(DE) Patterns over Hype: Why the Tech Industry Systematically Forgets its Most Important Lessons

Read the full article in German on Midrange.

‍

In the innovation and investment business, looking back is considered a weakness. The unspoken doctrine is that those who drive in the rearview mirror have already missed the next big thing. The focus is always on the future – on the next buzzword, the next disruption, the next valuation round. But this very fixation on the future carries a systematic risk that materializes with alarming regularity: roughly every 15 years, the entire industry suffers from a collective amnesia.

The pattern is robust and historically verifiable. Roughly every two entrepreneurial generations, familiar concepts reappear under new names, known mistakes are repeated without any discernible learning curve, and historical warning signs are routinely ignored. Particularly where technological enthusiasm meets a surplus of capital, the transfer of experiential knowledge from previous technology cycles reliably fails.

There is ample concrete evidence: The stock market blind pool of the early 1980s mutated into the blank-check company of the late 1990s, experienced its next incarnation as an ICO in the crypto age after the bursting of the dot-com bubble – and returned to the front pages a few years later as a SPAC. In the 2000s, investors stubbornly ignored, when building the first cloud-based SaaS models, those key performance indicators for customer retention and capital efficiency that had already proven costly for telecommunications companies in the 1980s and 1990s. And when drone-based air taxis were celebrated as an urban mobility revolution from 2014 onwards, hardly anyone seemed to remember the spectacularly failed CargoLifter project of the Neuer Markt (New Market) – along with its already almost insurmountable regulatory hurdles.

Even today, some deep-tech promises reflect more enthusiasm for what is technically feasible than a historically informed, sober planning for the future.

Structural convergence instead of singular breakthroughs

Those who take something away from this story gain a crucial analytical advantage: looking back is not nostalgia, but pattern recognition. And the most important lesson is this: profound, sustainable innovations almost never arise from a single source.

This has been true since the beginnings of the semiconductor industry. The breakthrough of the internet in the mid-1990s was not a product of chance, but the result of several parallel developments: a modern transmission protocol from DARPA research, finally deregulated and massively expanded fiber optic networks, millions of personal computers available – and finally Netscape's browser as the catalyst. The iPhone followed the same logic: not a sudden inspiration, but the logical convergence of existing technologies and market conditions – a mass audience already accustomed to mobile phones, sufficiently powerful second-generation mobile networks, and, for the first time, a practical balance between mobile computing power, display quality, and battery life.

It is such structural convergences that give rise to genuine innovation platforms – foundations upon which sustainable value creation becomes possible for many years to come. These genuine paradigm shifts contrast sharply with one-dimensional trends: driven by an isolated technological breakthrough, a regulatory tailwind, or short-lived market euphoria. They may begin spectacularly, but sooner or later encounter strong headwinds.

Such hype cycles are often based on a self-reinforcing investor dynamic: a single spectacular exit is enough to drive entire industries into overfunding. The MEMS euphoria following some extraordinary optical networking exits in the early 2000s and the next-generation solar cell boom after Q-Cells' IPO illustrate this logic, as do the regulatory-fueled boom-bust cycles in climate tech and new space in the recent past. Even the currently burgeoning new defense sector has so far primarily rested on a political foundation: sharply rising defense budgets. This alone is not sufficient evidence of convergence.

AI: When the hype is exceptionally justified

The current wave of artificial intelligence is remarkable in this context – precisely because it represents a rare exception. It rests on an unusually clear multidimensionality: Decades of refined mathematical models from academic fundamental research meet exponentially increased computing power, now made affordable by GPU cloud services – and training data of a scope and diversity that only the scaled internet has made possible. None of these strands alone would have sufficed. Only their convergence from around 2020 onward transformed years of AI promises into tangible disruption.

The underlying lesson is old, yet it is repeatedly forgotten: Only when technological developments in hardware, software, and market structure converge in an economically viable way does a new innovation platform emerge. Not before.

Aiming for the right moment

For market participants – whether investors, technology providers, or corporate clients – timing remains the central strategic discipline. Hunters speak of leading the shot: aiming is not at the target's current location, but at the point where it will be when the bullet arrives. Those who jump on a trend when it's already making headlines are aiming behind the target – they're investing in a valuation, not a future.

...

Read the full article in German on Midrange.