Cycle on Aug 24, 2026

I will list the open threads and read the existing notes for this topic to understand the past research and see what needs to be pursued or updated.

I will read the existing notes to see what subjects and slugs have already been created for this topic.

I will search for recent news and events related to enterprise software, SaaS, AI disruption, and companies mentioned in the watches, such as ZoomInfo, HubSpot, Pegasystems, GitLab, NICE, Sierra AI, Salesforce, and Oracle. Let's see what has been happening in August 2026. I'll start by doing a web search for general enterprise software earnings or news in August 2026. Let's do a search on enterprise software companies or specific earnings in August 2026. Let's search for "enterprise software" "AI displacement" or "seat compression" or "earnings" in August 2026. Let's run a search.

Let's look at the Futurum Group's 2H 2026 Enterprise Software Decision-Makers Survey and other reports. Let's search for "Futurum Group" "2H 2026 Enterprise Software Decision-Makers Survey" or "Why Core Software and AI Can't Share One Commercial Model". Let's fetch that page: `https://futurumgroup.com/insights/the-pricing-split-why-core-software-and-ai-cant-share-one-commercial-model/`. Let's see if we can read it. Or let's search specifically for Futurum Group enterprise software pricing.

This is an incredibly rich and highly relevant article published on August 17, 2026! It directly addresses several of our threads and prior findings. Let's see what else we should look for. We have watches for: 1. ZoomInfo's transition to hybrid consumption pricing and whether their non-seat Operations business continues to grow at 20%+ to offset seat compression. Let's do a search on ZoomInfo Q2/Q3 2026 earnings or recent news. 2. HubSpot's Q3 2026 earnings / September 2026 Analyst Day. Wait! The current date is August 24, 2026. HubSpot's Q3 earnings are expected early November 2026, and their Analyst Day is September 2026. Since today is August 24, 2026, the September 2026 Analyst Day has not happened yet! We should check if there's any pre-announcement or preview or if Q2 earnings (reported on August 5, 2026) are fully documented. (We have a note about HubSpot Q2 2026 earnings). Let's double check if there are any updates or if we should register/check other things. 3. Pegasystems (PEGA) Q3 2026 earnings expected early November 2026. Wait! The Futurum article says Pegasystems' ETR TSIS Net Score fell sharply in July 2026 to +0.9% (from +12.4% in April 2026). Let's see if we should record an event or update our notes on Pegasystems. Let's research Pegasystems and its recent performance. 4. GitLab's Q3 2026 results expected early December 2026. Wait, what about Q2 2026 (or Q2 FY2027) results for GitLab? GitLab's fiscal year is ahead by 1 year, so Q2 FY2027 would be reported in late August or early September 2026. Let's search for GitLab Q2 FY2027 or Q2 2026 earnings. 5. Vercel's AI gateway traffic share crossing 40% threshold. Let's search for Vercel AI gateway traffic share, open source models, etc. 6. NICE's cloud revenue growth and Q3 2026. It is expected in November 2026. Let's check if there is any other recent news on NICE. 7. Sierra AI ARR growth beyond $200M milestone or IPO/M&A. 8. Salesforce acquisition of Fin (Intercom) closing in Q4 FY2027. 9. Oracle FY2027 Capex and Free Cash Flow. 10. Salesforce organic revenue growth reacceleration in H2 FY2027. Let's check if Salesforce has reported Q2 FY2027 (which should be late August 2026). Oh! Salesforce Q2 FY2027 (ended July 31, 2026) is typically reported in late August 2026. Let's search for "Salesforce Q2 FY2027 earnings" or "Salesforce Q2 2026 earnings" to see if they've reported or are about to report. Let's do some web searches to find the latest data. First, let's search for GitLab's latest earnings (Q2 FY2027 / Q2 2026) and Salesforce's latest earnings.

Ah, GitLab's Q2 FY2027 results are scheduled for September 1, 2026. This is in about 8 days (since today is August 24, 2026). Let's keep that in mind. Now let's search for Salesforce's Q2 FY2027 earnings (which would be for the quarter ended July 31, 2026). Let's see when they report. Usually, Salesforce reports in late August. Let's do a search.

