Ben’s Portfolio update end of August 2026
Returns and portfolio holdings:
| Portfolio | Notes | |
|---|---|---|
| 2022 | -15.6%* | *Jul-Dec, since I started posting my portfolio on Saul’s and fully adopting my version of Saul’s investing approach. |
| 2023 | 77.8% | |
| 2024 | 31.7% | |
| 2025 | 24.9% | |
| 2026 | YTD | Month |
| Jan | -8.2% | -8.2% |
| Feb | -15.9% | -8.3% |
| Mar | -14.7% | 1.3% |
| Apr | -9.4% | 6.3% |
| May | 25.2% | 38.2% |
| Jun | 28.9% | 2.9% |
| Jul | 31.3% | 1.9% |
| Aug | 39.4% | 6.1% |
*time-stamp: August 31st, after market close
These are my current positions:
| August 2026 | July 2026 | First buy* | |
|---|---|---|---|
| Cloudflare | 22.7% | 22.0% | 11/2/2020 |
| Nvidia | 16.2% | 15.8% | 5/13/2020 |
| Datadog | 15.8% | 19.2% | 5/13/2020 |
| Crowdstrike | 14.5% | 12.9% | 5/13/2020 |
| Snowflake | 13.2% | 12.5% | 2/8/2021 |
| Axon | 6.5% | 6.0% | 4/2/2024 |
| Astera Labs | 4.4% | 3.8% | 11/18/2025 |
| Samsara | 2.9% | 2.8% | 1/8/2024 |
| AppLovin | 2.1% | 3.4% | 11/18/2025 |
| Zscaler | 1.8% | 1.5% | 3/4/2021 |
*held through today
*time-stamp: August 31st, after market close
Company comments
Datadog:
Datadog’s results for the second quarter of 2026 confirm exactly what we had already observed in Q4 2025 and Q1 2026: This platform isn’t just scaling - it’s actively accelerating and evolving into a cash cow. With revenue of $1.12b, a YoY growth rate of 35.6%, which met my expectation of 35.2%, the company posted a huge sequential net revenue increase of $115m. CFO David Obstler noted during the conference call that this was by far a sequential record (nearly double the previous record). Even better is the operating leverage. While revenue rose by just under 36%, operating expenses grew by only about 26%. That explains why operating margins rose to 22.9% and free cash flow came in at $279 million, giving us a margin of 24.9%. The financial profile here is spectacular as the company scales beyond a projected annual run rate of $4.4 billion.
When it comes to customer metrics, there’s an obvious elephant in the room: Datadog added only 200 net new customers this quarter - the lowest sequential growth I’ve recorded since 2018. At first glance, this 0.6% quarter-over-quarter (QoQ) growth seems alarming. But when you dive deeper into the underlying structure, it becomes clear that the thesis of enterprise consolidation, which I highlighted last quarter, is playing out perfectly. They added 170 new customers with an ARR of over $100k, far surpassing the 80 new customers from the second quarter of last year and easily exceeding my expectations. This shows us that the SMB segment at the base is still turbulent and experiencing churn, but Fortune 500 companies are standardizing on Datadog. This momentum in the enterprise segment led to a sharp increase in billings (invoiced revenue) of 38.5% YoY to $1.18 billion. And the platform’s customer stickiness continues to strengthen: the cohort using more than 10 products now stands at 13% (up from 11% in Q1), and 58% of customers now use 4 or more products. This is precisely what keeps the net retention rate in the low 120s and drives this outstanding 10.8% QoQ jump in implied ACV:
Implied ACV is calculated as [ARR (revenue × 4)] / customers in thousands of USD.
One of the most encouraging updates from the Q2 earnings call was the additional context management provided on the breakdown between its AI-native and traditional enterprise cohorts. CEO Olivier Pomel specifically highlighted that revenue growth for non-AI customers accelerated again this quarter, reaching the high 20% range (YoY); up from the mid-20% range in Q1 and just 18% a year ago. This is an important signal. It proves that this reacceleration is not just a byproduct of a few AI startups burning through venture capital. Traditional companies are aggressively moving beyond the optimization phase and migrating back to the cloud and modern infrastructure. Meanwhile, the AI-native cohort remains incredibly strong: 31 customers now spend over $1 million annually, and all 10 of the top 10 AI market leaders rely on the platform.
