US Equity Growth: investor letter Q1 2026

As with any investment, your capital is at risk.

“When everything feels like it’s falling apart, long-term thinking can seem like a distraction – a luxury, even. Crisis narrows our attention. The amygdala fires, breath shortens, and time compresses into a panicked sense of now. Our response is natural and adaptive – designed to help us survive in the moment. But it’s also limited, leaving little room for reflection, foresight, or collective action.” – Patrick Dowd, board president of the Long Now Foundation, in Pace Layers.

Dowd is right. In moments like this, long-term thinking can feel like a luxury. And we’re living through one of those moments: the world is noisy, the market is jumpy, and our recent performance has disappointed. None of that deserves softening. 

At the same time, something else is true: we feel a genuine sense of excitement. Periods that compress everyone’s time horizon are the same periods that create dislocations, when share prices become a referendum on fear rather than a reflection of durable value. It can be a fleeting moment that long-term investors spend years waiting for.

Fundamentals as a starting point

If you have been a reader of our letters over the years, you’d be forgiven for assuming that the answer to the question of ‘what has us excited?’ is… ‘strong fundamentals’. But you’d only be half-right. Yes, fundamentals in the portfolio are strong. And yes, that’s exciting. 

• Average 2026 revenue growth across the portfolio is 20 percent.
• Fundamentals are largely improving, with average 2026 sales estimates having been revised upward by 3 percent over the last six months.
• Profitability continues to trend higher. Average earnings before interest and taxes (EBIT) margins are 11 percent, up from under 9 percent last year.

However, today’s enthusiasm goes beyond the fundamental snapshot. 

Valuation detachment

Over the long term, share prices tend to follow fundamentals. Intermittently, however, there are periods when market narratives emerge and break that pattern – a divergence in which uncertainty runs deep and, for the long-term investor, opportunities are born. 

Today, that gap between narrative and fundamentals feels unusually wide. On both a price-to-sales and price-to-earnings basis, the portfolio has not been this inexpensive relative to the S&P 500 since before 2015. 

Digging deeper, since June 2025, the S&P 500’s P/S has re-rated upwards by 7 percent whereas our holdings have de-rated by 12 percent on average. But because these are averages, they don’t quite do justice to the de-rating many of the companies in the portfolio have experienced. Over the past nine months, 20 companies in the portfolio have de-rated by greater than 25 percent.  

If not fundamentals, then what?

Some of the de-rating is, of course, company-specific: a handful of holdings have faced idiosyncratic fundamental challenges, execution hiccups, or near-term uncertainty that the market has chosen to punish. But that is not the common thread across the bulk of the drawdowns. In fact, the median revenue growth for these 20 companies in the fourth quarter of 2025 was 28 percent (average 31 percent), and estimate revisions have been more upward than downward. So overall, deteriorating fundamentals are not the culprit.

Instead, the dominant narrative shaping sentiment today centres on the ‘AI scare’ – a growing fear that rapid AI adoption will compress pricing power, shorten product cycles, and ultimately erode the durability of earnings in much of the software ecosystem. 

Our take is different. 

In our view, much of the Street’s response has been too blunt – treating ‘software exposure’ as a single bucket and marking down almost anything with a meaningful software component. The reality is more nuanced. One of the ways we aim to understand complexity better is through systems thinking, a holistic problem-solving approach that analyses how individual parts of a system interact and influence one another to produce overall behaviours and outcomes.

Thinking in systems

The market’s current ‘AI kills software’ narrative strikes us as too reductive. AI will almost certainly make certain software features cheaper, faster to build, and easier to replicate. But that does not mean value disappears. More often, it means value moves.

This is why we find it helpful to think in systems. A workflow is not just the interface a user sees. It is the full chain that turns intent into outcome: data moving, permissions being checked, money changing hands, goods being delivered, risks being managed, and trust being maintained. When one part of that chain becomes easier, the whole system does not suddenly become frictionless. The bottleneck usually shifts.

That matters because, in most systems, relieving one constraint reveals the next one. The profit pool often migrates accordingly. If the interface layer commoditises, value tends to accrue to the gates and the rails inside the system: the gates that control access, trust, and decision-making, and the rails that allow transactions, workflows, and execution to happen reliably at scale.

A simple analogy is a balloon. Squeeze it in one place, and the air does not disappear; it bulges somewhere else. AI works in much the same way. It may dramatically increase the speed and volume of certain activities such as writing code, generating content, answering questions and automating routine tasks. But the harder parts of many workflows remain. Trust still has to be earned. Compliance still has to be satisfied. Operations still have to be coordinated. Goods still have to move. Decisions still have to be made. 

So rather than asking whether AI kills software, we think the better question is: 

‘Which constraints are being relaxed, and where do the new gates and rails emerge?’

