“I propose to raise a revolution against the lie that the majority has the monopoly of the truth”
Dr. Stockmann, in Henrik Ibsen’s play An Enemy of the People
When Dr. Stockmann discovers that the beloved spas of his small Norwegian town are poisoned, he assumes that everyone will be grateful. Instead, the town gathers against him to hide the truth and keep the tourists coming. Stockmann ends the play alone - righteous and ruined.
The stock market often resembles Ibsen’s Norwegian town. At any point in time, it is ruled by a short-term focused, profit-maximising momentum herd. Meanwhile, the minority in the market always feels…righteous and ruined.
No prize is bigger than winning the battle for the best idea in the investment marketplace. And the highest returns will come to those that create their thesis in the minority and later manage to convince the majority of its brilliance. This struggle has fuelled countless investment folklore. Graham held out hope that in the long run the best idea would always win. Keynes took that hand away by observing that the markets can “remain irrational longer than you can remain solvent”.
Both concepts assume that we have a chance of knowing what is a right or wrong. Karl Popper pioneered the idea of an unpredictable future. His works inspired the famous investor George Soros to add his theory of reflexivity to the same investor folklore. Soros thought of the market as a constant battle between different interpretations of reality and that those interpretations impacted people’s actions, and thus the reality itself: “The prevailing view is bound to be biased, and that bias can influence the course of events”, George Soros, The Alchemy of Finance.
Over the last weeks, the AI majority herd has dethroned the long reigning SaaS and marketplace majority. After more than a decade of casually reporting Rule of 40 performance (% growth + % EBITDA margin) while enjoying ~10x revenue multiples, valuations have collapsed in only a couple of months. The new rule is questioning the realism with which software investments have been made and the sanity of the sector’s rising leverage.
The rise of the offended voice of the disgruntled minority is why you may be drowning in investor updates saying that there is nothing to worry about. There is a chance these will be right. But this will not depend on any investor reporting. It will depend on the actions the disgruntled minority decides to take.
Owners and managers of the dethroned generation of tech companies are mistaken if they think share prices are falling because of what the new technology brings to the table. For most of them, share prices are falling because of what they have stopped bringing to the table themselves. When most companies were founded, they were launching revolutionary products to underserved customers. They probably priced them cheaply to reach scale as fast as possible, which created benefits for customers in two ways – growing network effects with other users and increasing scale benefits in product development. Their flywheel was spinning faster and faster.
At some point their mission drifted away from their customers to secondary priorities. They started to raise prices without improving the product, they invested in the workplace without demanding a similar increase in efficiency from the employees. When customers were no longer super happy, they stopped measuring NPS and instead measured employee happiness and posted ‘workplace of the year awards’ on LinkedIn. The board and shareholders awarded the finance department free rein to increase leverage and pay more dividends.
Business is an unforgiving game. Schumpeter’s model for creative destruction was written with companies like that in mind. The average life span of a listed company has decreased from 35 years to 15 years over the last half century and only ~1% of the listed companies will deliver higher returns than the treasury bond yield over time (1990-2018). Complacent companies are waiting for a new technology, a new market, a new organisation form to take them out.
As the new battle lines are drawn up, nothing is pre-destined. The new ruling narrative is draining the old sectors of financial and human capital and customer attention, which can make the narrative self-fulfilling. On the other hand, the new paradigm creates a crisis in the old sectors that creates a catalyst for change and makes the cost of it lower. It allows the best companies to come out of the challenge stronger than before. For these companies AI may be the greatest gift ever as they are allowed to turn their ‘10x better than average’ engineers to ‘1000x better than average’ engineers before any new start-up catches up.
Meanwhile, those companies that spend their time hoping and arguing that the new majority is wrong will soon be added to the large and growing statistic of underperforming and failed companies.
The German incubator Rocket Internet was mass producing code and companies long before vibe coding was a term. As soon as an internet model took off anywhere in the world, the Rocket machinery started cloning it. Everything that was difficult with building a tech company was made easy, almost automatic, by its centralised team of experts in Berlin. The back end and front end (nicknamed Bob and Alice) were the spine of the product development. The design was developed to align as closely as possible to the original. Some of its genius rested with marketing. Rocket pioneered search engine optimisation and search engine marketing before it became mainstream and it ran more aggressive marketing campaigns earlier than almost all its global peers.
