HUBSPOT, INC. (HUBS)
Hold

The Leading Indicators Broke: HubSpot Pulls Its Net New ARR Commitment as Retention Slips to 102% and Net Adds Shrink

Published: By A.N. Burrows HUBS | Q2 2026 Earnings Analysis

Key Takeaways

  • Q2 beat the guide again, by the smallest margin in the six quarters we track. Revenue of $911.7M (+20% reported, +17.5% constant currency) cleared the top of the guide by $13.7M and consensus by 1.5%; the non-GAAP operating margin of 20.3% beat the 19% guide, and non-GAAP EPS of $3.26 beat $3.02.
  • Every leading indicator moved the wrong way. Net customer additions fell to 6,988 against a 9,000–10,000 guide and are now guided to 5,000–6,000. Net revenue retention fell to 102% with the full-year view cut to flat. Management now expects 2026 net new ARR to grow slower than constant-currency revenue, the reverse of the expectation it kept in May, and guides Q3 to 15% constant-currency growth.
  • The slowdown is no longer only self-inflicted. Management added a second headwind: tighter customer budgets, larger buying committees, board-level approvals and downgrade pressure at renewal. It said the pressure continued into July, assumed it lasts all year, and could not say when it ends.
  • Margins and buybacks are now the floor, not the story. The 21% margin guide held on lower revenue, management expects 2–3 more points in 2027, and HubSpot bought back $531.9M of stock at $202.80 and added a $1B authorization. The stock fell 19.1% to $202.43, 15.3x the raised EPS guide.
  • Rating: Downgrading to Hold from Outperform; base-case target cut to $216 from $254. Three of the four Hold triggers we set in May fired: NRR below 103%, the net new ARR commitment withdrawn, and a lower constant-currency guide. Valuation and capital return keep us from going further.

Results vs. Consensus

Q2 2026 Scorecard

MetricQ2 2026 ActualConsensusCompany Guide (May 7)Beat/MissMagnitude
Total revenue$911.7M~$898M$897.0–898.0MBeat+$13.4M (+1.5%) vs. consensus; +$13.7M vs. top of guide
Revenue growth, constant currency+17% (17.5%)n/a+16%Beat+1 to 1.5 pts vs. guide
Revenue growth, as reported+20%n/a+18%Beat+2 pts vs. guide
Non-GAAP operating income$185.3Mn/a$173.0–174.0MBeat+$11.3M (+6.5%) vs. top of guide
Non-GAAP operating margin20.3%n/a19% (19.3% at midpoint)BeatAbout +100bp vs. guide midpoint
Non-GAAP EPS (diluted)$3.26$3.02$3.00–3.02Beat+$0.24 (+7.9%)
GAAP EPS (diluted)$0.86n/an/an/avs. ($0.06) in Q2 2025
Non-GAAP free cash flow$167.9Mn/an/an/a18.4% of revenue
Net customer additions6,988n/a9,000–10,000MissAbout 2,000 below the bottom of the range
Net revenue retention102%n/a+1 to 2 pts for FY 2026Miss(1 pt) YoY and QoQ
Q3 2026 revenue guide (midpoint)$924.5M~$942Mn/aMiss(1.8%) to (1.9%)
Q3 2026 non-GAAP EPS guide (midpoint)$3.26$3.45n/aMiss(5.5%)
FY 2026 revenue guide (midpoint)$3,682M~$3,710M$3,700–3,708MMissAbout (0.7%) vs. consensus; cut $22.0M
FY 2026 non-GAAP EPS guide (midpoint)$13.27$13.10$13.04–13.12Beat+1.3% vs. consensus (share count)
Quality-of-beat headline: The reported quarter was clean; the forward picture was not. Revenue beat the top of the guide by $13.7M and constant-currency growth beat the guided rate by one to one and a half points, but both are the smallest beats in the six quarters we track (the prior five ranged from $15.1M to $22.5M and two to three points). Currency added about three points to reported growth, down from five in Q1. The operating margin beat the guide midpoint by about 100bp. The 330bp YoY margin expansion came entirely from operating expenses: about 485bp of leverage, mostly from a flat research and development budget, less about 150bp of non-GAAP gross margin as AI and hosting costs rose. The full-year EPS raise came from a lower share count; at the May share count the new net income guide would have produced EPS of about $12.81, 2.1% below the May midpoint.

Year-Over-Year Comparisons

MetricQ2 2026Q2 2025YoY Change
Total revenue$911.7M$760.9M+19.8% (+17.5% CC)
Subscription revenue$894.0M$744.5M+20.1%
Professional services and other$17.7M$16.3M+8.5%
Gross profit$750.9M$638.7M+17.6%
GAAP gross margin82.4%83.9%(150bp)
Non-GAAP subscription gross margin85.3%87.1%(180bp)
GAAP operating income (loss)$43.3M($24.6M)n/m (swing to profit)
Non-GAAP operating income$185.3M$129.1M+43.5%
Non-GAAP operating margin20.3%17.0%+330bp
GAAP net income (loss)$43.3M($3.3M)n/m
Non-GAAP diluted EPS (as reported)$3.26$2.19+48.9%
Non-GAAP diluted EPS (Q2 2025 at the 15% tax rate)$3.26$2.33+39.9%
Non-GAAP free cash flow$167.9M$116.2M+44.4%
Stock-based compensation (% of revenue)14.1%18.5%(440bp)

Quarter-Over-Quarter Comparisons

MetricQ2 2026Q1 2026QoQ Change
Total revenue$911.7M$881.0M+3.5%
Subscription revenue$894.0M$862.3M+3.7%
Professional services and other$17.7M$18.7M(5.4%)
GAAP gross margin82.4%83.5%(110bp)
Non-GAAP subscription gross margin85.3%86.6%(130bp)
Non-GAAP operating income$185.3M$156.8M+18.2%
Non-GAAP operating margin20.3%17.8%+250bp
Non-GAAP diluted EPS (both at 15% tax)$3.26$2.72+19.9%
Non-GAAP free cash flow$167.9M$153.7M+9.2%
Customers306,446299,458+6,988
Average subscription revenue per customer$11,800$11,722+0.7%
Net revenue retention102%103%(1 pt)

Five-Quarter Trend

QuarterRevenueGrowth (reported / CC)Non-GAAP Op. MarginNet AddsASRPCBillings Growth (reported / CC)NRR
Q2 2025$760.9M19% / 18%17.0%9,724$11,31026% / 20%103%
Q3 2025$809.5M21% / 18%19.9%10,898$11,57818% / 19%103%
Q4 2025$846.7M20% / 18%22.6%9,826$11,68327% / 20%105%
Q1 2026$881.0M23% / 18%17.8%10,752$11,72219% / 17%103%
Q2 2026$911.7M20% / 17%20.3%6,988$11,80014% / 17%102%

The table shows a business whose income statement is still improving and whose customer metrics have turned. Constant-currency revenue growth slipped to 17%, the first quarter below 18% in the six quarters we track. Net adds, NRR and reported billings growth are each the lowest of the five quarters shown, and constant-currency billings growth matched Q1's 17%, the low of the series. The operating margin rose 330bp YoY on the same seasonal pattern as 2025. In May the guide was where the change happened; this quarter the change is in the reported KPIs as well.