History offers a useful parallel. Before the internet, one of retail’s key bottlenecks was physical shelf space and local distribution. When commerce moved online, that constraint loosened. But value did not vanish. It shifted, first toward discovery, trust, and digital distribution, and later toward fulfilment speed, logistics, and infrastructure. The bottleneck moved, and value moved with it.

Something similar may happen in software. AI may commoditise parts of the interface layer, the visible features users click on and interact with. But that can make the gates and rails beneath the interface more important, not less. When interfaces become easier to reproduce, value often shifts toward the systems that govern traffic, transactions, permissions, compliance, reliability, and execution. That distinction matters for the portfolio. Our private company research reinforces this view, offering a close look at how the most capital-efficient new models are being built around these dynamics.

Across many of your holdings, we see businesses that sit not merely at the interface, but at the gates and along the rails of the system’s real workflows.

Shopify is one example. Shopping interfaces may proliferate across chat, voice, and AI agents. But commerce still requires a secure, compliant path from intent to purchase: payments, fraud management, fulfilment, returns, and the merchant tools that enable the transaction to occur. Shopify’s importance lies not only in what the customer sees, but also in the rails that enable commerce to move from interest to a completed order.

DoorDash is another. The ordering interface may evolve, and AI agents may eventually compare prices or place orders across platforms. But the hard part of delivery is not the screen. It is the underlying logistics system: balancing courier supply and demand, coordinating merchants, optimising routes and batching, and resolving refunds, mistakes, and fraud across a three-sided marketplace. Here too, the interface may change, but the execution rails remain critical.

Cloudflare is slightly different. It sits at a gate in front of internet traffic, where access is controlled, threats are filtered, reliability is enforced, and, increasingly, automated activity by bots and agents can be governed and monetised. If AI increases the volume and complexity of traffic, that may make control points like these more valuable, not less.

Axon and Samsara illustrate a related but equally important form of resilience. In both cases, the company’s position is anchored by hardware deployed in the real world, allowing it to capture proprietary data at the source. That gives them a gate at the point of capture and, increasingly, rails for turning real-world events into trusted decisions and actions.

Samsara helps businesses run fleets and field operations through connected devices on vehicles and equipment, turning real-time data into safer driving, better maintenance, regulatory compliance, and more efficient routing. Axon supports public safety agencies by capturing and managing video evidence in ways that preserve trust and usability throughout investigations and the justice process. In both cases, AI may commoditise dashboards, summaries, or analytics at the surface level. But the harder thing to replicate is the combination of trusted data capture, system integration, and workflow relevance. Here, AI looks less like a disruptor of the core system and more like an accelerant within it.

Recent datapoints already point in this direction. Cloudflare saw AI agent traffic on its network double in January. Shopify has seen agentic search traffic rise roughly 15x, albeit from a small base. And at Axon, 2025 bookings reached $7.4bn, with $2bn coming from newer products, up roughly threefold year-over-year, while AI-specific products now account for about 10 percent of total bookings. Early though they are, these signals support the broader point: for the right companies, AI is not just a source of disruption. It is also a blessing: increasing volume, deepening relevance, and raising the value of the gates and rails already embedded in the workflow.

That is why we think the market’s current framing is too simplistic. The question is not whether AI changes software. It will. The better question is where AI commoditises value, where it concentrates value, and which companies already sit closest to the next bottleneck. In many cases, we believe the winners will not be the businesses with the flashiest interface, but the ones embedded in the gates and rails that real work still has to pass through.

Performance

Our excitement stands in sharp contrast to the current market fears. 

2026 so far has been shaped by geopolitics and conflict. The first is geopolitical: a world that feels more brittle than it did a year ago. We’ve watched events in the Middle East with deep concern. The region's destabilisation is troubling, not least because the US and Iran appear locked into an escalation dynamic with no clear exit strategy, a possible scenario we are factoring into our thinking about portfolio resilience.
 
Markets price second-order effects, and with the Strait of Hormuz effectively closed, energy has become the choke point. Prices have spiked, and our lack of energy exposure showed in relative terms. Yet, despite feeling profoundly impactful in the moment, over sufficiently long time horizons, events such as this have historically had little bearing on the long-run value creation of the US stock market.
 
The second, discussed earlier, is the ‘AI kills software’ scare. It began as a software derating and then broadened into a wider sell-off in anything that looks like ‘digital’ or ‘data’. That is how narratives work in stressed markets. Narratives don’t handle nuance well. The result? Software has suffered its worst sell-off in 30 years, and even tech bellwether Microsoft has underperformed the S&P 500 for eight consecutive months, its longest streak ever. 

The result has been a sharp divergence in the short run: energy up strongly, software down sharply. That move has not been about fundamentals. It has been about the multiple the market is willing to pay while it wrestles with uncertainty. Against this backdrop, the software industry (21x), the technology sector (21x), and even NVIDIA on 20x (growing about 60 percent year-over-year) are now clustered around a market level multiple of 20x on a forward price-to-earnings basis.

