So this week two of the more important bellwether names in B2B software reported earnings. And neither of them just “beat.” Both accelerated. Materially. At $7B and $5.6B+ run rates respectively. They crushed the quarter, in fact.
This really matters right now.
Because for the past 18 months the narrative on public software stocks in general has been brutal. SaaS is dead, AI-native is everything, the legacy SaaS leaders are in structural decline, and even the great ones are stuck at sub-15% growth forever. We’ve all been writing about the SaaSpocalypse, the great repricing of every B2B software company that hasn’t materially repositioned around AI agents.
And then yesterday happened.
- Atlassian: $1.79B in revenue, +32% YoY. Cloud accelerated to 29%. EPS of $1.75 vs $0.98 expected (a 78% beat). Stock up 22.77% today to $84.21.
- Twilio: $1.41B in revenue, +20% YoY, the fastest growth in more than three years. Raised full-year guidance. Q2 EPS guided to $2.50 to $2.60 vs $1.29 expected. Stock up 19.59% today to $177.05.
Both stocks ripped. But context matters. TEAM came into yesterday at $68.59, down ~70% over the past year. Cantor Fitzgerald had cut its price target from $146 to $98 just three days earlier. BTIG cut from $140 to $110 the day after that. The market had largely given up on Atlassian going in. Even after a +22.77% day, the stock is still down ~65% from its 52-week high of $242.
So a fair framing of yesterday: both companies delivered prints strong enough to break a brutal narrative. Whether they’re strong enough to change that narrative is the actual question.
So the obvious question: Is the SaaSpocalypse over?
Let me give you 5 takeaways from each, then my honest answer.
Atlassian: 5 Takeaways. Revenue Up 32%, Stock Price Still Down -63% Last 12 Months
1. 32% Growth at a $7B+ Run Rate Is Frankly Unbelievable
The headline number is $1.79B in revenue, up 32% YoY. At this scale, $7B+ annualized, basically nobody outside of Nvidia is growing at this rate. To put this in perspective: a year ago Atlassian did $1.36B in this same quarter. They added more than $400M of quarterly revenue YoY. That’s a top-50 SaaS company added in a single year on top of the existing business.
2. Cloud Accelerated. That’s the Whole Game.
Cloud revenue crossed $1.1B and accelerated to 29% growth. Acceleration is the rarest event in public SaaS. Almost nothing accelerates at scale once you’re past a few billion in ARR. The fact that Cloud is speeding up, not slowing down, is the single most important data point in this report.
The catch: about $50M of the beat came from upfront term license revenue pulled forward from FY27 due to the March pricing change. Strip that out and underlying growth is more like 28%. Still extraordinary. But it does mean FY27 comps will be harder than they look.
3. Rovo It’s AI Offering Is Actually Working, and Driving Real Expansion
This is the line that should make every SaaS CEO take notice: customers using Rovo are growing their ARR at roughly 2x the rate of customers who don’t use it. AI Rovo credit usage is growing 20%+ month-over-month.
This is the proof point everyone has been looking for. AI in SaaS isn’t just about deflecting tickets or automating a feature. When deployed as a real expansion lever, it materially increases NRR. Atlassian just gave us the cleanest data point we’ve seen on this from a $7B+ company.
4. The Largest-Ever Quarter for Competitive Displacements
Mike Cannon-Brookes called this “our largest-ever quarter for competitive displacements from a major ITSM provider.” Read: ServiceNow. Jira Service Management is taking real share at the enterprise. This is the kind of comment that gets a CFO at a competitor on a plane.
If you’re wondering whether AI is creating new openings to displace incumbents in categories everyone thought were locked up, the answer is yes. Apparently even in ITSM.
5. RPO Up 37% YoY to $4B = The Pipeline Is Real
Remaining performance obligations rose 37% to $4B. RPO is the truest forward indicator in SaaS. It’s contracted but unrecognized revenue. 37% RPO growth on 32% revenue growth tells you the deal flow is real and lengthening. Customers are signing bigger and longer commitments. That’s not a sugar high. That’s a structural shift.
Twilio: 5 Takeaways. Revenue Up 20%, Stock Price Up +82% (!) Over Last 12 Months
1. 20% Growth, the Highest in 3+ Years. Twilio Re-Accelerated.
Twilio was largely left for dead in 2023 and 2024. Growth had collapsed to single digits. Activist investors. CEO change. The whole story.
Q1 2026: $1.41B in revenue, up 20% YoY, the fastest growth rate in more than three years. Organic growth of 16%. This is one of the more remarkable re-accelerations in modern public SaaS history. If you owned this stock through the dark period, congratulations.
2. Voice Is the Engine, and It’s All AI
Voice revenue grew 20% YoY, the highest in 19 quarters. Almost five years. The driver is AI agent workloads.