Quality of Beat

Revenue: The $13.4M beat against consensus and $13.7M beat against the top of the guide are organic, but they are smaller than any beat in the prior five quarters. Constant-currency growth of 17.5% (17% in the release) was one to one and a half points above the 16% guide, against two to three points in every earlier quarter we track. Currency added about three points to reported growth, down from about five in Q1, so the reported rate overstates the business by less than it did. Subscription revenue grew 20.1% and services 8.5%. The 10-Q again attributes subscription growth mainly to customer count (+14.4%, led by Starter products) and to higher average subscription revenue per customer from Professional and Enterprise demand plus currency.

"Q2 revenue grew 17.5% year-over-year in constant currency. We delivered 3 points of non-GAAP operating margin expansion year-over-year, bringing our operating margin to 20.3%."
— Yamini Rangan, CEO

Margins: The non-GAAP operating margin of 20.3% expanded 330bp YoY, and operating expenses did all the work. Non-GAAP research and development spending grew 1.0% in dollars while revenue grew 19.8%, taking it to 17.6% of revenue from 20.9%; general and administrative fell to 6.7% from 7.8%; sales and marketing fell to 39.4% from 39.8%. Non-GAAP total gross margin fell about 150bp. Headcount was flat sequentially at 9,016. The 10-Q shows subscription cost of revenue up 36.8% against subscription revenue up 20.1%, with hosting and capitalized-software amortization making up about 99% of the increase.

"This expansion reflects our continued disciplined approach to head count spend, partially offset by AI costs."
— Kathryn Bueker, CFO

EPS: Non-GAAP EPS of $3.26 rose 48.9% as reported, but the comparison crosses the change in HubSpot's non-GAAP tax rate from 20% to 15%. On the company's 15% recast, Q2 2025 EPS was $2.33 and growth was 39.9%. Operating income growth of 43.5% drove it; interest income fell 49% to $9.4M as cash went to the buyback, and diluted shares fell to 50.6M from 53.5M. GAAP EPS of $0.86 compares with a loss of ($0.06), helped by stock-based compensation falling to 14.1% of revenue from 18.5%. The 10-Q repeats that a release of the tax valuation allowance is reasonably possible within 12 months.

"GAAP operating margin was 5% in Q2 compared to a negative operating margin of 3% in the year ago period. This 8 points of expansion reflects our non-GAAP operating income expansion and a more than 4-point reduction in stock-based compensation expense as a percentage of revenue."
— Kathryn Bueker, CFO

Segment Performance

HubSpot reports one operating segment. As in our earlier reports, we use revenue by type, revenue by geography and the customer KPIs as the segment lens.

Revenue by Type: Q2 2026

Revenue TypeQ2 2026Q2 2025YoY (reported)% of RevenueGross Margin Q2 2026Gross Margin Q2 2025
Subscription$894.0M$744.5M+20.1%98.1%83.7% GAAP / 85.3% non-GAAP85.7% GAAP / 87.1% non-GAAP
Professional services and other$17.7M$16.3M+8.5%1.9%15.8% GAAP5.2% GAAP
Total$911.7M$760.9M+19.8%100%82.4% GAAP83.9% GAAP

Subscription: 20% Growth, 180bp of Non-GAAP Margin Given Back

Subscription revenue of $894.0M grew 20.1% as reported and 3.7% sequentially. The non-GAAP subscription gross margin fell to 85.3%, 180bp lower than a year ago and 130bp lower than in Q1, and below the 86% bar we set at initiation for the first time. Subscription cost of revenue rose 36.8%: hosting costs rose $25.8M with customer growth, platform usage and AI functionality, and capitalized-software amortization rose $13.1M. The 10-Q describes total gross margins as relatively consistent; the subscription line, at 200bp lower on a GAAP basis, is the one that moved.

"Q2 subscription revenue grew 20% year-over-year, while services and other revenue increased by 8%, both on an as-reported basis."
— Kathryn Bueker, CFO

Assessment: The revenue line is still healthy; the margin under it is not holding. In May we warned that cutting the price per resolved conversation while credits grow would narrow the gap between AI revenue and AI cost, and this quarter shows it: more usage at lower prices with free trials. At 85.3% the margin sits about 30bp above the 85% level we set as an Underperform trigger, so the Q3 print matters.

Professional Services and Other: Payments, Again

Services and other revenue of $17.7M grew 8.5% YoY and fell 5.4% sequentially; the 10-Q again names Payments as the driver. The GAAP gross margin improved to 15.8% from 5.2% as HubSpot used Solutions Partners for onboarding and cut its own services staff costs.

Assessment: At 1.9% of revenue, this line has no bearing on the thesis.

Revenue by Geography

Geography (Q2 2026)YoY Growth (reported)YoY Growth (CC)Share of Q2 RevenueComment
Domestic (U.S.)+17%n/a (USD)~51% (by difference)Down from 18% in Q1
International+23%+18%49%5-pt currency tailwind (11 pts in Q1)
Total+20%+17%100%Currency added about 3 pts
Region (by recording operation)Q2 2026Q2 2025YoY (reported)Share Q2 2026Share Q2 2025
Americas$533.7M$457.3M+16.7%58.5%60.1%
Europe$301.1M$243.9M+23.4%33.0%32.1%
Asia Pacific$77.0M$59.6M+29.2%8.4%7.8%
Total$911.7M$760.9M+19.8%100%100%

Domestic: 17%, a Point Slower Than Q1

U.S. revenue grew 17% in Q2, down from 18% in Q1, with no currency effect. On the 10-Q's regional cut, the Americas grew 16.7%. In May we called the domestic rate the one to watch to separate a sales-process problem from a demand problem; it moved down a point in the first quarter of our coverage in which management described a weaker demand environment.

"Domestic revenue grew 17% year-over-year in Q2. International revenue growth was 23% as reported and 18% in constant currency, representing 49% of total revenue."
— Kathryn Bueker, CFO

Assessment: One point is small, but it reverses Q1's one-point acceleration and it lines up with the call: budget scrutiny, larger buying committees and longer approvals. The domestic business no longer shows the strength that let us call Q1's problem self-inflicted.

International: 18% in Constant Currency, 23% Reported

International revenue grew 18% in constant currency and 23% as reported, so the currency tailwind narrowed to five points from eleven in Q1. Europe grew 23.4% and Asia Pacific 29.2% on the reported regional cut. Revenue from customers outside the United States was 49% of the total, up from 48% a year ago.

Assessment: International grew a point faster than domestic in constant currency after the two grew at the same rate in Q1, so the slowdown is centered in the U.S. The fading currency benefit also explains part of the guide: Q3 is guided to 14% reported and 15% constant currency, the first quarter in our coverage in which the guided reported rate sits below the constant-currency rate.

Key KPIs

KPIQ2 2026Q1 2026Q2 2025YoYTrendvs. Expectation
Customers306,446299,458267,982+14.4%6,988 net addsAbout 2,000 below the 9,000–10,000 range
Avg. subscription revenue per customer$11,800$11,722$11,310+4% reported / +2% CC+0.7% QoQLow end of the CC guide; no ramp
Net revenue retention102%103%103%(1 pt)Downgrade pressure at renewalBelow our 104% bar and our 103% review line
Customer dollar retentionHigh 80sHigh 80sn/an/aStablen/a
Calculated billings$929.7M$912.3M$814.3M+14% reported / +17% CCCC growth flat with Q1n/a
Total deferred revenue$1,062.7M$1,044.7Mn/an/a+1.7% QoQn/a
RPO (contracts over one year)~$1.7B~$1.7Bn/an/a~89% recognized within 24 monthsn/a
Net new ARR growthFY 2026 now below CC revenue growthA bit below CC revenue growthn/an/aFull-year commitment reversedCommitment withdrawn
Agent activationsData 16,000+; Prospecting ~17,000; Customer 10,000+Data 9,000+; Customer 9,000+n/an/aData +80% QoQ; Prospecting +28% QoQ (company rates)Adoption rising
Credits consumed"Grew"+67% QoQn/an/aNo growth rate givenNo dollar figure, again

Customers and Net Adds: The Funnel Is Now the Problem

HubSpot added 6,988 net customers, ending at 306,446 (+14.4%). That is about 2,000 below the bottom of the 9,000–10,000 range the CFO reaffirmed in May, the lowest figure in our five-quarter table and 35% below Q1's 10,752. The CFO blamed weaker conversion and buyer hesitancy, and guided the next two quarters to 5,000–6,000. At that pace, customer growth falls to about 10% YoY by December from 16% in Q1.