Wait, look at this! There is a tweet or some preview saying: "🇺🇸 $CRM | SALESFORCE Q2 FY2027 ... Q2 FY2027: Salesforce is showing that AI monetisation is finally moving into the numbers. Key numbers: • Revenue: $11.3B | +10% YoY • ..." But wait, Salesforce's official earnings release date is August 26, 2026. Today is August 24, 2026. So the actual earnings announcement hasn't happened yet (it's in 2 days, on Wednesday, Aug 26). The tweet or preview must be a placeholder or a preview or an error or maybe a leak? No, let's look at the dates. The official release is August 26, 2026. Let's see if there are any other earnings reported in mid-August 2026. Let's search for "ZoomInfo" Q2 2026 or "HubSpot" Q2 2026 or other companies. Wait! We have a note: "SaaSpocalypse Vulnerabilities: How AI Agents are Displacing ServiceNow, ZoomInfo, and HubSpot" (updated August 17, 2026) and "ZoomInfo's Strategic Pivot: Q2 2026 Goodwill Impairment, GTM.AI Headless Layer, and Hybrid Consumption Migration" (updated August 10, 2026). Let's read these notes to see what was covered. Wait, I can read their full bodies if I need to, but actually I don't need to re-fetch if they are already in my context. Let's do a search on ZoomInfo or HubSpot to see if there are newer updates, or let's search for "Workday" or "Snowflake" or "Autodesk" or other enterprise software earnings in August 2026. Let's search for "Workday" Q2 FY2027 or Q2 2026 earnings (reported in August 2026). Let's do a search.

Okay, so Workday is reporting on August 27, 2026 as well. Let's see if there are other enterprise software earnings in August 2026. Let's search for Snowflake. Snowflake's Q2 FY2027 (quarter ended July 31, 2026) is usually reported in late August. Let's do a search.

So Snowflake reports on September 2, 2026. Let's search for "Salesforce" "Agentforce" in August 2026. Or let's search for some other enterprise software earnings or news that happened recently in August 2026, or other topics. Wait! Let's search for "enterprise software" "earnings" "August 2026" or "AI displacement" to see if there are other companies reporting. Let's see: what about Autodesk? Or what about others? Let's do a search on: `enterprise software earnings "August 2026" OR "July 2026"` to see which companies recently reported.

Let's look at the ICONIQ State of AI report, July 2026 edition: "The Builder's Economy". Let's search for "ICONIQ State of AI" "July 2026" or "The Builder's Economy". Let's run a search.

Let's fetch the OnlyCFO post or the SaaSletter post about the ICONIQ State of AI July 2026 report. Wait, the SaaSletter post is: `https://www.saasletter.com/p/iconiq-state-of-ai-july-2026-ai-revenue-mix`. Let's fetch it. Or the OnlyCFO post: `https://www.cfopilot.com/p/iconiq-releases-2026-state-of-ai`. Let's fetch both if possible, or start with OnlyCFO. This will give us excellent data on AI COGS, pricing models, token spend, and margin compression/expansion. Let's fetch OnlyCFO first.

This is excellent. Let's look at the actual PDF if we can get some details, or let's search for "SaaSletter" "ICONIQ" to see the specifics they mentioned. Let's fetch the SaaSletter URL `https://www.saasletter.com/p/iconiq-state-of-ai-july-2026-ai-revenue-mix`. Let's fetch it. This is a great way to understand the changing gross margin dynamics of AI products! Let's do that.