Finally, we need to discuss the Q3 forecast and the comments regarding their largest customer. The revenue forecast for Q3 was $1.14 billion (28.7% YoY). Historically, I would view this as setting the stage for a final result of 35% YoY. However, management provided a crucial piece of context: They have signed a nine-figure contract extension with a leading AI company, but the forecast deliberately factors in a “reduction in usage” for this specific account starting in Q3. Obstler made it clear that they have fully adjusted the future outlook for this customer to account for this risk. While this headwind makes the Q3 forecast look somewhat weaker sequentially, the underlying business momentum is largely masked by this optimization for a single customer. When you pair this risk-adjusted forecast with the remaining performance obligations (RPO), which rose by an incredible 42.8% YoY to $3.47 billion - representing cRPO growth of around 40% year-over-year and a 3%(!) increase in the full-year revenue forecast - the broader pipeline is quite clearly exceptionally strong. Given this level of execution, cash generation, and sustained enterprise growth, the risk-reward profile for Datadog right now looks very compelling to me.
Overview of how Datadog performed versus my prior expectations:
- Revenue expectation: $1118M (11.1% QoQ, 35.2% YoY), implying a 4.0% beat.
→ Revenue was $1121M (11.4% QoQ, 35.6% YoY), a 4.3% beat. - Q3 new revenue guide: $1152M (3% QoQ, 30% YoY) which I would interpret as $1196M (7% QoQ, 35% YoY) expecting YoY growth will stay close to 35%.
→ Q3 new revenue guide was $1140M (1.7% QoQ, 28.7% YoY), which I now interpret as $1195M (6.5% QoQ, 34.9% YoY), implying a 4.8% beat. - My Q2 revenue expectation implies about $111M raw sequential revenue increase (up from $65M last Q2).
→ QoQ net new revenue was $115M. - I would like to see RPO at around $3.58b (3% QoQ, 47.5% YoY growth) and cRPO YoY growth at 40%.
→ RPO was $3.47b (-0.3% QoQ, 42.8% YoY) and cRPO grew about 40% YoY. - I would like to see Billings at around $1.13b (10% QoQ, 33% YoY growth).
→ Billings was $1.18b (14.6% QoQ, 38.5% YoY growth). Note: good to see stronger than expected billings in quarters with weaker than expected RPO. - I would like to see QoQ customer growth around 2.5% (~830 new) and for the $100k+ cohort, around 2.5% QoQ (~114 new).
→ Customers grew only 0.6% QoQ with only 200 net adds in the quarter the lowest sequential adds I have on record going back to 2018 with the second lowest adds being around 500. What happened here??? but large customer growth was much better than I expected: the $100k+ cohort grew 3.7% QoQ, 170 new, which is up from 80 last Q2. - I would like to see continued multi-product adoption progress with 2+, 4+, 6+, 8+ and 10+ products cohort percentages to stay stable at 85%, 56%, 35%, 21% and 12%.
→ 2+, 4+, 6+, 8+ and 10+ products customer cohort percentages were 85%, 58%, 37%, 22% and 13%, respectively. - I would like to see NRR around 122.5% (low-120s %).
→ NRR was in the low-120s %. The continued strong multi-product adoption and NRR is also reflected in stellar implied ACV ([ARR (rev x4)]/customers) improvements over the years, growing 10.8% QoQ this Q2, which is a sharp step up from recent sequential growth steps on the order of 1% to 5.5%. - I would like to see OM ~20%, NM ~24%, FCFM ~21%. Furthermore, I’d like to see operating expense YoY growth below revenue YoY growth.
→ very strong profitability metrics this quarter, driven by huge operational leverage where operating expenses grew only 25.8% YoY, while revenue grew 35.6% YoY: Operating margin was 22.9%, up from 19.8% last Q2. Net margin was 26.5%, up from 24.7% last Q2. FCF margin was 24.9%, up from 20.0% last Q2. - Thoughts from previous quarter: Datadog’s 1Q26 earnings recap.