In its first decade (2008-2018) Rocket was one of the most successful incubators ever. At its peak it employed almost 50,000 people across 200 companies operating in almost 100 countries. Only three of these companies have achieved sustainable scale and profitability: Zalando, Delivery Hero (the result of a merger with one of Rocket's food delivery companies) and Hellofresh. The three were IPO'd with a combined value of EUR 11bn. Only a couple of years later their valuations peaked at EUR 60bn.
What did those three companies do that all the others did not manage to achieve? Interestingly, the three success stories were the ones that either were never built on the centralised Rocket platform or graduated from it early. They all built their own in-house tech teams and invested heavily in strengthening a deeper unique offering, tailor-made for their customers' needs. They invested to be the best at everything from the broadest supply to the most efficient fulfilment.
The battle between Airbnb and Rocket’s clone, Wimdu, is a similarly telling case study. Airbnb was founded in 2008 and hit its unicorn milestone in 2011, backed by Sequoia and a16z. A unicorn round, backed by leading Silicon Valley VCs, was what made a company appear on Rocket’s ‘cloning candidates’ radar screen.
Rocket quickly saw the potential of the emerging model and raised one of its largest seed rounds ever - EUR 90m to launch its clone in record time. Only a couple of months after birth it already had 400 employees and an almost global coverage. It even launched a Chinese site next to its international site to cement the network effects. After this show of offensive strength, Rocket called the founders of Airbnb to offer a truce in the form of a merger.
“We all just kind of looked at each other and said, “Uh-oh,”’ Brian Chesky, co-founder of Airbnb describes it in the book The Upstarts by Brad Stone.
“It was pretty impressive.”
Coming from design rather than execution, there was not a doubt in his mind that Airbnb had met its superior in efficiency and speed. The commercial benefits that would derive from a merger were obvious. But there was one major obstacle that made Chesky hesitate: Rocket didn’t share their values, their design sensibilities, or their desire to build a close-knit community.
Weeks after Rocket launched its assault, these values were tested when an Airbnb rental was trashed. The property owner went public, lambasting the company's weak response on social media. The story went viral. The community Airbnb was trying to build began to fracture.
Chesky realised that he had to act decisively to save his company. He introduced a 24/7 customer service hotline, doubled the size of the customer-support staff, created an in-house trust and safety department that was separate from customer service and focused on longer term measures. He introduced a reimbursement guarantee of $50,000 that would soon grow to $1 million. On the technical front, Airbnb started to verify users’ identities and ask for their social media profiles. The response to the crisis strengthened the trustful community that would become a cornerstone of Airbnb’s success.
Wimdu was never built with a customer-centric purpose. It was built for shock and awe and to grab as much market share as possible while competition was still paralysed. When it failed with its first assault it ran out of stamina. A couple of years later it folded and ceased operations. Meanwhile, Airbnb went on to build one of the largest and strongest marketplaces in the world. Since hitting its unicorn valuation and winning its fight with Rocket, it has 75x'd its valuation.
Culture alone did not save Airbnb. Chesky knew that he also needed to match his competition in speed of execution. While in Berlin to meet Rocket, Chesky also met with another manager of similar calibre to Oliver Samwer, namely Oliver Jung. Jung had grown up in the same German start-up factory mould and had even co-operated with the Samwer brothers in the early days of their respective company cloning careers. He knew intimately the force that Airbnb was facing and how to get the same agility and efficiency into its more carefully built culture. The combination of efficient execution and customer-centric passion proved extremely successful.
What does this have to do with an AI update? The point we are trying to make is that innovation and aggressive competition is nothing new. We have seen rapid launch of complex world class code and dramatically more efficient and aggressive modus operandi before. It may be impressive but that in itself is not a recipe for long-term sustainable, great company building. Competing in the era of AI means building the most efficient execution machine to deliver your core values faster and better than before. Being good at only one will not be enough to build a sustainable winning organisation.
“I very frequently get the question, ‘What’s going to change in the next ten years?’ And that is a very interesting question; it’s a very common one. I almost never get the question: ‘What’s not going to change in the next ten years?’ And I submit to you that that second question is the more important of the two because you can build a business strategy around the things that are stable in time. When you have something that you know is true, even over the long term, you can afford to put a lot of energy into it.”