"We added 7,000 net new customers in Q2, bringing our total customer count to over 306,000 growing 14% year-over-year. This was below our 9,000 to 10,000 expectation and driven primarily by weaker conversion rates and increased buyer hesitancy."
— Kathryn Bueker, CFO

Assessment: In May we wrote that customer count was not where the problem sat. It is now. Customer count has driven most of HubSpot's subscription growth, and net adds guided to about half of Q1's level with ASRPC rising only 2% in constant currency point to low-teens revenue growth in 2027 unless the trials convert into paid seats and credits.

ASRPC: 2% in Constant Currency, and the Ramp Language Is Gone

Average subscription revenue per customer was $11,800, up 4% as reported and 2% in constant currency, and up 0.7% sequentially. In May management said ASRPC growth would ramp through 2026; the new guide keeps the low-to-mid-single-digit range but drops the ramp.

"Average subscription revenue per customer was $11,800 in Q2, up 4 points year-over-year as reported and 2 points in constant currency."
— Kathryn Bueker, CFO

Assessment: Flat at 2% for two quarters, ASRPC is where seats and credits were supposed to show and have not. It neither reached our bullish 3% nor fell to our Underperform trigger (flat or down). With lower agent prices, free trials and customers swapping seats for credits, we see no driver for a ramp in the second half.

Net Revenue Retention: 102%, Below Our Review Line

Net revenue retention fell to 102%, down a point both sequentially and YoY. The CFO said seat and credit expansion still helped, but the April price cuts and trials slowed credit expansion and customers renewing at lower spend created downgrade pressure. Customer dollar retention held in the high 80s, so customers are spending less rather than leaving. The full-year view moved to roughly flat from one to two points of improvement.

"Net revenue retention does tend to be higher in Q3 and Q4 with UNBOUND and Q4 is obviously a big quarter for us, both from a new business perspective, from a renewal perspective and also from an upmarket perspective."
— Kathryn Bueker, CFO

Assessment: At initiation we called NRR the thesis-break metric, with 103% as the floor and 104% as the level at which we would add. Q2 printed 102% without a seasonal excuse, and the guide that underpinned our pillar is gone. The stable dollar retention keeps this a downgrade to Hold rather than a broken franchise.

Billings, Deferred Revenue and RPO: Holding at 17% in Constant Currency

Calculated billings of $929.7M grew 14% as reported and 17% in constant currency, the same constant-currency rate as Q1 and the low of our five-quarter series. Total deferred revenue rose 1.7% sequentially to $1,062.7M. Remaining performance obligations on contracts longer than one year held at about $1.7B, with about 89% converting within 24 months.

"Q2 calculated billings were $930 million, growing 14% year-over-year as reported and 17% in constant currency."
— Kathryn Bueker, CFO

Assessment: Constant-currency billings did not weaken further, which is the one forward series that held. Reported billings growth of 14%, three points below the constant-currency rate, shows that currency, a tailwind to revenue, was a headwind to billings this quarter. Flat long-dated RPO fits a quarter in which large deals slipped by weeks.

Key Topics & Management Commentary

Overall Management Tone: Management was candid about the miss and could not give a timeline for recovery. It opened by saying the quarter it expected did not fully materialize, then spent most of the call on adoption metrics and margins while declining to say when budgets would loosen. Neither the February nor the May call cited budget pressure or the demand environment; this call returned to them in answer after answer, which makes it the most defensive of the three calls we have covered.

1. Two Headwinds, Not One: The Demand Environment Joins the Reset

In May the slowdown had one cause, which management chose: the April pricing, trial and sales-training changes. In August it had two. The CEO repeated the first and added a second that management does not control: buyers are scrutinizing budgets, buying committees are larger, more deals need C-suite and board sign-off, and existing customers are cutting spend because AI costs elsewhere are squeezing their budgets.

"Let me be direct about what happened. April got off to a slow start and the quarter we expected did not fully materialize. Two factors drove the headwinds. The first was deliberate."
— Yamini Rangan, CEO
"The second factor was a shift in the demand environment in Q2 with increased budget sensitivity. Businesses want greater confidence that their investments will position them for the AI platform shift. As a result, purchase decisions are facing greater scrutiny, buying committees are larger and more deals require C-suite and Board approval, leading to longer sales cycles. For existing customers, unpredictable AI costs across the broader landscape are impacting budgets, leading to budget optimization and downgrade pressure."
— Yamini Rangan, CEO

The CFO said the pressure was steady through the quarter and continued into July, and the guide assumes it lasts through December.

Assessment: This is the change that breaks our May framing. We treated the April reset as a self-inflicted speed bump with a two-quarter window; a demand headwind that management cannot time is a different risk. It also blurs the read on the April changes, because investors can no longer tell how much of the slowdown is the reset and how much is the market.

2. Net New ARR: The Full-Year Commitment Is Withdrawn

In February management said it expected 2026 net new ARR to grow faster than constant-currency revenue. In May, after Q1 came in a bit below revenue growth, the CFO kept that expectation. In August she withdrew it: net new ARR growth will now be below constant-currency revenue growth for the year.

"We now expect net new ARR growth to be below constant currency revenue growth for the fiscal year of 2026."
— Kathryn Bueker, CFO
"Now, the factors that I talked about resulted in slower customer acquisition and net new ARR headwinds, and that is what we are contemplating in the guidance for the rest of the year..."
— Yamini Rangan, CEO

Assessment: Net new ARR running ahead of revenue was the reason we expected growth to re-accelerate. With it running behind, 2027 revenue growth starts below 2026's 16%. This is one of the four Hold triggers we set in May, and it fired.

3. NRR at 102% and a Flat Guide

Net revenue retention of 102% was down a point YoY, and the full-year guide fell to roughly flat from one to two points of expansion. The CFO broke the number into its parts: dollar retention steady in the high 80s, seats and credits still adding, credit expansion slowed by the April price cuts and trials, and other upgrades offset by customers who renewed at lower spend.

"Customer dollar retention remained healthy in the high 80s, while net revenue retention was 102%, down 1 point year-over-year as continued seat and credit expansion was offset by other net upgrade headwinds as a result of customer budget optimization."
— Kathryn Bueker, CFO
"For the full year of 2026, we expect net revenue retention to be roughly flat year-over-year."
— Kathryn Bueker, CFO

Assessment: Our May downgrade review was set at NRR below 103%, and Q2 crossed it. The pressure is on expansion, not on churn, which is why we move to Hold rather than further. A guide that leaned on credits in May now concedes that credits will not offset budget cuts this year.