Cloudflare:
On August 6, Cloudflare announced its results for the second fiscal quarter of 2026, exceeding virtually all of my prior expectations. Revenue totaled $696.1M, representing an 8.8% increase QoQ and a 35.9% increase YoY. This easily surpassed my expectation of $684M and further accelerated the 34% YoY growth rate from Q1. The outlook was equally strong: The revenue forecast for Q3 was set at an average of $736.5M, representing 31% YoY growth. I now interpret this forecast as $762M (9.5% QoQ, 35.6% YoY), implying a 3.5% beat. Even more revealing is that management raised its full-year revenue forecast by 2.0% - the largest upward revision for a fiscal year we’ve seen from them in four years. The narrative from Q1, which focused heavily on internal restructuring to address the “Agentic Web,” is now clearly reflected in external momentum. During the conference call, Matthew Prince cited a fascinating statistic: for the first time in human history, non-human traffic accounted for more than 50% of the data flowing through Cloudflare’s network.
Following management’s decision in Q1 to drop total customer count as a metric, the focus has shifted entirely to the enterprise segment, which continues to exceed my expectations. I had anticipated growth in large customers ($100k+ ARR) of about 5.1% QoQ (approximately 224 net new customers). Instead, they added 282 large customers - a 6.4% increase QoQ and a significant rise compared to the 118 new customers in the first quarter and the 185 in the same quarter last year (Q2). This is a new record for any second quarter reported to date, with the company adding a record 986 large customers YoY. More importantly, after my initial disappointment over the decline to 118% in the first quarter, the Net Retention Rate has now rebounded to 120%, clearing the 119% threshold I had set. The “land-and-expand” strategy in the enterprise segment is in full swing and recorded record net additions in every single enterprise customer cohort, from $100l+ to $5M+ ARR customers. As a result, enterprise customers now account for 73% of total revenue, up from 71% a year ago.
Beneath the surface of these enterprise figures, adoption of the developer platform is skyrocketing. While the addition of 1M new active developers in Q1 was already staggering, the numbers in Q2 were almost unbelievable: the number of active developers grew by 34.5% QoQ, representing an increase of nearly 2M new developers, bringing the total to 7.4M. As Prince emphasized, more developers joined in Q2 alone than in all of 2025. This hypergrowth is directly linked to the “Agentic Web” narrative, as Cloudflare Workers has become the default, elastic & cost-effective container infrastructure for AI agents. This is also reflected in the contract structures, as several massive “pool of funds” deals were signed this quarter - including a $7.5M contract with a generative AI company, which wanted to avoid hyperscaler egress costs, a $4M contract with a tech company from the APAC region, and a $6M contract with a tech company building a platform for AI agents (to name just a few). To support this shift toward machine-to-machine traffic, Cloudflare used the second quarter to lay the groundwork for an “Agentic Economy.” With the launch of Monetization Gateway, Wallets, and cloudflare.pay, the foundation was laid to enable microtransactions for agents across the web.
In Q1, the most significant warning sign I identified was a sharp slowdown in the growth of Total RPO (total remaining performance obligations). That concern was alleviated this quarter. Total RPO accelerated again, rising 7.4% QoQ to $2.73 billion (up 38% YoY), easily surpassing my target of $2.70 billion. Current RPO also met my exact expectation of $1.75 billion, with growth of 6.7% QoQ and 35% YoY. At 64%, cRPO continues to account for an above-average share of total RPO, which signals excellent short-term revenue transparency. CFO Thomas Seifert did point out that, due to the business’s increasing shift away from purely flat-rate SaaS contracts toward these large, usage-based “pool of funds” and “T-shirt size” deals, there could be slightly higher QoQ volatility as customers use up their quotas and renew early. However, the underlying demand profile continues to show a steep upward trend.
Finally, regarding profitability, the margin pressure we saw in Q1 is stabilizing, exactly as management promised. The gross margin rose slightly from its low of 72.8% in Q1 to 73.1%, successfully surpassing my threshold of 73%. Seifert noted that, although paid traffic continues to drive a reallocation of network costs relative to free traffic, this dynamic is slowing, and the focus is now on expanding overall unit economics. On the operating side, while operating expense growth rose to 29.7% YoY (compared to 25.3% in the first quarter), revenue growth of nearly 36% significantly outpaces this. This resulted in an operating profit of $96.1M and a margin of 13.8%. Even though I will be keeping a close eye on operating expenses to ensure they do not rise much further - since revenue growth is likely capped at around 36% - the operating leverage remains intact. Free cash flow and CapEx were both lower this quarter - the FCF margin stood at 8.1% ($56.4M), and CapEx was sharply reduced to just 7% of revenue. Crucially, the restructuring in Q1 - which resulted in a charge of $151M to prioritize speed - did not derail cash generation. The long-term trend of TTM FCF versus TTM CapEx remains fully intact. Cloudflare is proving that it has the right infrastructure for the AI transformation, is maintaining aggressive growth, and at the same time has a firm grip on its bottom line.