This is incredible. We have two major reports from late July / August 2026: 1. **The Futurum Group's 2H 2026 Enterprise Software Decision-Makers Survey** (published August 17, 2026): - Shows a fascinating "Pricing Split" where per-seat pricing ranks *last* for core software (12.6%) but *first* for generative AI add-ons billed separately (42.3%). - ETR TSIS (Technology Spending Intentions Survey) data shows that pure outcome-based players like Zendesk and Intercom have flatter or more volatile Net Scores (Zendesk at -1.7% in July 2026, Intercom at -11.1%), whereas hybrid players like ServiceNow (+42.8%), Salesforce (+19.0%), and Adobe (+15.5%) have steadier spending momentum. - Pegasystems (PEGA) is a major exception in the hybrid camp, with its ETR Net Score crashing from +12.4% in April 2026 to +0.9% in July 2026, alongside a Pervasion dip. This is highly relevant to our Pegasystems watch thread! - Futurum predicts that pricing optionality (at least consumption/outcome for core and predictable/seat for AI) will become a shortlist gate by Q4 2026. 2. **ICONIQ's "State of AI: The Builder's Economy" Report** (published July 16, 2026): - Survey of over 300 software companies building AI products. - Shows AI gross margins are expanding from 45% in 2025 to a projected 53% in 2026 and 59% in 2027. This is driven by "blocking and tackling" cost-reduction efforts: 66% of respondents improved cost per query by 10%+, and 19% by 30%+. - Key drivers of unit economics: reducing inference costs, implementing routing strategies, revenue growth through cost leverage, and switching to lower-cost/open-source models. - Revenue mix for non-AI-native startups is shifting heavily to AI: 32% of total revenue in 2025, 42% in 2026, and projected 53% in 2027. - Companies are using an average of 3.3 model providers, with cheaper/open-source models gaining traction. Let's double-check if we can find any other recent information on ZoomInfo or other companies. We have a watch for ZoomInfo: "Track ZoomInfo's transition to hybrid consumption pricing and whether their non-seat Operations business continues to grow at 20%+ to offset seat compression." Let's search for ZoomInfo's recent earnings or updates in August 2026. Wait! We noted earlier that ZoomInfo reported Q2 2026 results on August 5, 2026. Let's see if there's any commentary on ZoomInfo's pricing model or operations business growth. Let's do a search on ZoomInfo Q2 2026.

Let's fetch the Yahoo Finance page or Seeking Alpha page on ZoomInfo's Q2 2026 earnings call, or let's search for "ZoomInfo" "Operations" growth rate Q2 2026 to see if we can get the specific percentage. Let's search for "ZoomInfo" "Operations" growth rate Q2 2026 or "non-seat" or "hybrid consumption" to see how that business grew in Q2 2026. This is directly related to our ZoomInfo watch thread! Let's do a search.

This is perfect. ZoomInfo's Operations business (which is primarily data and not tied to seats) grew by exactly 20% year-over-year in Q2 2026. Let's double-check the watch description for ZoomInfo: "Track ZoomInfo's transition to hybrid consumption pricing and whether their non-seat Operations business continues to grow at 20%+ to offset seat compression." Ah! The watch is "ongoing". We should record an event on it because ZoomInfo reported Q2 2026 results on August 5, 2026, showing that their Operations business grew exactly 20% year-over-year, meeting the target threshold. Wait, let's look at the other watches. Let's see: - HubSpot: Expected Q3 2026 in early November. This is in the future. - Pegasystems: Expected Q3 2026 in early November. This is in the future. Wait! Is there an event on Pegasystems? The watch says: "Track whether Pegasystems (PEGA) Q3 2026 earnings show a recovery... or if buyer paralysis continues... Check if Pegasystems' ACV growth recovers and if management reports easing of AI-related customer confusion in Q3 2026 earnings." This is a one-shot expected in November. But wait, we just saw in the Futurum brief that Pegasystems' ETR TSIS Net Score fell sharply in July 2026 to +0.9% (from +12.4% in April 2026), showing that buyer paralysis and confusion are very much ongoing and possibly worsening! This is a great data point to add to our notes, and we should keep the watch open. - GitLab: "Track if GitLab's 14% workforce reduction and generational rebuild... result in accelerated revenue growth... Q3 2026 results are expected in early December 2026." This is a one-shot expected in December. - Open-Source AI Gateway Traffic Share Crossing 40% Threshold: Let's search for "Vercel AI Gateway" "open source" "traffic share" or similar phrases to see if there is any new data on this. Let's do a search.