4. Net Adds Cut to 5,000–6,000 a Quarter

The new net-add guide is about 40% below the 9,000–10,000 range management gave in February and reaffirmed in May. Management attributed the shortfall to weaker conversion and hesitancy, including in the down-market segment where HubSpot's Starter funnel sits.

"Our Q3 and full year guidance reflects our expectation that the headwinds we saw in Q2 will persist throughout the remainder of the year. We now expect quarterly net additions to be approximately 5,000 to 6,000, and ASRPC growth to be in the low to mid-single digits in constant currency."
— Kathryn Bueker, CFO

Assessment: This is the most mechanical damage in the guide. At 5,000–6,000 a quarter, the customer base grows about 10% YoY by year-end, and with ASRPC at 2% constant currency the arithmetic points to low-teens growth in 2027. The free-trial motion should help conversion over time, but the April changes were meant to lift adoption and net adds fell by a third from Q1.

5. Longer Deal Cycles: Upmarket Still Wins, Down Market Hesitates

Asked to separate the segments, the CEO said the two ends of the market are slowing for different reasons: down-market buyers are more cautious, while upmarket deals are still being won but take longer because more people approve them. Deals over $120,000 of annual recurring revenue grew 38% YoY, down from 64% in Q1, and 64% of new Pro+ customers landed with more than one hub.

"In down market, we just saw a more cautious buying environment. And that is leading to higher scrutiny when it comes to making the decisions."
— Yamini Rangan, CEO
"Upmarket momentum continued to be strong, deals over $120,000 ARR grew 38% year-over-year, reflecting continued demand from larger customers for a unified AI-powered customer platform. And multi-hub momentum continue. 64% of new Pro+ customers landed with multiple hubs, up 3 points year-over-year."
— Yamini Rangan, CEO

Assessment: Our second bull pillar holds, but with less margin. Large deals still grow fast and still close, a few weeks later than before, and multi-hub attach is steady. The deceleration from 64% to 38% in large-deal growth is the first sign in our coverage that the upmarket engine is not immune to the budget scrutiny.

6. AI Adoption Is Broadening; Credits Still Have No Dollar Figure

The adoption data kept improving. Data Agent passed 16,000 activated customers (+80% QoQ), Prospecting Agent almost 17,000 (+28%) and Customer Agent over 10,000, with a 72% resolution rate. More than 55% of Pro+ customers use agents or Breeze Assistant, and agent adoption among Pro+ customers rose from high single digits to mid-teens this year. Credit consumption grew, but this time management gave no growth rate, after citing 67% sequential growth in May.

"On AI, customer adoption is accelerating. Data agent has over 16,000 customers activated, up 80% quarter-over-quarter. Prospecting agent has almost 17,000 customers activated, up 28%, and customer agent reached over 10,000 customers."
— Yamini Rangan, CEO
"Finally, growth. Is AI driving growth of our business? Total credit consumption grew in Q2 despite the pricing changes we made in April. Credit usage is now evenly distributed across data agent, prospecting agent, customer agent and buyer intent."
— Yamini Rangan, CEO

Assessment: Usage is real and broader than in May, when Customer Agent alone was over half of credits. But the measure investors need, AI revenue, is still missing, and the one usage number HubSpot gave last quarter was replaced by a verb. Our Bear-2 point stays at its highest level: the AI transition is costing bookings and has not yet shown up in dollars.

7. Seats for Credits: Budget Optimization Inside the Base

One answer described a pattern that matters for the seat-compression debate. Some customers adopt AI by swapping seats for credits to stay within budget, and HubSpot is letting them. In May the CEO cited active core-seat users up 90% YoY; this call did not mention core seats at all.

"The other pattern that we see is having flexibility to be able to swap seats for credits and managing their budget, wanting to drive AI adoption, but managing the budget across seats and credits."
— Yamini Rangan, CEO

Assessment: This is the first time in the three calls we have covered that management has described seats being traded for AI usage inside the base. It is not disintermediation (customers stay on HubSpot and dollar retention is stable), but it is the mechanism by which agents could cap seat revenue. Our first bear point stays emerging, with its first evidence in the data.

8. Subscription Gross Margin: 85.3%, and AI Costs Arrive

The non-GAAP subscription gross margin fell 180bp YoY to 85.3%, the largest YoY decline in the three quarters we have covered (Q4: 140bp; Q1: 50bp). The call did not discuss gross margin directly. The CFO's only reference was that AI costs partly offset operating-margin gains, and the 10-Q shows research and development software spending up $8.0M on AI tools as well as the hosting increase in cost of revenue. The pricing side came from the CEO: HubSpot cut entry prices and tied agent prices to outcomes because customers want predictable AI costs.

"Businesses have been hit with unpredictable token costs. And they want pricing that is transparent and tied to value. In response, we introduced outcome-based pricing for several of our HubSpot agents, lowered entry price points and are providing customers clear visibility and control over usage and spend, including the ability to set thresholds that fit their budget."
— Yamini Rangan, CEO

Assessment: Free cash flow was still 18.4% of revenue, but the gross-margin trend is moving faster than HubSpot's own guidance language. Lower agent prices, free trials and rising inference volume all push the same way. We cut our 86% floor to a watch level of 85.5% for Q3 and keep 85% as the line at which we would consider an Underperform rating.

9. Margins Keep Rising: Two Points This Year, Two to Three in 2027

Management held the full-year operating-income guide in dollars despite cutting revenue, which lifts the implied margin, and gave a first 2027 marker: two to three more points of non-GAAP operating margin. The CEO credited a reorganization into smaller teams working in six-week sprints and AI use across the company. Stock-based compensation fell to 14.1% of revenue from 18.5%.

"Our guidance contemplates 2 points of non-GAAP operating margin leverage this year. Looking ahead, we expect to deliver 2 to 3 points of incremental non-GAAP operating margin leverage in 2027. In addition, we remain committed to driving stock-based compensation as a percentage of revenue down year-over-year."
— Kathryn Bueker, CFO
"We have reorganized into smaller, more focused teams operating in 6 weeks sprints."
— Yamini Rangan, CEO

Assessment: Our third bull pillar strengthened again, and it is now the main support for the stock. The concern we flagged in May has become explicit: margin commitments are rising as growth commitments fall. A margin near 23% in 2027 on low-teens growth is a good business, but it deserves a different multiple than the one we underwrote at initiation.

10. $531.9M of Buybacks and Another $1 Billion

HubSpot bought back 2,622,496 shares in Q2 for $531.9M at an average of $202.80, more than half of its February $1B authorization in a single quarter, with the heaviest buying in May at $196.68. The board added a new $1B program on August 3. With $257.1M left on the February program, capacity is about $1.26B, roughly 12% of the shares outstanding at the reaction close.

"During the quarter, we bought back more than $500 million of stock under our current $1 billion share repurchase program. Our Board of Directors has authorized an additional share repurchase program of up to $1 billion, reflecting the confidence we have in our business and the growth opportunity ahead."
— Kathryn Bueker, CFO

Assessment: The buyback cleared our $150M bar more than three times over, and shares outstanding fell 2.6% between May 1 and July 31. It is also why the EPS guide rose while net income guidance fell. Cash and investments dropped to $1.38B from $1.78B, and the Q2 pace was about three times quarterly free cash flow, so it cannot be sustained from cash flow alone. We treat the buyback as a floor under the stock, not a reason to own it.