- Reporting Fiscal Q2 2026 on 8/6/26.
- Revenue expectation: $684M (7.0% QoQ, 33.6% YoY), implying a 3.0% beat.
→ Revenue was $696M (8.8% QoQ, 35.9% YoY), a 4.7% beat. - Q3 new revenue guide: $726M (6% QoQ, 29.1% YoY) which I would interpret as $746M (9.0% QoQ, 32.7% YoY) expecting YoY growth rate to slightly decelerate.
→ Q3 revenue guide was $736.5M (5.8% QoQ, 31.0% YoY), which I now interpret as $762M (9.5% QoQ, 35.6% YoY), implying a 3.5% beat. Also noteworthy: They raised their FY guide by 2.0%, the largest FY revenue guidance raise we’ve seen in 4 years. - I would like to see NRR at 119%.
→ NRR was 120%. - I would like to see large customer growth around 5.1% QoQ (~224 net adds, compared to last Q2’s 185 adds).
→ large customers grew 6.4% QoQ, adding 282 new ones to this cohort, a significant step up from last Q2’s 185 and Q1’s 118, and a new record add for any Q2 previously reported. Also active developers grew an incredible 34.5% QoQ, to 7.4M!! - I would like to see RPO grow around 6.0% QoQ to $2.70b (36% YoY).
→ RPO grew 7.4% QoQ to $2.73b and is up 38.2% YoY. - I would like to see cRPO grow around 6.7% QoQ to $1.75b (34% YoY).
→ current RPO grew 6.7% QoQ to $1.75b and is up 34.0% YoY. - I would like to see Gross Margin greater or equal to 73%.
→ Gross Margin was 73.1%, up from a bottom of 72.8% in Q1. - I would like to see operating income around $100M (14.6% margin vs. 14.1% last Q2).
→ operating expense growth has been on the rise, growing 29.7% YoY, up from 25.3% in Q1 and up from 22.6% last Q2, but so have revenue growth rates which are at 35.9% YoY, up from 33.5% in Q1 and up from 27.8% last Q2. So operating leverage stays in tact, but I expect revenue growth rates will not accelerate beyond 36%, so will keep an eye on operating expense growth not going up much more than current levels. This Q2, operating income was $96M, a 13.8% margin. - I would like to see a FCF margin around 14.0% ($97M) and Capex around 11% of revenue ($75M).
→ both FCF income and Capex were significantly down this quarter. FCF margin was 8.1%, compared to 13.1% last Q, but up from 6.5% last Q2. Capex was only 7.2% of revenue. The TTM FCF vs. TTM Capex trend stays fully intact though where TTM Capex was going down while TTM FCF was going up. - Thoughts from previous quarter: Cloudflare’s 1Q26 earnings recap.
Astera Labs:
Astera Labs announced its results for the second fiscal quarter of 2026 on August 4, and the revenue figures have definitively demonstrated that its growth engine is accelerating. The company reported revenue of $392.4M (up 27% QoQ and 104% YoY), exceeding its own forecast by 9.0%. But the real highlight was the forward-looking guidance. The revenue forecast for Q3 averaged $550M, completely exceeding my expectation of $435M. Applying my historical interpretation model, I now expect this forecast to point to an actual Q3 result of around $591M, which would represent incredible growth of 51% QoQ and 156% YoY. This completely shifts the narrative: from a steady acceleration back to an absolute breakout phase. It proves that the wave of infrastructure spending is directly translating into Astera’s revenue.
Looking at the product mix, the shift toward Scorpio is happening even faster than expected. Last quarter, management anticipated that the Scorpio family would overtake Aries by the end of the year, but now they forecast that Scorpio will become their largest product line as early as the third quarter - a full quarter ahead of schedule. This is driven by the start of volume production of the Scorpio X series, particularly the rollout of the high-port-density (high-radix) 320-lane configurations at its leading hyperscaler. Management noted that the revenue potential per unit (content opportunity) for Scorpio X alone is expected to rise significantly above $1000 per XPU in future platforms. Meanwhile, Aries remains extremely resilient and posted record quarterly revenue in Q2. The transition to PCIe 6.0 is going exceptionally well; PCIe 6.0 products across the Aries and Scorpio series now account for more than 50% of total company revenue, up from just one-third in Q1.