Software add-ons like Conversational Intelligence and Branded Calling both grew more than 100% YoY. That’s not a rounding error. That’s a re-pricing of what Twilio is, from messaging-and-voice infrastructure to the orchestration layer for AI customer interactions.
3. The Customer List Is the Real Tell: Sierra. Bland.ai. Posh.
The customer wins this quarter weren’t just Fortune 500 logos. They were the AI-native leaders in voice and customer experience:
- Sierra (Bret Taylor’s company) signed a cross-sell deal for global expansion
- Bland.ai committed to a multi-year deal for messaging, voice, and software add-ons
- Posh is using Conversation Relay as the voice infrastructure for their bank/credit union AI
- Sela AI signed an expansion deal
- Scorpion, a digital marketing partner, built an AI agent on Conversation Relay that drove $8.4M in revenue and a 39% boost in booking rates in three months
Twilio isn’t selling against the AI agent companies. It’s the picks-and-shovels underneath them. Every voice AI startup needs phone numbers, low-latency connections, compliance, and routing. That’s all Twilio.
4. Multi-Product Attach Up 29%. The Platform Strategy Is Finally Working.
For years, Twilio talked about being a platform, and for years, most customers just bought messaging. That’s changed. Multi-product customer count grew 29% in Q1. The percentage of deals closing with multiple products is climbing every quarter.
This is what a real platform consolidation looks like in practice. Customers consolidating spend with one vendor across voice, messaging, and AI orchestration because the data and the channels need to live in the same place.
5. Guidance Raise Was Massive. Margins Improving Too.
Twilio raised full-year revenue guidance from $5.65B to $5.70B, to a new range of $5.78B to $5.83B. Q2 EPS guidance: $2.50 to $2.60 vs. consensus of $1.29. That’s roughly 2x the Street estimate.
This is what happens when you re-accelerate at scale and operating leverage kicks in simultaneously. Non-GAAP operating income of $279M in Q1, FCF of $132M. Twilio has gone from a “growth at all costs” reputation to a real operating discipline story while also re-accelerating growth. That combination is incredibly rare.
The Tell: Customer Counts Reveal Where Growth Is Actually Coming From
Here’s the question that matters more than any headline revenue number: are these companies adding new customers, or just extracting more from the existing base? Revenue growth without logo growth means you’re milking the installed base. That works for a few quarters. It’s not a long-term strategy.
The five-company cohort tells a clearly bifurcated story. Two are slowing on customer adds. Three are accelerating, sharply.
Decelerating customer growth (the application SaaS layer):
- Atlassian’s >$10K Cloud ARR cohort over the last 7 quarters: 17%, 15%, 14%, 13%, 13%, 12%, 10%. A clean grind down. Net new adds in the most recent quarter were only ~544, versus 2,352 the prior quarter.
- HubSpot’s total customer count growth has gone 21%, 19%, 18%, 17%, 16% over 5 quarters. Same pattern.
The high end is doing better at both companies. Atlassian has 600+ customers >$1M ARR, up ~40% YoY. HubSpot’s enterprise tier is expanding well. Revenue is being driven by bigger deals, pricing changes, and existing-customer expansion. Not by net-new logos.
Accelerating customer growth (the AI infrastructure layer):
- Twilio added 43,000 net new accounts in Q1 2026 alone. That’s more than the company’s full-year net adds in 2023.
- Cloudflare added a record 37,000 sequentially in Q4 2025 and now has 332,000+ paying customers, up 40% YoY.
- Snowflake’s quarterly net-new customer adds have grown 40% YoY in the most recent quarter. Net expansion is rising at all three (DBNER 109% at Twilio, NRR 120% at Cloudflare, 125% at Snowflake).
A fair caveat on Twilio: they’re killing the Active Customer Accounts metric this quarter, explicitly because they think it’s “less informative.” Many of those new accounts are AI startups spending the minimum and don’t materially move revenue yet. So the +42% is partly inflated by long-tail signups. But every Bland.ai-style company, every voice AI startup, every Cursor-style agent app needs a Twilio account. The flood of small accounts today is the pipeline for material accounts in 2027.
The pattern is clear.
The companies selling to AI builders (Twilio, Cloudflare, Snowflake) are picking up tens of thousands of new customers because every AI-native startup needs their infrastructure. The companies selling per-seat application software to humans (Atlassian, HubSpot) are seeing logo growth slow even as they extract more from existing accounts.
That’s not a one-quarter accident. It’s a structural divergence that’s been building for 2+ years and is getting starker. Revenue acceleration looks the same on a chart whether it comes from new customers or installed-base extraction. Customer count growth tells you which is which. The answer is uncomfortable for the application SaaS side of the cohort.
So… Is the SaaSpocalypse Over?
No. The market loved these 2 beats, but the bear case isn’t gone. It’s just on pause.