11. New Products and a New Director: Agent Builder, Revenue Hub and an AI Board Hire

Product cadence stayed high. HubSpot launched AEO as a stand-alone product and inside Marketing Hub (activated by 32% of Marketing Hub Pro+ customers, with nearly 16,000 stand-alone trials), Revenue Hub for quoting, billing and payments, and in July Agent Builder and Agent Hub for custom and partner agents. It also closed the purchase of Futurepedia, an AI media platform, for $19.1M. On August 5 the board added a former Google executive whose most recent Google role was leading the development of AI agents for customer service and sales.

"In July, we launched Agent Builder and Agent Hub. Agent Builder lets anyone build custom agents, agentic workflows or mix and match both."
— Yamini Rangan, CEO

The quarter also included a governance misstep that the call did not address. On July 1 HubSpot changed its terms of service to share customers' enrichment data across accounts by default, then reversed the change within days after customer objections.

Assessment: The product strategy is coherent and the board hire fits it. None of these launches is sized, and the reversed data-sharing change shows how sensitive the customer base is to HubSpot monetizing its data layer, which matters for the paid intelligence layer management described in May.

Guidance & Outlook

MetricPrior FY 2026 Guide (May 7)New FY 2026 Guide (Aug 5)Change at MidpointClassificationQ3 2026 Guide
Revenue$3,700–3,708M$3,678–3,686M($22.0M), or (0.6%)Lowered$924.0–925.0M
Growth, as reported+18%+18%Unchanged (rounded)Maintained+14%
Growth, constant currency+17% (16.6%)+16%LowerLowered+15%
Non-GAAP operating income$762.0–766.0M$762.0–766.0MUnchangedMaintained$187.0–188.0M
Non-GAAP operating margin21%21%Unchanged (rounded)Maintained20%
Non-GAAP diluted EPS$13.04–13.12$13.23–13.31+$0.19 (+1.5%)Raised$3.25–3.27
Diluted shares~51.8M~50.0M(1.8M)Lowered (buyback)~49.3M
Free cash flow~$750M~$750MUnchangedMaintainedn/a
Capex (% of revenue)5–6%5–6%UnchangedMaintainedn/a
SBC (% of revenue, from the reconciliation)~14.2%~13.9%About (0.3) ptLowered (favorable)n/a
Net revenue retention+1 to 2 ptsRoughly flat(1 to 2) ptsLoweredn/a
Net customer additions9,000–10,000 per quarter~5,000–6,000 per quarterAbout (4,000)Lowered~5,000–6,000
ASRPC (CC)Low-to-mid single digits, rampingLow-to-mid single digitsRamp droppedSoftenedn/a
Net new ARR vs. CC revenue growthAbove, for FY 2026Below, for FY 2026ReversedWithdrawnn/a
2027 non-GAAP operating marginAnalyst Day+2 to 3 ptsNewNewn/a

Read line by line, this is a cut. Revenue, constant-currency growth, net revenue retention, net adds and the net new ARR expectation all moved down; operating income, free cash flow and capex held; EPS rose only because the share count fell 3.5%. At the May share count, the new net income guide would have produced EPS of about $12.81, 2.1% below the May midpoint. The CFO said the guide assumes the second-quarter headwinds continue for the rest of the year, and she did not split the change into currency and organic parts as she did in May.

"And for the full year of 2026, total-as-reported revenue is now expected to be in the range of $3.678 billion to $3.686 billion, up 18% year-over-year on an as-reported basis and 16% in constant currency. We continue to expect non-GAAP operating income to be in the range of $762 million to $766 million, representing a 21% margin."
— Kathryn Bueker, CFO
"We continue to expect CapEx as a percentage of revenue to be 5% to 6% for the full year of 2026 and continue to expect free cash flow to be about $750 million."
— Kathryn Bueker, CFO

Implied Q-over-Q ramp: The Q3 midpoint of $924.5M is up 14.2% as reported on Q3 2025's $809.5M and only $12.8M (1.4%) above Q2's $911.7M; a year ago the Q2-to-Q3 add was $48.7M. The full-year midpoint of $3,682M leaves $1,889.3M for the second half, up 14.1% as reported on 2025's second half and $36.2M below what the May guide implied. That implies Q4 revenue of about $964.8M (+13.9% reported) and Q4 non-GAAP operating income of about $234.4M, a 24.3% margin. Constant-currency growth goes from 18% in Q1 to 17.5% in Q2 and a guided 15% in Q3.

Street at: The Q3 revenue guide midpoint sits 1.8–1.9% below the ~$942M consensus, and even the top of the range is 1.8% below it. The Q3 EPS guide midpoint of $3.26 is 5.5% below the $3.45 consensus. The full-year revenue midpoint is about 0.7% below the ~$3.71B consensus, while the full-year EPS midpoint is 1.3% above $13.10. In May the next-quarter revenue guide sat 0.1–0.5% below the Street, the next-quarter EPS guide sat above it, and we found no usable full-year revenue consensus. This time the next-quarter revenue gap is about 1.9%, the next-quarter EPS guide is also below the Street, and the full-year revenue guide sits below a consensus confirmed by two independent providers.

Guidance style: Still conservative by construction, now less so in practice. HubSpot has beaten the top of its revenue guide in each of the six quarters we track, but the margin of the beat shrank to $13.7M from $15.1–22.5M, and the guide now explicitly assumes no recovery. We assume a Q3 beat of about $9M against the top of the range, smaller than Q2's.

Analyst Q&A Highlights

Were There More Disruptive Choices, and Are More Coming?

The opening question asked whether anything beyond the known April changes had disrupted the quarter, and whether HubSpot needed to make more aggressive moves. The CEO walked through the trials and pricing changes again and defended them as the right trade, but did not say whether more changes were coming. The question also referred to the July terms-of-service change and its reversal; the answer did not address it.

Q: "I guess the question that I have just processing Kate's guidance and all the information that you gave, were there additional choices in the quarter that disrupted the results and influenced the outlook beyond what we've already talked about. And then maybe looking ahead, are there more aggressive choices the company needs to make in this fast-moving AI backdrop."
— Samad Samana, Jefferies

A: "Now these decisions create a near-term headwind, and the decision we made is we'd rather remove friction and build customer confidence at the beginning of the AI journey and then help them drive much more adoption as they continue. ... Those are all the deliberate changes that we made, and we feel very good because, yes, we take half a step back, but we are doing that so that we can take multiple steps forward."
— Yamini Rangan, CEO

Assessment: "Half a step back" understates a quarter in which net adds fell by a third and the full-year revenue guide was cut. The answer defended the April changes but did not answer the second half of the question, so investors cannot rule out further pricing or go-to-market changes.

Is Token Spending Crowding Out HubSpot?

A question asked whether customers' spending on AI tokens is eating into software budgets, and whether newer AI-native vendors are confusing buyers. The CEO described a "sorting phase" in which customers will shift from spending on tokens to paying for outcomes, and gave the clearest picture of the new buying process: decisions that a marketing vice president used to approve now go to the board.

Q: "So my question for you is, has that kind of token burn impacted budgets and how companies both new and existing, are thinking about deploying HubSpot?"
— Rishi Jaluria, RBC

A: "Look, in Q2, we saw a more cautious spending environment. And look, there will be a sorting phase, I'm sure. ... For new customers, we're seeing larger buying committees and more C-suite involvement. Many times, boards and PE firms are involved in the final approval. And that's not what we saw a few quarters ago where things that would get like approved at the VP of Marketing level now requires a Board approval."
— Yamini Rangan, CEO

Assessment: The answer confirms that AI budgets elsewhere are pressuring HubSpot's spend, which is a version of the bear case we have tracked since initiation. Board-level approval for a mid-market CRM purchase is a meaningful change in the sales motion, and it is not one HubSpot can reverse with pricing alone.