Ah! My web search budget is spent. No problem, I have incredibly rich material from the searches I already performed! Let's analyze what I have. Let's review the findings and watches: 1. **ZoomInfo Watch**: - The watch is: "Track ZoomInfo's transition to hybrid consumption pricing and whether their non-seat Operations business continues to grow at 20%+ to offset seat compression." - In ZoomInfo's Q2 2026 earnings (reported August 5, 2026), management confirmed that their Operations business (which is primarily data-driven and not tied to seats) grew by exactly 20% year-over-year. They also announced plans to introduce more flexible consumption-based pricing later in Q3 2026, with customer migrations continuing through late 2026 and 2027. - This represents a successful check of the "ongoing" watch. I will call `record_event` to register that the Operations growth rate met the 20% target and consumption-based pricing details were shared. Since it's an ongoing watch, I will set `resolves: false`. 2. **Pegasystems Watch / ETR TSIS Net Score Decline**: - The watch is: "Track whether Pegasystems (PEGA) Q3 2026 earnings... show a recovery in ACV growth (above Q2's 7-8%) and a resolution of customer AI confusion..." - The Futurum brief (published August 17, 2026) reveals that Pegasystems' ETR TSIS Net Score fell sharply from +12.4% in April 2026 to +0.9% in July 2026. This is the sharpest single-quarter decline among the six enterprise vendors with ETR coverage tracked in the brief. This suggests that the customer AI confusion and paralysis we have been tracking are having a massive, immediate impact on buyer spending intentions ahead of their Q3 earnings. - I will update the Pegasystems finding or write a new note that integrates this ETR data. Wait, let's look at our existing notes. - Our existing notes include: - `ai-buyer-paralysis-legacy-software-slowdown-2026` ("AI Buyer Paralysis: "Value Maxing" and Elongated Sales Cycles Drive Historic Software Pullbacks"). This note is about Pegasystems and others. - Let's read this note or write an update to it! - Wait, the slug is `ai-buyer-paralysis-legacy-software-slowdown-2026`. I can update this note to include the new ETR TSIS data showing Pegasystems' Net Score collapse to +0.9% in July 2026, confirming that buyer paralysis is worsening. - Let's check if there is an existing note specifically on Pegasystems. No, `ai-buyer-paralysis-legacy-software-slowdown-2026` is the closest. Let's update it in place. 3. **Enterprise AI Pricing Split (New Finding)**: - This is a massive, highly strategic theme from the Futurum Group's August 17, 2026 brief: "Why Core Software and AI Can't Share One Commercial Model". - It shows a stark bifurcation in buyer preferences: buyers rank per-seat pricing *last* for core software (12.6%) but *first* for generative AI add-ons billed separately (42.3%). - Buyers want consumption/outcome flexibility for core software, but predictability (per-seat or capped) for AI because AI usage costs are too volatile and hard to model. - This cuts directly against the popular narrative that "per-seat pricing is dead across the board." Instead, it's a "Pricing Split." - Zendesk and Intercom (pure outcome-based) are showing flatter or volatile ETR Net Scores (Zendesk -1.7% in July 2026, Intercom -11.1%), whereas hybrid players like ServiceNow (+42.8%), Salesforce (+19.0%), and Adobe (+15.5%) have steadier spending momentum. - We should write a new finding on this: `enterprise-ai-pricing-split-core-vs-add-on-2026` or reuse an existing slug. Wait, let's look at existing slugs: - `saas-pricing-model-siege-build-vs-buy-2026` ("SaaS Pricing Models Under Structural Siege: Enterprise Customers Pay More for the Same Tools") - `outcome-based-ai-pricing-disruption-ccaas-crm-2026` ("The Seismic Shift to Consumption and Outcome-Based AI Pricing: Microsoft, Salesforce, and Startups Pivot to Mitigate Seat Compression") - `zendesk-outcome-pricing-autonomous-service-2026` ("Zendesk's Triple-Layered Pricing and the Customer Friction Threat") - Is there a note that tracks the broader pricing shift? Yes, `outcome-based-ai-pricing-disruption-ccaas-crm-2026` or `saas-pricing-model-siege-build-vs-buy-2026`. - Let's look at `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. It's titled "The Seismic Shift to Consumption and Outcome-Based AI Pricing: Microsoft, Salesforce, and Startups Pivot to Mitigate Seat Compression". We can update this note or create a more specific one on the "Pricing Split" or update `saas-pricing-model-siege-build-vs-buy-2026`. - Wait, the Futurum report is extremely specific about the "Pricing Split: Why Core Software and AI Can't Share One Commercial Model." Let's write a new finding with slug `enterprise-software-pricing-split-2026` or update `saas-pricing-model-siege-build-vs-buy-2026` or `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. Let's update `outcome-based-ai-pricing-disruption-ccaas-crm-2026` because it's the stable subject tracking the transition of pricing models (consumption, outcome, seat-based) and how vendors pivot. - Wait, let's see if there is another note: `saas-pricing-model-siege-build-vs-buy-2026`. Let's update `outcome-based-ai-pricing-disruption-ccaas-crm-2026` with this new, highly detailed data, or write a dedicated note on the "Pricing Split" since it represents a major counter-narrative to the "seat-based models are dying" consensus. Let's write a dedicated note or update `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. Let's update `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. Let's re-read the slug name: `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. Yes, this is perfect. 4. **ICONIQ "State of AI: The Builder's Economy" (New Finding)**: - This is a major bi-annual report published in July 2026. - It shows that AI gross margins are expanding (from 45% in 2025 to 53% in 2026 and 59% projected in 2027) because of "blocking and tackling" cost-reduction efforts like reducing inference costs, routing strategies, and lower-cost/open-source models. 66% of respondents improved query cost by 10%+, and 19% by 30%+. - It also shows that the AI revenue mix for non-AI-native startups is shifting massively: 32% of total revenue in 2025, 42% in 2026, and 53% in 2027. - This is a direct update to `ai-cogs-saas-gross-margin-compression` ("AI COGS Problem: SaaS Gross Margins Compress from 80% to 52-65% Range"). The previous note highlighted the gross margin compression, and this new ICONIQ report shows how startups are fighting back and successfully expanding those margins through query cost optimization and model routing. This is a perfect update for the `ai-cogs-saas-gross-margin-compression` slug! Let's record the event for ZoomInfo first. Wait, let's verify the ZoomInfo watch details: - Title: `ZoomInfo Hybrid Consumption Pricing and Operations Business Growth Tracking` - Thread ID: `019feaef-34f4-71c2-be12-85f0016000fe` - Trigger: `Monitor ZoomInfo's Operations business growth rate (target 20%+) and details on hybrid consumption pricing rollout in upcoming earnings calls.` - Let's call `record_event` with: - `thread_id`: `019feaef-34f4-71c2-be12-85f0016000fe` - `cycle_id`: `5f413919-52b7-4829-8955-b22e393bb204` - `summary`: `ZoomInfo Q2 2026 Operations business grew 20% YoY, meeting target; consumption pricing rollout confirmed for Q3 2026.` - `outcome`: `Operations grew exactly 20% YoY, matching expectations. Consumption pricing migrations will start in Q3 2026 and run into 2027.` - `resolves`: `false` (since it's an ongoing watch tracking future quarters too) - `salience`: `4` Let's make this tool call.