Here’s what these two reports actually tell us:
1. AI is re-accelerating some mature SaaS, when deployed as expansion, not deflection.
Atlassian’s Rovo customers expanding 2x faster. Twilio’s voice AI workloads driving 20% growth in a 19-year-old product line. The companies that have wired AI deeply into their core product and pricing are seeing real acceleration. The companies that bolted a chatbot on the side are not.
2. Infrastructure for AI agents is its own boom.
Twilio’s customer list this quarter, Sierra, Bland.ai, Posh, Sela AI, tells the story. The picks-and-shovels layer for AI agents is a new category of B2B software demand that didn’t really exist 18 months ago. If you sell critical infrastructure that AI agents need (voice, payments, identity, data, communications), you’re booming.
3. The bear case on Atlassian specifically is real, and the pop didn’t make it go away.
This is the part of the analysis that needs more honesty than I gave it in the first draft. Going into yesterday, Atlassian had been the worst performer in this cohort by a wide margin. TEAM was down ~70% over the past year. Two analysts cut price targets in the three days leading into the print: Cantor Fitzgerald from $146 to $98 on April 27, BTIG from $140 to $110 on April 28. KeyBanc trimmed too.
The bear thesis is straightforward and it didn’t disappear yesterday:
If AI agents do the coding, what happens to seat-based developer tools?
Atlassian sells Jira, Confluence, and Bitbucket primarily on a per-seat basis. The whole model assumes you have ~N developers each filing tickets, writing docs, and pushing code. But:
- GitHub Copilot for Jira went into public preview on March 5, 2026. You assign a Jira ticket to Copilot, the agent reads the description, opens a branch, writes the code, and creates a PR. GitHub has 15M+ Copilot users already. The “ticket” still exists, but the human writing it might not.
- Cursor’s cloud agents now run ~35% of internal PRs at companies that have adopted it. Claude Code handles autonomous coding tasks. Devin, Cognition, and a wave of agentic coding products are all moving from “AI assistant inside IDE” to “AI engineer that owns tickets end-to-end.”
- Google’s ADK ecosystem added Jira, Asana, Linear, and Notion integrations the same week.
If a 10-person engineering team genuinely becomes a 4-person team plus AI agents, and there’s now real evidence this is starting to happen, Atlassian has to figure out how to monetize the agents. Otherwise seats decline structurally.
Atlassian’s answer is Rovo (its own AI agent inside Jira/Confluence) and the Rovo MCP server, which went GA in February 2026. The 32% growth this quarter and the 5M+ Rovo MAUs (up 50% QoQ from last quarter’s 3.5M) say the strategy is working so far. Mike Cannon-Brookes said yesterday “all AI is not built equal,” meaning their bet is that proximity to enterprise data and workflow context (the Teamwork Graph) makes Rovo stickier than third-party agents.
But this is a structural debate, not a one-quarter debate. The market gave Atlassian a +22.77% relief rally yesterday because the print was strong enough to suggest Rovo is working. It did not give them a pass on the question of whether AI strengthens or erodes the per-seat developer-tools moat. That question will be litigated quarter by quarter for the next two years.
4. Twilio’s bear case is different, and arguably weaker.
Twilio’s bear case used to be “AI deflects calls, kills minutes, kills usage.” That bear case is dying because the AI agents themselves use Twilio. Voice, messaging, and conversational infrastructure are inputs to the agent stack, not victims of it. Sierra, Bland.ai, Posh, all using Twilio as the layer underneath. The picks-and-shovels framing is structurally different from Atlassian’s seat-based model.
5. Most of the cohort isn’t here yet.
We haven’t seen Salesforce, MongoDB, Datadog, or others report. Cloudflare has been the most consistent grower in this group, 27%+ for 11 straight quarters with FY26 guidance of 28% to 29%. Snowflake just printed 30% product revenue growth. HubSpot has held in the 19% to 21% range. So the cohort isn’t uniformly broken. It’s bifurcating between AI-attached infrastructure and AI-threatened seat models.
The Bifurcation Is Now the Whole Story
There is no longer “SaaS” as a single category. There are now three:
- AI infrastructure (Cloudflare, Twilio, Snowflake): the agents need somebody’s network, somebody’s data warehouse, somebody’s voice layer. Re-accelerating or stable at high rates.
- AI-attached application SaaS (HubSpot, Atlassian-via-Rovo): legacy seat or subscription models being rebuilt around AI as an expansion vector. Growth is holding up if the rebuild is real. This is the contested category.
- Static SaaS: companies that have not repositioned. Still in decline. Multiples still compressing.
Atlassian’s print yesterday was a powerful argument that the second category, the contested one, can work. But one great quarter doesn’t end a multi-year debate, and Cantor and BTIG didn’t cut price targets last week because they’re stupid. The risk that GitHub, Cursor, and the agentic coding stack ultimately erode developer seats is the central debate in software for the next 24 months. Yesterday’s data point is bullish. It is not the verdict.