When Does It Get Better?

A question asked whether the current caution is a phase of exploration that will pass or a new baseline, drawing the comparison with the years after COVID when software buyers never returned to prior behavior. The CEO described the transition in general terms and then stated the guidance assumption directly.

Q: "Do you envision getting to a more steady state environment? Or is this like the new normal, like what we were talking about post-COVID, where we kept waiting for things to get better and it just turned out this is the new baseline."
— Gabriela Borges, Goldman Sachs

A: "Now we are going to assume that what we saw in Q2 is what we will see in the second half, and that is what is contemplated within the guidance. But we're doing everything from our response perspective to provide the comfort for our customers as they navigate this transition."
— Yamini Rangan, CEO

Assessment: A guide that assumes no improvement is easier to beat, and that is the bull argument for the second half. But management did not dispute the "new baseline" framing, and we now model it.

Where Is the Budget Going Instead?

A follow-up asked what customers are spending on instead of HubSpot and how long the scrutiny will last. The CEO pointed to heavy spending on AI tokens in the first half and said she could not name a month or quarter for a recovery.

Q: "I'm just kind of struggling to think, okay, customers are scrutinizing their budget more, what is it that they are opting to spend their money on instead of HubSpot at the moment? And what gives you the confidence that this is -- it's something temporary, and how long does that temporary last?"
— Jackson Ader, KeyBanc

A: "Look, I wish I had like a very clear crystal ball to give you exact answers there. But what happened, as you all know, in the first half of this year is that there's just a lot of token maxing. ... I can't tell you an exact month or a quarter where that happens, but we are doing everything possible to be super clear in communicating the value that we deliver and the predictability of costs that we provide for customers that will help them through the shift."
— Yamini Rangan, CEO

Assessment: An honest answer and the most important one on the call for the rating. Management cannot time the recovery, so neither can we, and a Hold rating is the right posture until the data show budgets moving back.

Why Net Revenue Retention Goes Flat

A question asked what supports a flat NRR view and whether churn is rising among customers that are not adopting AI. The CFO answered with the components: stable dollar retention, slower credit expansion after the price cuts, and downgrade pressure as customers renew at lower spend. She also noted that NRR is seasonally higher in the second half.

Q: "And in terms of NRR, what gives you that confidence that NRR will be flat? Are you seeing any kind of churn for the customers, those who are not upticking AI?"
— Sitikantha Panigrahi, Mizuho

A: "In Q2, customer dollar retention remained really strong and stable in the high 80s. ... But as Yamini talked about, the agent pricing reduction, the introduction of trials created a near-term headwind for credit expansion. ... We did see pressure in other upgrade motions and how that's manifesting itself is that as customers come up for renewal, we're seeing some downgrade pressure as they sort of optimize their overall spend. Given those trends, we took down our full-year view for net revenue retention to be basically flat to 2026."
— Kathryn Bueker, CFO

Assessment: A clear decomposition. Churn is not the issue; expansion is. The April price cuts that were meant to build credit revenue are, for now, a headwind to the retention metric that the full-year guide relied on in May.

How Much Credit Consumption Is Paid?

A question asked how much credit consumption comes from paid usage versus credits bundled into subscriptions or used in trials. The CEO described HubSpot's adoption funnel (reach, depth, outcomes) and said consumption covers both included and additional credits, but gave no split.

Q: "And I recognize that it's very early, but is there any way to think about how much of total credit consumption today is coming from paid usage versus bundled or trial usage?"
— Eamon Coughlin (for Raimo Lenschow), Barclays

A: "Then it's the depth of usage, then it's the number of outcomes that they're driving that then drives the credit consumption, which includes both the included as well as additional consumption. ... All of those are the pre leading indicators in terms of credit consumption, and as I mentioned, even though we reduced pricing, we've seen credit consumption increase, and we are continuing to see that trend go into July."
— Yamini Rangan, CEO

Assessment: The question asked for the one number that would show whether credits are a revenue line, and the answer did not provide it. Three quarters into the credit model, the paid share is still undisclosed; we continue to give AI revenue no value in our base case.

Did the Pressure Ease Into July?

The last question asked whether the demand pressure was steady through the quarter or worsened toward June. The CFO said it was steady and continued into July.

Q: "Kate, just one for you on the 2 pieces that impacted the quarter here, the deliberate actions you took and then the shift in the demand environment, was the demand environment fairly level and the pressures that you saw consistent throughout the quarter? Or is that something that started to become more pronounced as you exited the quarter?"
— Parker Lane, Stifel

A: "I think that there's not a distinct difference between the pressure at the beginning of the quarter and the pressure at the end of the quarter. And just to be clear, we did see it continue into July."
— Kathryn Bueker, CFO

Assessment: No deterioration through the quarter is better than the alternative, but no improvement into July means Q3 started where Q2 ended. That closes the door on the argument that the April disruption was a one-month event.

What They're NOT Saying

  1. How far below revenue growth net new ARR is running: Management reversed the full-year net new ARR expectation but gave no rate, no dollar base and no half-year figure. The metric is disclosed only annually, so the size of the shortfall stays unknown until February.
  2. Credit revenue, and now credit growth: In May HubSpot gave a 67% sequential growth rate for credits consumed; this quarter it said only that consumption grew. The 10-Q describes HubSpot Credits as a revenue source but gives no amount, the release does not mention credits, and the CEO did not answer how much consumption is paid.
  3. Core seats: The May call cited active core-seat users up 90% YoY and discussed core seats repeatedly; this call did not mention them. The only seat comment was that some customers are swapping seats for credits.
  4. The terms-of-service reversal: HubSpot changed its data-sharing terms on July 1 and reversed the change within days. The opening question referred to it; neither the answer nor the prepared remarks addressed it.
  5. Gross margin: The non-GAAP subscription gross margin fell 180bp YoY to 85.3%, and gross margin was not discussed on the call. The only reference was that AI costs partly offset the operating-margin gain.
  6. A currency bridge for the guide cut: In May the CFO split the revenue change into currency and organic parts. This time the $22.0M cut and the $36.2M reduction in implied second-half revenue came with no currency figure.
  7. The 20% growth goal and the 2027 revenue framework: The 20% revenue growth goal restated in February went unmentioned for a second call. The 2027 framework now has a margin marker (two to three points) and no revenue marker, which is deferred to the September 17 Analyst Day.
  8. The intelligence layer: The paid API layer for third-party agents that management described in May was not mentioned, and no pricing was published.

Market Reaction

  • Pre-print setup: HubSpot closed the print day at $250.21, up 29.1% over the trailing 30 days and 46.9% above its June 25 closing low of $170.28, the 52-week closing low. It was down 37.7% year to date and 49.2% over 12 months, against an S&P 500 up 12.8% year to date. Between prints the stock rose from the $197.34 May reaction close to $262.20 on June 1, fell 35.1% to the June 25 low, then rallied into the print. At $250.21 it traded at 19.1x the May EPS guide.
  • After-hours move: Shortly after the release, which carried the lower full-year guide, the stock was described as down more than 20%; a later after-hours report had it down 19.2% at about $202.
  • Reaction session (August 6): HubSpot opened at $193.19 (down 22.8%), traded between $185.05 and $205.59, and closed at $202.43, down 19.1% ($47.78). The S&P 500 fell 0.2%, a relative move of about 19 points. The close was 9.4% above the intraday low, 1.5% below the high, 2.6% above the May 8 reaction close and 18.9% above the June 25 low. It was the second straight one-day drop of about 19% after a print.
  • Volume: 8.6M shares, 5.6x the 30-day average of 1.5M.
  • Peer reaction (August 6 session): Salesforce (3.2%), ServiceNow +0.1%, monday.com (4.2%), Wix (5.7%) and Atlassian (2.8%). None reported during the session: Atlassian reported after that day's close and Wix two sessions earlier. HubSpot's decline was about 13–19 points worse than its peers'. The five entered the session down 24–38% year to date.