Now let's check if there are other watches we can resolve or record events on. Wait! What about the Pegasystems watch? - Title: `Pegasystems Q3 2026 Earnings: AI Buyer Paralysis and ACV Recovery Tracking` - Thread ID: `019fa2d2-368c-7440-9909-d6c996c4d1d5` - Trigger: `Check if Pegasystems' ACV growth recovers and if management reports easing of AI-related customer confusion in Q3 2026 earnings.` - Since Q3 2026 earnings are not expected until early November 2026, we cannot resolve this watch yet. However, we have a major leading indicator: the ETR TSIS July 2026 survey showed a massive drop in Pegasystems' Net Score to +0.9% (from +12.4% in April). This is a critical development, so I will record an event on this watch but keep it open (`resolves: false`). Let's do that! - `thread_id`: `019fa2d2-368c-7440-9909-d6c996c4d1d5` - `cycle_id`: `5f413919-52b7-4829-8955-b22e393bb204` - `summary`: `Pegasystems ETR Net Score crashes to +0.9% in July 2026, signaling deepening buyer paralysis.` - `outcome`: `Net Score fell from +12.4% in April to +0.9% in July, the sharpest drop among enterprise peers, indicating that AI buyer confusion is worsening ahead of Q3 earnings.` - `resolves`: `false` - `salience`: `4` Let's make this tool call.

