The comfortable familiarity dilemma
When you stop questioning the limits of the systems you've built, you fall into what I call the comfortable familiarity dilemma, a trap that once held back big tech companies like Twitter and Google.
Software companies often fall into what I call the “comfortable familiarity” dilemma. Once they know their system inside out, they stop noticing its limits. They work under the assumption that they have total freedom to improve or change things, but in reality they’ve already accepted the boundaries and stopped questioning them.
It’s like a video game. At first you try to walk through walls or break out of the map, but once you’ve played long enough, you don’t even bother. You’ve accepted the rules, and inside that box it feels like you can go anywhere. The illusion of freedom comes from forgetting the boundaries. In fact, the phrase “think outside the box” comes from a classic puzzle where you're given nine dots arranged in a square (a 3×3 grid), and the challenge is to connect all nine dots using just four straight lines without lifting your pen.
Most people assume they have to stay “inside the box” of the square, but the only solution is to extend the lines outside that invisible boundary.
Software companies, more often than not, also fall into the same trap. Once they establish routines, assumptions and rules, they rely on them to avoid risk and uncertainty. But innovation always requires both. Twitter is a great example of this. For years, product leaders saw it only as a messaging app and never questioned whether it could also be a platform for writing blog posts, sharing videos, posting jobs, or even hosting online meetings. The 140-character limit was Twitter’s defining feature, and for the teams building the product, 140 became the natural boundary, and once accepted it was rarely challenged. It was not until Elon Musk acquired the company and brought in new product leadership that those assumptions started to be questioned.
In contrast, WeChat treated messaging as just the beginning and grew into an entire ecosystem with voice calls, payments, ride-hailing, food delivery and even utilities. It took WeChat just 2 years to grow from a mobile-first messaging app into a social feed with a digital wallet, and 6 years to become a full super app with Mini Programs acting like an app store. They never saw messaging as a boundary and managed to turn the app into a full ecosystem by constantly asking questions, creating prototypes, testing assumptions and pushing the edges of the system. They succeeded because its product team understood the needs of people using mobile phones and managed to see beyond the limitations of their basic messaging app.
The risks of comfortable familiarity
“The more familiar things feel, the harder it is to change.”
Once product teams get too comfortable, they develop tunnel vision and stop questioning the basics of the product, the users and the market. They start looking for information that backs up what they already believe about the system and ignore anything that challenges it. That’s how teams end up holding on to old features or directions, even when the data is telling them to take a different path.
In the last couple of years, AI has challenged the limits of what’s possible with software and pushed product teams out of their comfort zone. That’s what big shifts in technology do, they expose teams that got too comfortable. Nothing shows this better than Google letting one of its most creative and talented employees walk away, only to realise later how valuable he really was. Noam Shazeer was one of the lead authors of the research paper “Attention is all you need,” which inspired OpenAI to create ChatGPT. He even went a step further and built his own chatbot named “Meena” before ChatGPT existed, but Google considered it too risky for their brand and refused to release it. Frustrated, Shazeer left the company in 2021. Just a year later, ChatGPT was released and Google panicked. When Larry and Sergey realised how comfortable their product leaders had become with their systems and how valuable Shazeer really was, they paid $2.7 billion to bring him back. From that point on, Google Labs has been releasing experimental prototypes one after another to test new ideas quickly, see what sticks with users, and avoid falling back into the same comfort zone that slowed them down with products like search.
How to overcome the comfortable familiarity dilemma?
When I was working at Sky on the innovations team, we asked uncomfortable questions in every brainstorming meeting. Two I remember clearly were: “Should we see Netflix, YouTube and other online streaming platforms as competitors?” And “will our set-top box become obsolete in the future?” These questions were uncomfortable because satellite dishes and set-top boxes were our core products, the technology that was hardest for competitors to replicate. The answers we got from executives almost always came from a place of comfortable familiarity with the systems we already had: “Netflix uses the internet, they don’t have the bandwidth to compete with satellite TV” or “TVs are not platforms, customers will always need a set-top box.” At the same time, we were encouraged by management to think 2–3 years ahead, build prototypes, and demo them internally to influence decision makers. Eventually, TVs became SmartTVs, set-top boxes became apps, and Netflix, Amazon Prime, Disney+, Hulu, Discovery+, YouTube and Twitch became the new controllers and distributors of television content. And Sky, thanks to its investment in R&D, never completely fell behind, but it did have to play catch-up in a market it once controlled. Just to give you an idea, Netflix alone has about 17.5 million UK subscribers, while Sky has around 12.7 million.
The only way around comfortable familiarity is to keep questioning limits and experimenting in the open, releasing prototypes one after another. Google and Sky taught me that the path forward is to test ideas quickly, take more risks and treat comfort as the real danger.
Many companies obsess over metrics, but if they never ask the uncomfortable questions, the data itself becomes part of the comfortable familiarity dilemma.
Are we measuring the right thing?
Why do our users actually choose us?
What are we afraid to change?
What limits do we treat as fixed that might actually be self-imposed?
If a competitor didn’t have our limits, what would they build that we can’t?
Twitter’s belief was that raising the tweet limit would alienate core users, that people would stop using the app if it went beyond 280 characters, not to mention 4,000.
Twitter was comfortable living inside its 140-character box. This slowed experimentation and made the company conservative with product decisions. The limit became part of its identity, a legacy constraint that turned into comfortable familiarity. Over time the team stopped challenging it, leaving space for others like Threads, Bluesky, Reddit, Discord, and Substack to step in.