A Second Guide-Down in Three Months: In May the market sold a Q2 revenue guide that sat 0.1–0.5% below consensus. In August the Q3 guide sat about 1.9% below, the full-year revenue guide was cut for the first time in our coverage, and the KPIs that had held in May (net adds and NRR) broke. The release beat on every financial line HubSpot guides, so the move was a verdict on the outlook and the customer metrics, not on the income statement.

A Recovery Rally Unwound: The stock had risen 46.9% from its June 25 low into the print and traded at 19.1x the May EPS guide, a multiple that assumed the April disruption would fade. The call said the opposite: the headwinds continued into July and the guide assumes they last all year. Four rating downgrades on the reaction morning, none to a sell rating, and target cuts across the remaining coverage turned that repricing into a penalty-box trade.

Buyers at the Buyback Price: The close recovered 9.4% from the intraday low and finished within 0.2% of the $202.80 average price HubSpot paid for its Q2 repurchases, on a new $1B authorization. We read the session as a reset to a mid-teens earnings multiple that the business can support, not as a sign that further downside is unlimited, which is why we move to Hold rather than Underperform.

Street Perspective

Debate: Cyclical Budget Freeze or Structural AI Squeeze?

Bull view: The bull case rests on management's framing that the budget scrutiny is a "sorting phase": mid-market buyers paused while they worked out what AI costs, and once spending shifts from tokens to outcomes, a platform that bundles agents with the CRM benefits. Dollar retention is stable, large deals still close, and a guide that assumes no improvement leaves room to beat.

Bear view: The bear camp contends that the pressure is structural: AI budgets are coming out of seat-based software, customers are already swapping seats for credits, and net adds have fallen by a third while HubSpot gives its agents away on trial. On this view the company is in a penalty box until growth inflects, and there is no near-term catalyst.

Our take: The evidence today fits both readings, and that is the problem. Stable dollar retention argues against the structural bear case; falling net adds, NRR and ASRPC that will not ramp argue against a quick recovery. We need two quarters of data before choosing, and a Hold rating reflects that.

Debate: Did HubSpot Trade Revenue for Adoption, and Will It Be Repaid?

Bull view: Some desks argue that the April pricing and trial changes are working on the metric they target: agent activations rose 28–80% in a quarter, more than half of Pro+ customers use AI, and credit consumption grew despite lower prices. Adoption leads revenue, and the 2027 margin target shows management can fund the transition.

Bear view: Skeptics point out that the company that cut prices to drive adoption now reports lower NRR, slower credit expansion and fewer net adds, and still gives no credit revenue figure. Each quarter without one makes the AI story harder to value, and the subscription gross margin is falling as usage grows.

Our take: The adoption data are the best part of the report, but the bears are right that adoption has not reached revenue. The Analyst Day on September 17 is the place for a credit revenue or paid-attach disclosure; without one, we keep AI revenue out of our estimates.

Debate: At 15x Earnings With $1.26B of Buyback Capacity, Floor or Value Trap?

Bull view: The bull case is that a business with a 21% operating margin guide, about $750M of free cash flow and 2–3 more points of margin coming in 2027 should not trade at 15x earnings and 11.6x enterprise value to free cash flow, and that the company itself is buying stock at these prices with capacity for about 12% of the shares.

Bear view: The bear camp argues that the multiple is low for a reason: growth is not near an inflection, net adds and NRR are still falling, and the stock stays in the penalty box until that changes. On our numbers, stock-based compensation also leaves free cash flow after dilution at only a 2.7% yield on enterprise value, and a buyback cannot offset shrinking customer growth.

Our take: We side with the floor view on valuation and with the bears on direction. At $202.43 the price already reflects slower growth, which limits downside, but it also fairly values the business we now model. That combination is a Hold.

Model Update & Valuation Framework

ItemPrior Estimate (Q1 2026 Recap)Updated Estimate (Q2 2026 Recap)Reason
Q2 2026 revenue$908M$911.7M (actual)Beat the top of the guide by $13.7M
Q3 2026 revenuen/a$934MAbout $9M above the top of the guide, a smaller beat than Q2's
FY 2026 revenue$3,730M$3,700M (+18.2% reported)Lower guide and no recovery assumed; 0.4% above the guide top
FY 2026 constant-currency growth~17%~16%Guide cut to 16%; Q3 guided to 15%
FY 2026 non-GAAP operating income$780M (20.9%)$775M (20.9%)Guide held at $762–766M; Q2 beat by $11.3M
FY 2026 non-GAAP EPS (15% tax)$13.36$13.47Lower share count (~50.0M) offsets lower other income (~$17M)
FY 2026 free cash flow$760M$755MGuide ~$750M
FY 2026 net revenue retention~104.5%~103%Guide cut to roughly flat; Q2 at 102%
Net customer additions~9,500 per quarter~5,500 per quarter (Q3–Q4)Guide cut to 5,000–6,000
2027 constant-currency growthn/aLow teensCustomer growth near 10% by year-end, ASRPC ~2% CC, net new ARR below revenue growth
Base-case P/E multiple19x16xNet new ARR commitment withdrawn; NRR below 103%; customer growth slowing
12-month price target (base)$254$21616x our FY 2026 EPS of $13.47

Valuation framework: At the reaction-day close of $202.43 and 49.87M shares outstanding, HubSpot's market cap is about $10.1B; net of $1.38B of cash and investments (no debt), enterprise value is about $8.7B. That is 2.4x the 2026 revenue guide midpoint, 11.6x guided free cash flow (a 7.4% yield on market cap) and 15.3x the 2026 EPS guide midpoint (15.0x our $13.47). We keep P/E as the anchor and EV to free cash flow as the cross-check.

ScenarioFY 2026 EPSMultipleValuevs. $202.43What It Assumes
Bull$13.8221x$290+43%$3.72B revenue, 21.3% margin; budgets loosen by Q4; credit revenue disclosed and growing; net adds back above 7,000
Base$13.4716x$216+7%$3.70B revenue, 20.9% margin, NRR ~103%; low-teens CC growth in 2027 with 2–3 pts of margin gain
Bear$13.0111x$143(29%)$3.67B revenue (below the guide), 20.5% margin; growth slides toward 10% as seats give way to credits

Valuation impact: Our base-case target falls to $216 from $254 (down 15%), which implies about +7% from the $202.43 close, close to what we expect from the market. Our EPS estimate rises 0.8% to $13.47, entirely because of the lower share count; the target falls because we cut the multiple to 16x from 19x to reflect the withdrawn net new ARR commitment, NRR below our review line and customer growth heading toward 10%. A 50/25/25 weighting of the base, bull and bear cases gives about $216. Upside to the base case is about $14 against about $59 of downside to the bear case, a ratio of 0.23 versus 1.1 in May: the price barely moved while our estimate of value fell.