Excellent. We have recorded events on two of our watches! Now let's write our notes. First, let's write the note for **SaaS Gross Margin Compression/Expansion** using slug `ai-cogs-saas-gross-margin-compression`. Wait, let's review our instructions: - Carry 1-2 REPRESENTATIVE VERBATIM QUOTES, copied exactly, each kept with the link it came from. - Name and link the actual thing. - Say what it means. - Reuse slug `ai-cogs-saas-gross-margin-compression` to update. Let's read the previous content of `ai-cogs-saas-gross-margin-compression` if possible, but actually we can just overwrite it with an updated, comprehensive note that covers the new findings. Wait, let's look at the preview of `ai-cogs-saas-gross-margin-compression`: `AI inference costs are structurally rewriting SaaS unit economics. For every $1M in AI product revenue a SaaS company books in 2026, roughly $230,000 exits as inference cost before any personnel…` We will rewrite it to integrate the July 2026 ICONIQ State of AI report, which shows a very interesting development: while early 2026 saw massive gross margin compression, by mid-2026 companies are fighting back and successfully expanding AI gross margins (projected to rise from 45% in 2025 to 53% in 2026 and 59% in 2027) through "blocking and tackling" unit economic optimizations (reducing inference costs, routing strategies, switching to cheaper or open-source models). Let's write this note. Let's make sure we quote the SaaSletter or OnlyCFO articles properly. Let's look at the SaaSletter article: > "Looking strictly at customer-facing AI product costs, the 14% projected increase in AI Gross Margins (2025: 45% → 2027: 59%) seems questionable. ... However, the change in the unit economics survey suggests gross margin improvement is supportable: 66% of respondents improved their cost per query by 10%+, with 19% at 30%+." And: > "Key unit economics drivers: #1 reducing inference costs, #2 implementing routing strategies, #3 revenue growth through cost leverage, #4 switching to lower-cost/open-source models." Let's draft the note body for `ai-cogs-saas-gross-margin-compression`. Wait, let's review the required format: ```markdown # AI COGS Problem: SaaS Gross Margins Compress from 80% to 52-65% Range <content> ## Sources - [SaaSletter - ICONIQ "State Of AI" Takeaways](https://www.saasletter.com/p/iconiq-state-of-ai-july-2026-ai-revenue-mix) — Analysis of ICONIQ's July 2026 State of AI report and AI gross margins. - [ICONIQ Releases 2026 State of AI Report - by OnlyCFO](https://www.cfopilot.com/p/iconiq-releases-2026-state-of-ai) — Breakdown of token spend, model strategy, and gross margins from the July 2026 report. ``` Let's write the note. I will formulate a very strong, detailed, and data-driven note. Let's do it.

Now let's write the second note. This will be about the **Enterprise AI Pricing Split (Core vs. AI Add-Ons)**. Wait! Let's check the existing notes again. Is there an existing note about pricing? Yes, `outcome-based-ai-pricing-disruption-ccaas-crm-2026` ("The Seismic Shift to Consumption and Outcome-Based AI Pricing: Microsoft, Salesforce, and Startups Pivot to Mitigate Seat Compression"). We can update `outcome-based-ai-pricing-disruption-ccaas-crm-2026` with this new insight on the "Pricing Split." Let's see: the Futurum Group report published on August 17, 2026, titled "Why Core Software and AI Can't Share One Commercial Model" is a perfect fit for this slug. It explains why core software and AI cannot share a single pricing model: - Core software is shifting to consumption and outcome-based pricing (per-seat is ranked last at 12.6%). - AI functionality is shifting *back* to per-seat pricing (ranked first at 42.3%) because buyers want predictability and budget ceilings for highly volatile AI costs. - Pure outcome-based pricing is showing flat/volatile ETR Net Scores (Zendesk at -1.7%, Intercom at -11.1%), whereas hybrid models (ServiceNow +42.8%, Salesforce +19.0%) are showing stronger spending momentum. - By Q4 2026, pricing optionality will become a shortlist gate for enterprise procurement. Let's write an updated note for `outcome-based-ai-pricing-disruption-ccaas-crm-2026`. This is a beautiful way to keep findings as a living document! Let's do it.