For the portfolio, the main detractors were swept up in this sell-off, compounded by company-specific near-term factors. Take DoorDash – the company executed well, growth was strong, but the shares fell 28 percent as investment accelerated and profitability in newer verticals such as grocery and retail pushed further out. The market treated that as a disappointment. We see it as a deliberate choice: build the rails, widen the moat, then harvest.

In other cases, the debate has been more existential. Duolingo has found itself in the AI crosshairs, with fears that AI translation makes language learning obsolete. That view mistakes a tool for a system. Yes, management has eased near-term monetisation to reaccelerate user growth and engagement, and that has weighed on bookings expectations. But the underlying engine, engagement, habit formation, and product iteration, is intact. If anything, AI is more likely to expand what the product can do than to wipe out the need it serves.

CoStar reflects a different kind of impatience: the market’s demand for near-term margins while a founder-led business reinvests aggressively to build the next leg of a franchise. Homes.com has weighed on reported profitability and tested the market’s patience. But this is a company with deep proprietary data, a proven playbook, and the capacity to fund ambition from a position of strength. In our experience, markets often underestimate how durable that edge is.

Not everything has been dragged down by market fears. Health care provided some ballast. Skilled nursing home provider Ensign has done what it often does in unsettled conditions, quietly compounding. Guardant Health delivered results that reinforce our belief in the breadth of its cancer testing platform. And Penumbra performed strongly, helped by excellent business results and Boston Scientific’s announced acquisition of the company.

Stepping back, the pattern is familiar: in periods like this, the market pays up for what feels safe. It marks down what feels exposed, even when the underlying businesses keep executing. That creates dislocation. And dislocation is not inherently bad news for the long-term investor. It is the mechanism by which short time horizons hand opportunity to long ones.

Outlook

When markets are stressed, the amygdala does what it is designed to do: it narrows attention and makes every headline feel urgent. That instinct may be useful for survival, but it is a poor guide for investing. Our job is to widen the aperture and return to first principles. That is why we have spent time studying systems. When narratives grow loud, we want analytical anchors that are harder to ‘vibe’, grounded not in emotion but in the underlying physics of how systems actually work. Those are the kinds of truths that tend to stay standing even as storylines rotate.

This is what leaves us energised. Periods of market stress and noise hand us clarity. They force the market to separate what is fashionable from what is functional and what is story from what is structural.

Looking ahead, progress depends on fundamentals compounding over time, more than geopolitical and macro shocks. Thus, time and patience are key parts of our competitive advantage. Patience is a resource we deploy when others cannot afford to wait. If we are right, today’s market anxiety will in hindsight read less like a rupture and more like the kind of reset that quietly sets up the next stretch of durable returns, especially for businesses that sit at critical control points and within the essential rails of real-world systems.

US Equity Growth

Annual past performance to 31 March each year (%)

Annualised returns to 31 March 2026 (%)

Source: Revolution, S&P. US dollars. Net returns have been calculated by reducing the gross return by the highest annual management fee for the composite. 1 year figures are not annualised. 
                    
Past performance is not a guide to future returns.
                    
Legal notice: The S&P 500 Index is a product of S&P Dow Jones Indices LLC, a division of S&P Global, or its affiliates (“SPDJI”). Standard & Poor’s® and S&P® are registered trademarks of Standard & Poor’s Financial Services LLC, a division of S&P Global (“S&P”); Dow Jones® is a registered trademark of Dow Jones Trademark Holdings LLC (“Dow Jones”). Neither S&P Dow Jones Indices LLC, Dow Jones Trademark Holdings LLC, their affiliates nor their third party licensors make any representation or warranty, express or implied, as to the ability of any index to accurately represent the asset class or market sector that it purports to represent and neither S&P Dow Jones Indices LLC, Dow Jones Trademark Holdings LLC, their affiliates nor their third party licensors shall have any liability for any errors, omissions, or interruptions of any index or the data included therein.

Risk factors 

The views expressed should not be considered as advice or a recommendation to buy, sell or hold a particular investment. They reflect opinion and should not be taken as statements of fact nor should any reliance be placed on them when making investment decisions.

This communication was produced and approved in April 2026 and has not been updated subsequently. It represents views held at the time of writing and may not reflect current thinking.

Potential for Profit and Loss 

All investment strategies have the potential for profit and loss, your or your clients’ capital may be at risk. Past performance is not a guide to future returns.

This communication contains information on investments which does not constitute independent research. Accordingly, it is not subject to the protections afforded to independent research, but is classified as advertising under Art 68 of the Financial Services Act (‘FinSA’) and Baillie Gifford and its staff may have dealt in the investments concerned.

All information is sourced from Baillie Gifford & Co and is current unless otherwise stated. 

The images used in this communication are for illustrative purposes only.

192401 10061942