's look closer at the error or try 'question'. Let's call `record_thread` with kind `question`.

The thread is recorded successfully! Now that I have: 1. Listed the threads. 2. Conducted thorough web research and fetched official resources and analysis. 3. Written a comprehensive and deeply grounded finding note (`incumbent-agentic-pricing-defense-models`). 4. Resolved the active thread. 5. Recorded an open question for future cycles. I am ready to call `finish_cycle` to complete this research cycle. I will proceed to do so.

I have successfully completed the research cycle for topic `019e471f-94c7-7037-bf14-6994c512ce55`. ### Summary of Activities & Findings 1. **Resolution of Active Thread:** * **Thread Investigated:** *Incumbent Agentic Pricing: How Salesforce, HubSpot, and Atlassian Defend Per-Seat Revenue Models* (ID: `019e5e2f-fe35-73dd-ad6b-c52ea821e5f7`). * **Findings:** We uncovered three distinct strategies legacy SaaS incumbents are using to defend against traditional per-seat revenue erosion caused by "seat compression" (e.g., Monday.com replacing 100 SDRs with AI agents, causing a 90% seat-count collapse in supporting tools): * **Salesforce (Agentforce):** After its initial $2/conversation model suffered severe backlash and low adoption, Salesforce overhauled its pricing on May 15, 2025. It introduced **Flex Credits** ($0.10 per action), the **Flex Agreement** (allowing customers to convert unused human seat licenses into digital labor credits to prevent contract value erosion), and **Per-User Unlimited Licenses** ($125/user/month add-ons or $550/user/month premium "Agentforce 1" editions) to offer a cost ceiling. * **HubSpot (Breeze):** Effective April 14, 2026, HubSpot shifted its Breeze Customer and Prospecting agents to a pure **outcome-based model** ($0.50 per resolved conversation and $1.00 per qualified lead). This removes forecasting risk for buyers and allows HubSpot to capture a share of the direct value of digital labor, offsetting any human seat contraction. * **Atlassian (Rovo):** Atlassian bundles its AI assistant directly into existing Cloud seat-based subscriptions (Standard, Premium, Enterprise) using a **pooled credit quota system**. This encourages enterprises to upgrade to premium tiers and "Collections" (like Teamwork Collection, which yields 10x more credits) to increase their organization-wide credit pool. Their Q3 2026 earnings (reported April 30, 2026) showed a 20% MoM growth in Rovo credit usage and a *pushback* against seat compression fears, with Rovo-active customers growing ARR at twice the rate of non-Rovo customers. * **Artifact Created:** Written as a comprehensive research note: `incumbent-agentic-pricing-defense-models`. 2. **Recorded Open Question for Future Cycles:** * **Thread Opened:** *The Auditing and Churn Risks of Outcome-Based AI Pricing: Verifying 'Resolutions' and 'Qualified Leads'* (ID: `019e6506-50e6-7e68-92d5-45048ad8008c`). * **Objective:** Investigate how enterprise buyers audit and verify "resolutions" and "qualified leads" to ensure they aren't paying for low-quality automated interactions that add customer friction, and explore the subsequent customer retention and churn risks for outcome-based vendors.