Wonderful. Now let's write our third note. Wait, let's look at the other existing notes. We have: - `ai-buyer-paralysis-legacy-software-slowdown-2026` ("AI Buyer Paralysis: "Value Maxing" and Elongated Sales Cycles Drive Historic Software Pullbacks"). This note covers buyer paralysis and Pegasystems. Let's see if we should update it with the new ETR TSIS data showing Pegasystems' Net Score crash to +0.9% in July 2026 (from +12.4% in April 2026). This is a perfect, concrete validation of the "AI Buyer Paralysis" thesis and directly updates our existing finding! Let's draft an update to `ai-buyer-paralysis-legacy-software-slowdown-2026`. Let's do it. Let's make sure we quote the Futurum article or other sources and link them properly. Wait, let's write the note.

Excellent. I have updated three key findings with rich, highly relevant data from late July and August 2026. Let's review if there are any other threads or watches I should record or update. Wait, let's list the open threads again to make sure I didn't miss anything. - `ZoomInfo Hybrid Consumption Pricing and Operations Business Growth Tracking` (watch) -> Event recorded! - `HubSpot Q3 2026 Earnings & Analyst Day: Net Adds and Outcome Pricing Pivot Tracking` (watch) -> Expected in November, no new event yet. - `Pegasystems Q3 2026 Earnings: AI Buyer Paralysis and ACV Recovery Tracking` (watch) -> Event recorded! - `GitLab Q3 2026 Earnings: AI Infrastructure Pivot and Margin Impact` (watch) -> Expected in December, no new event yet. - `Open-Source AI Gateway Traffic Share Crossing 40% Threshold` (watch) -> No new event yet. - `NICE Q3 2026 Cloud Revenue Recovery and AI Backlog Conversion` (watch) -> Expected in November, no new event yet. - `Sierra AI ARR Growth and IPO Tracking` (watch) -> No new event yet. - `Salesforce Acquisition of Fin (Intercom) Closing in Q4 FY2027` (watch) -> Expected in January 2027, no new event yet. - `Oracle FY2027 Capex and Free Cash Flow Tracking` (watch) -> No new event yet. - `Salesforce Organic Revenue Growth Reacceleration in H2 FY2027` (watch) -> No new event yet. Wait! Are there any other threads we should open? Let's see: are there any new, forward-looking triggers we should watch for? Yes! In the Futurum Group report: - **Pricing Optionality Shortlist Gate (Tripwire):** "By the fourth quarter of 2026, Futurum expects vendors without pricing optionality... to be excluded from consideration on commercial architecture alone..." This is a perfect forward-looking tripwire / thesis breaker! We can register a watch for this: - Watch Type: `tripwire` - Title: `Enterprise RFPs Mandating Pricing Optionality by Q4 2026` - Cardinality: `one_shot` - Description: `Track if enterprise RFPs and procurement guidelines begin explicitly requiring vendors to offer multiple pricing models (such as consumption/outcome for core and predictable/seat for AI) as a gating shortlist criterion by the end of Q4 2026.` - Let's call `record_watch` to register this! Let's do that.

This is fantastic. We have: 1. Listed the open threads and read the prior findings. 2. Searched the web, found and fetched primary sources (specifically Futurum Group's August 17, 2026 report and ICONIQ Capital's July 2026 State of AI report). 3. Grounded our findings in these sources and extracted precise data, metrics, and verbatim quotes. 4. Updated three major existing findings: - `ai-cogs-saas-gross-margin-compression` with the new ICONIQ State of AI data on gross margin expansion and engineering cost-containment strategies. - `outcome-based-ai-pricing-disruption-ccaas-crm-2026` with the Futurum survey data on the "pricing split" (core software vs. AI add-ons) and ETR TSIS spend momentum. - `ai-buyer-paralysis-legacy-software-slowdown-2026` with the dramatic collapse of Pegasystems' ETR Net Score to +0.9% in July 2026. 5. Recorded events on two of our active watches (ZoomInfo's 20% Operations growth and Pegasystems' Net Score crash). 6. Registered a new forward-looking watch (a tripwire for Q4 2026 procurement RFPs requiring pricing optionality). We are now ready to call `finish_cycle` and conclude this highly successful research cycle. Let's make that call!

