Startups are, by definition, organizations that are just beginning to form and mature. Culture is not a fixed asset at this stage; it is in constant flux. What works at 20 people rarely works at 50, and what works at 50 often breaks at 150. People join, people leave, more experienced hires arrive, new leaders come in, old leaders move on. A few months can completely change who is in the room and how decisions are made.

Because of this instability, many startups try to “install” a culture program to create some sense of continuity or control. Typically, these programs try to respond to a mix of wishes:

Preserving elements of the early culture that people enjoyed and don’t want to lose.

Emulating companies that look “grown up” and successful, often by copying their values or principles (like Amazon’s leadership principles).

Wanting people to perform at their best and looking to research-backed approaches (for instance, work like Google’s Project Aristotle on effective teams).

These are legitimate motivations. The problem is not the intention. The problem is how culture programs are designed and operationalized inside a startup context.

Who owns culture?

Most of the time, culture programs are led by HR, with sponsorship from leadership. On paper, this sounds reasonable. In practice, it hides a few challenges:

Culture is hard to measure and hard to prove in business terms.

Many HR professionals are learning on the job or are borrowing practices from other companies without fully adapting them.

Many founders and leaders are also operating from previous habits or incomplete models, unless they are serial founders who already scaled and exited a business.

This creates an odd situation: culture is treated as a specialized topic, but often handled without the depth of skill and clarity it actually requires.

The measurement and KPI gap

Business speaks the language of plans and numbers: roadmaps, KPIs, OKRs, revenue, churn, runway. Culture programs rarely translate their goals into this language in a concrete way.

Some questions often go unanswered:

How are we going to measure whether this program is working?

What do we expect to see in terms of hiring, performance, retention or decision-making if the culture changes in the way we want?

Where, exactly, do these values show up—in hiring decisions, in promotion criteria, in how we run meetings, in who we reward?

Without clear links to outcomes, culture becomes easy to deprioritize when other “hard” topics come up. If culture is not in the same operating system as KPIs and OKRs, it lives on the side—nice words, no real weight.

Managers as the real vectors of culture

Because resources are limited, managers end up being the main vector through which culture spreads (or doesn’t). They are the ones turning high-level values into everyday decisions: who gets hired, who gets promoted, what behaviors are tolerated, what gets ignored.

Here’s where things get messy:

Values are often abstract and hard to observe. A manager might notice that someone regularly behaves in ways that feel misaligned with “how we want to work,” but it is hard to make that concrete and objective.

Micro-behaviors—interruptions, subtle blame-shifting, cutting corners, how people talk in Slack—are often too small to justify formal action, but they add up.

Without clear behavioral examples and agreed consequences, managers default to their own judgment.

You can easily imagine a company where every manager has their own interpretation of “what good looks like.” At a small size, these differences are noticeable but manageable. As the startup scales, these deviations compound and grow exponentially. Suddenly, people in different teams feel like they work for different companies.

The exponential drift problem

Because culture is not static, small deviations don’t stay small. When you go from 20 to 50 to 150 people, each new hire arrives into a slightly different local culture, shaped by their manager, peers and current pressures.

If there is no deliberate mechanism to:

Check where things are drifting,

Align on what is non-negotiable, and

Reinforce that in systems (hiring, performance, promotions, rituals),

then the “official” culture and the lived culture slowly separate. At some point, the posters, the values deck and the reality on the ground have very little to do with each other.

Limited capacity and competing priorities

On top of all this, HR and leadership usually operate with limited time and resources. HR is busy with hiring, contracts, payroll, and urgent people issues. Leaders are pulled into fundraising, sales, crisis management, product fires.

Culture work is slow, subtle and hard to credit for wins. When priorities clash, it tends to lose. It is not that people don’t care; it’s that culture rarely screams as loudly as a missed quarter or a production outage.

Even the “good examples” are more complex than they look

Occasionally, we point to successful examples and say, “Look, it can work.” Buffer is often mentioned in this context: a company built around transparency, with public salaries and a strong narrative about openness.

But even there, what matters is not just that they “have transparency as a value.” It’s that this value is wired into actual practices: salary formulas, open dashboards, explicit communication norms. It is not a single culture program installed once. It is many small, consistent choices made over time.