Jeff Bezos
We hold ourselves to the same standard. We have always believed in leverage through code - that individual work gets amplified by technology. We built our tech platform, SPINE, for this reason. Recent advances in AI have let us go much further than before. We use LLMs to extract and assess company characteristics against our investment checklist and feed previously untapped data into our automatic scoring. Our data platform merges and summarises information from sources that used to be fuzzy and disconnected but are now becoming useful. We run voice-of-customer studies across languages and markets in hours, not weeks, without large consultancies. We are doing more with less, more consistently.
Meanwhile, our ‘value perspective on tech investing’ naturally selects for both customer purpose and operating efficiency. We believe those core values were as true when we started Sprints as they are today and as they will be ten years (or more) from now.
But our book of misery - the internal nickname for our investment checklist – is continuously collecting learnings from the ever-changing markets. Just like search did with Google or mobile did with the iphone, AI will bring a new business logic that we need to understand and adjust for. Such as focusing on software products with a system of record containing scarce and valuable data (like Modular Finance and Flightradar24). We look for companies that cultivate network effects and positive feedback loops; great products that get harder to displace the more they're used, whether by humans or agents. Furthermore, we evaluate product and tech quality not only on its ability to solve problems well today, but whether they make sense in an agentic era and if the tech teams can evolve into developing at the new pace.
The ideal management team we look for is a Chesky-Jung combo (not Bob and Alice). The ones that can simultaneously disrupt their own organisation and stubbornly double down on the company’s core values.
Our portfolio company Intercom, for example, has reinvented itself with AI product Fin which handles customer support tickets from start to finish. This allows massive efficiency and cost savings. Instead of charging per support agent seat (~$100/month whether they're working or not), they charge $0.99 each time Fin completely resolves a customer's problem. A human handled resolution can cost ~$15-30 vs. Fin which does it for $1 with consistently strong Customer Satisfaction Scores (CSAT) levels.
Intercom captures a fraction of the value, customers pocket most of the savings and don't have to manage employees. Both win. But the customers win more, which is a priority totally aligned with Sprints’ culture. The customers pay back with their loyalty. Fin AI agent is getting close to $100m ARR standalone business, growing ~3x YoY.
Although AI is a change of a calibre we rarely see, the counter measure is going to be more of the same that defines great companies - doubling down on building for your customers while constantly upgrading the quality and efficiency in the organisation. We don’t think we will ever live to see the day when ‘are the customers happy?’ is not the starting point for our company analysis. Why else would anyone want to run a business, regardless of the technology used? Looking at the rate with which we currently on- and offboard new AI products, this emerging industry has some way to go.
Another thing the best performing companies have in common is that their share prices are often very volatile. They are constantly questioned and dethroned to minority position by the markets. For the best companies it is a blessing in disguise - they never waste a crisis. It is much easier to change when you are out of power than when you are sitting on the throne. A way to think about it is that the most purposeful companies survive almost any crisis if they have a clear enough ‘why’, to paraphrase Nietzsche.
In his 2005 letter to the shareholders of Berkshire Hathaway, Warren Buffett described the winning mindset in one of the best ways we have seen:
“Every day, in countless ways, the competitive position of each of our businesses grows either weaker or stronger. If we are delighting customers, eliminating unnecessary costs and improving our products and services, we gain strength. But if we treat customers with indifference or tolerate bloat, our businesses will wither. On a daily basis, the effects of our actions are imperceptible; cumulatively, though, their consequences are enormous. When our long-term competitive position improves as a result of these almost unnoticeable actions, we describe the phenomenon as “widening the moat.” And doing that is essential if we are to have the kind of business we want a decade or two from now. We always, of course, hope to earn more money in the short-term. But when short-term and long-term conflict, widening the moat must take precedence. If a management makes bad decisions in order to hit short-term earnings targets, and consequently gets behind the eight-ball in terms of costs, customer satisfaction or brand strength, no amount of subsequent brilliance will overcome the damage that has been inflicted.”
Although this modus operandi has delivered a majority of the returns in the market, we have noticed that it is only practiced by a minority of companies and appreciated by a minority of investors, which makes us hopeful that we will be able to continue to deliver differentiated, high returns supported by the continuous evolution of technologies and markets.
Kindest regards,
Henrik and the Sprints team
Milan, 13th February 2026