I have completed this research cycle on the enterprise software landscape and AI-native displacement. Here is a summary of the key findings, updates, and tracking events processed during this cycle: ### 1. Key Findings & Strategic Updates During this cycle, I integrated primary evidence from two major research releases from late July and August 2026: **The Futurum Group's 2H 2026 Enterprise Software Decision-Makers Survey** (published August 17, 2026) and **ICONIQ Capital's "State of AI: The Builder's Economy" Report** (published July 16, 2026). * **The "Pricing Split" (Core Software vs. AI Add-Ons):** * *Finding Updated:* `[[outcome-based-ai-pricing-disruption-ccaas-crm-2026]]` * *The Story:* Traditional seat-based licensing is undergoing a highly bifurcated shift. While per-seat pricing has fallen to last place (**12.6%**) for core platform software (where consumption and outcome-based models now dominate), it has actually surged to first place (**42.3%**) for generative AI add-ons billed separately. Buyers want the flexibility of usage-based models for mature core products, but demand the predictability of per-seat caps for highly volatile AI costs to prevent invoice surprises. * *ETR Spending Data:* Pure outcome-based pricing has not yet translated into stronger spending momentum. Zendesk's Net Score has run flat at **-1.7%** in the July 2026 TSIS (N=237) and Intercom's remains volatile at **-11.1%**, while hybrid incumbents like ServiceNow (**+42.8%**), Salesforce (**+19.0%**), and Adobe (**+15.5%**) maintain steady, positive spend trajectories. * **AI COGS Optimization & Gross Margin Recovery:** * *Finding Updated:* `[[ai-cogs-saas-gross-margin-compression]]` * *The Story:* While early 2026 saw severe gross margin compression as AI inference costs ate into SaaS economics, software builders are successfully fighting back. ICONIQ's survey of over 300 executives shows AI gross margins are expanding from 45% in 2025 to a projected **53% in 2026** and **59% by 2027**. * *The Mechanics:* This recovery is driven by engineering-level cost containment: **66% of respondents improved their cost per query by 10%+** (and 19% by 30%+) through model routing (directing simpler tasks to cheaper/open-source models like Llama 3 or Claude Haiku) and aggressive prompt optimization. Companies now use an average of 3.3 model providers to manage cost. * **Pegasystems and the Deepening of AI Buyer Paralysis:** * *Finding Updated:* `[[ai-buyer-paralysis-legacy-software-slowdown-2026]]` * *The Story:* The thesis of AI-induced buyer paralysis and elongated sales cycles has been validated by quantitative spending data. **Pegasystems (NYSE: PEGA)**, which has struggled with customer confusion around generative AI integration, saw its company-level ETR Net Score **collapse from +12.4% in April 2026 to just +0.9% in July 2026**—the sharpest single-quarter spending intent decline among major enterprise software peers. --- ### 2. Tracked Events & Watches * **ZoomInfo Watch Checked (Ongoing):** Recorded an event on the watch `ZoomInfo Hybrid Consumption Pricing and Operations Business Growth Tracking` (Thread ID: `019feaef-34f4-71c2-be12-85f0016000fe`). In ZoomInfo's Q2 2026 earnings (reported August 5, 2026), management confirmed that their non-seat Operations business grew by exactly **20% year-over-year**, hitting our target threshold. They also scheduled the rollout of more flexible consumption-based pricing for Q3 2026. * **Pegasystems Watch Checked (Ongoing):** Recorded an event on the watch `Pegasystems Q3 2026 Earnings: AI Buyer Paralysis and ACV Recovery Tracking` (Thread ID: `019fa2d2-368c-7440-9909-d6c996c4d1d5`) to reflect the ETR Net Score collapse to +0.9% in July 2026, signaling that buyer paralysis is worsening ahead of their Q3 earnings in November. * **New Watch Registered (Tripwire):** Registered a new forward-looking watch: `Enterprise RFPs Mandating Pricing Optionality as Shortlist Gate in Q4 2026` (Thread ID: `01a03301-96e7-70ec-82ab-b9b9bd6b0253`). This tracks Futurum's prediction that by Q4 2026, enterprise procurement teams will explicitly mandate pricing optionality (consumption/outcome for core, predictable/capped for AI) as a strict gating criterion to even enter shortlists.