What would change our mind:

  • Upgrade back to Outperform: NRR back at 103% or better with management restoring an expansion outlook; net adds above 6,000 in Q3 with conversion improving; a credit revenue or paid-attach figure at the September Analyst Day; a 2027 revenue framework of mid-teens or better; or a further fall in the price below the June closing low without further deterioration in the KPIs.
  • Downgrade to Underperform: Non-GAAP subscription gross margin below 85% without disclosed credit revenue; customer dollar retention falling out of the high 80s; constant-currency ASRPC flat or down; another full-year revenue guide cut in Q3; or a 2027 framework that implies growth near 10%.

Signposts for Q3 2026 earnings:

SignpostWhat to WatchBullish if...Bearish if...
Q3 revenuevs. $924.0–925.0M guide (+15% CC)$935M+ and CC growth of 16%+At or below the top of the guide
Net customer additionsvs. ~5,000–6,000 guideAbove 6,000 with conversion improvingBelow 5,000
Net revenue retention102% in Q2; FY roughly flat103% or betterBelow 102%
ASRPC (CC)+2% for two quarters3%+ in constant currencyFlat or down
Non-GAAP subscription gross margin85.3% in Q2Back to 85.5%+Below 85%
Credits"Grew" in Q2, no rateAny revenue or paid-share figureUsage only, again
Demand environmentPressure continued into JulyManagement describes improvement into Q4Worse than Q2
FY 2026 guide$3,678–3,686M; 16% CC; 21% marginHeld or raised by more than the Q3 beatCut again
Analyst Day (September 17)2027 framework; +2–3 pts of marginRevenue framework of mid-teens or betterLow-teens or no revenue framework
Buyback~$1.26B of capacity$250M+ repurchased in Q3A pause, or revolver draws to fund it

Thesis Scorecard Post-Earnings

Q1 2026 Signposts and Commitments Revisited

Our May report set ten signposts for Q2 and our own bar for the quarter, and management made forward commitments on the May call. The table grades them against what Q2 showed.

Signpost or CommitmentBar Set in MayQ2 2026 ActualVerdict
Q2 revenueBullish at $905M+ with CC 17%+$911.7M; 17.5% CCBullish
Net new ARRBullish if back above CC revenue growth; bearish if the FY commitment softenedFY 2026 now expected below CC revenue growthBearish
Sales cyclesBearish if the extension runs into H2Longer cycles plus a weaker demand environment, continuing into JulyBearish
Net revenue retentionBullish 104%+; bearish below 103%102%; FY guide cut to flatBearish
ASRPC (CC)Bullish 3%+; bearish flat or down+2%Neutral
CreditsBullish on any revenue or attach figureConsumption "grew"; no rate and no dollarsBearish
Non-GAAP subscription gross marginBullish 86%+; bearish below 86% without credit revenue85.3%Bearish
FY 2026 guideBearish if the CC rate is lowered16% CC from 16.6%; revenue cut $22.0MBearish
BuybackBullish $150M+ at lower prices$531.9M at $202.80; new $1BBullish
Intelligence-layer pricingBearish on no follow-throughNot mentionedBearish
Commitment: Q2 guide ($897.0–898.0M, 19% margin, $3.00–3.02 EPS)Guide$911.7M; 20.3%; $3.26Beat
Commitment: a modest acceleration in CC growth after Q2GuideQ3 guided to 15% CC after Q2's 17.5%Missed
Commitment: net adds back to 9,000–10,000 from Q2Guide6,988; guide cut to 5,000–6,000Missed
Commitment: ASRPC growth ramping through 2026Guide+2% CC again; ramp language droppedMissed
Commitment: NRR +1 to 2 pts for FY 2026GuideCut to roughly flatMissed
Commitment: FY net new ARR growth above CC revenue growthGuideReversed to belowWithdrawn
Commitment: sales cycles moderate as trials convertCallCycles lengthened furtherMissed
Commitment: FY margin 21%, FCF ~$750M, SBC ~14%GuideHeld; SBC now ~13.9%; 2027 +2–3 pts addedMet
Commitment: fall Analyst Day with an updated frameworkCallSet for September 17; 2027 margin marker given earlyOn track

Scorecard summary: Two of ten signposts were bullish, one neutral and seven bearish. Of our own seven-part bar, three parts were met (revenue, constant-currency growth, buybacks) and four missed (NRR, subscription gross margin, net new ARR, credit revenue). Management beat every Q2 financial line it guided and missed or withdrew every customer and growth commitment it made in May. Three of the four Hold triggers we set in May fired; none of the three Underperform triggers did. The seat-for-credit swaps do not trip the first Underperform trigger: that trigger named its own test, constant-currency ASRPC flat or down, and ASRPC grew 2%, so we log the swaps as first evidence under Bear #1.

Thesis PointStatusNotes
Bull #1: Durable high-teens compounder with leading indicators turning upBroken (At Risk → Broken)Every leading indicator turned down: net adds 6,988 and guided to 5,000–6,000, NRR 102% with a flat FY guide, net new ARR now below revenue growth for 2026, Q3 guided to 15% CC and the FY CC guide cut to 16%
Bull #2: Upmarket and multi-hub consolidation expand wallet shareConfirmed, weaker (On Track)Deals over $120,000 ARR +38% (Q1: +64%); 64% of new Pro+ customers multi-hub (+3 pts); upmarket still described as the strongest part of the business, with longer approvals
Bull #3: Operating leverage plus capital return at a trough multipleConfirmed (On Track, strengthened)Margin 20.3% (+330bp); FY 21% held on lower revenue; 2027 +2–3 pts; $531.9M bought back at $202.80 and $1B added; SBC 14.1% of revenue
Bear #1: AI disintermediates the seat-based system of recordNeutral (Emerging, first evidence)Customers swapping seats for credits; AI costs elsewhere driving downgrades at renewal; core seats unmentioned. Dollar retention stable in the high 80s, so no evidence of customers leaving
Bear #2: AI monetization unproven while non-seat expansion is stalledChallenged (Materializing, deepened)Price cuts and trials slowed credit expansion and NRR; credit growth no longer quantified; still no credit revenue; subscription gross margin down 180bp
Bear #3: SBC-heavy cost structure makes non-GAAP cheapness overstate valueNeutral (Contained)SBC 14.1% of Q2 revenue vs. 18.5%; FY ~13.9%; SBC-burdened FCF yield on EV 2.7%; 2.3M new shares reserved in June (~4.6% of shares outstanding)

Overall: Thesis weakened materially. The leading-indicator pillar we placed at risk in May is broken, and the AI-monetization bear point deepened within its tag. What remains intact is the upmarket pillar, with less momentum, and the margin and capital-return pillar, which is now doing all the work. That is the profile of a well-run, slower-growing software company at a fair price, not of the re-accelerating compounder we initiated on.

Action: Downgrade to Hold with a $216 base-case target. In May we said we would add if Q2 showed NRR at 104% or better and a recovery in net new ARR; it showed 102% and a withdrawn commitment. We do not move to Underperform because none of our Underperform triggers fired, dollar retention is stable and the price already reflects slower growth. We would revisit the upgrade case on a Q3 print with NRR back at 103% or better, net adds above 6,000 and a credit revenue disclosure at the Analyst Day.

Independence Disclosure As of the publication date, the author holds no position in HUBS and has no plans to initiate any position in HUBS within the next 72 hours. Aardvark Labs Capital Research maintains a firm-wide policy of not trading any security we cover. No compensation has been received from HubSpot, Inc. or any affiliated party for this research.