Beyond core scale-up switching, the broader connectivity portfolio is reaching key milestones. Taurus began pre-production of 100G-per-lane modules to support 800G scale-out applications, and introduced new 200G-per-lane smart retimers and redrivers to double its Taurus TAM to $4b by 2030. In the CXL (Compute Express Link, an open standard for high-speed connections between the CPU and devices as well as between the CPU and memory) segment, they secured a new design win with a U.S. hyperscaler for their standard Leo memory controller and expect to begin volume shipments to two U.S. hyperscalers by 2027. Management also presented a clear optical roadmap: They are targeting Near Packaged Optics (NPO) chipsets for production in 2027, followed by fully integrated Scorpio X-series switches with Co-Packaged Optics (CPO) for 2028 and beyond.
However, the most significant mechanical updates this quarter relate to margins. The gross margin stood at 73.7%, missing my target of 76% but slightly exceeding the forecast of 73%. CFO Des Lynch attributed the wider margin range to the shift in revenue mix toward module-based Scorpio shipments rather than pure silicon components. With this structural shift, they expect gross margins to converge toward a long-term target of 70%, starting with a forecast of 72% for Q3. However, the operating leverage they are demonstrating fully offsets the decline in gross margin. The operating margin reached 39.1% (exceeding my expectation of 37%), and they set an incredible operating margin target of 43% for Q3: “Our non-GAAP operating margin is expected to be around 43%, a sequential increase of 400 basis points, which reflects significant operating leverage given our revenue growth.” While the net margin, at 37.2%, remained slightly below my target of 40%, the fact that revenue growth is translating so efficiently into profit demonstrates that they have the operating leverage to easily absorb their aggressive R&D investments in custom silicon and optical interconnects.
Overview of how Astera Labs performed versus my prior expectations:
- Revenue expectation: $382M (23.8% QoQ, 98.8% YoY), implying an 6% beat.
→ Revenue was $392M (27.3% QoQ, 104.5% YoY), a 9.0% beat. - Q3 new revenue guide: $435M (14% QoQ, 89% YoY) which I would interpret as $458M (20% QoQ, 98.6% YoY).
→ New Q3 revenue guide was $550M (40% QoQ, 139% YoY), which I now interpret as $591M (51% QoQ, 156% YoY), implying a 7.5% beat. - just wow! - I would like to see a Gross Margin around 76%.
→ Gross Margin was 73.7%. - I would like to see an Operating margin around 37%.
→ Operating margin was 39%. - I would like to see a Net margin around 40%.
→ Net margin was 37.2%. - Thoughts from previous quarter: Astera Labs 1Q26 earnings recap.
Wrap up
Another earnings season is already over again, and while I didn’t have much time yet to dig deeper into the reports of my other companies, I’ve been very pleased with this earnings season as the tech and security pillars of my portfolio, namely DDOG, SNOW, NET, CRWD, and NVDA, all delivered spectacular results, proving that enterprise budgets aren’t shrinking; they are simply hyper-concentrating on mission-critical AI, data infrastructure, and platform consolidation.
Wishing you all a great rest of September!
Ben
Past recaps
2022: Jul 2022 | Aug 2022 | Sep 2022 | Oct 2022 | Nov 2022 | Dec 2022
2023: Jan 2023 | Feb 2023 | Mar 2023 | Apr 2023 | May 2023 | Jun 2023 | Jul 2023 | Aug 2023 | Sep 2023 | Oct 2023 | Nov 2023 | Dec 2023
2024: Jan 2024 | Feb 2024 | Mar 2024 | Apr 2024 | May 2024 | Jun 2024 | Jul 2024 | Aug 2024 | Sep 2024 | Oct 2024 | Nov 2024 | Dec 2024
2025: Jan 2025 | Feb 2025 | Mar 2025 | Apr 2025 | May 2025 | Jun 2025 | Jul 2025 | Aug 2025 | Sep 2025 | Oct 2025 | Nov 2025 | Dec 2025
2026: Jan 2026 | Feb 2026 | Mar 2026 | Apr 2026 | May 2026 | Jun 2026 | Jul 2026