A CEO asked us this on a call, almost word for word. She ran a regional nonprofit newsroom, and she was blunt about what she'd inherited. "There is no culture in that newsroom," she said. "It's a little Wild West." Then came the real question, the one under every one of these calls: "Can you pull it back? Is there a point where it's too late? Are we too late?"
If you're asking that, you already know something is wrong. Maybe a prior culture push fizzled. Maybe growth or a downturn or an acquisition knocked the wheels off. Maybe you've watched good people leave and can't fully explain why. So, the honest answer to whether it's too late, and the more useful question underneath, is this: How do you change company culture when the one you have is broken?
Almost never. The reason people fear it's too late is that they're picturing the wrong fix. They imagine culture as a collection of personalities they'd have to replace, and replacing half your team is impossible, so it feels hopeless.
But culture isn't the personalities in the building. It's the behaviors those people repeat every day. As we often say, culture "isn't about words on a wall, it's about how people act when no one's watching." You can't reprogram someone's personality. You can change what they do on a Tuesday. That's the whole game, and it's why "too late" is the wrong question.
The instinct with a bad culture is to find the bad people and remove them. Sometimes one person does have to go. But most of the time, the problem isn't who your people are. It's that no one ever defined how they're supposed to act, and no one reinforced it.
Take that newsroom CEO. Her real problem wasn't that her journalists were bad people. It was structural: "They weren't doing any evaluations, no performance management, no anything." Nobody had ever named the behaviors of a functioning team or built a habit around them. Firing brilliant writers wouldn't fix that. Installing the behaviors would.
This is the difference between values and behaviors, and it's where most culture fixes die. "Respect" on a poster changes nothing, because no one agrees what it looks like on a Tuesday. Naming the specific, observable behavior ("we respond to teammates within the day, even if the answer is 'not yet'") gives people something they can do, coach, and repeat. That's how to change company culture without trying to change the humans: you change the actions, and the actions compound into a new normal.
If cultures could only decay, "too late" would be real. They don't. Culture is directional, and it moves both ways.
Consider a consumer-lending company we talked to. For years they'd won Best Place to Work in their city. In the founder's words: "We had a great culture and people came into work and it was fun every day." Then the market turned, volume collapsed to a quarter of its peak, layoffs came, and the culture went with it. Nobody's personality changed. The behaviors did. Leaders got lenient because they were scared to lose anyone. As one put it: "I don't need to lose any more people. So we're going to give people a little longer leash than we used to." The leash became the culture.
If a culture that good can slip that fast, a culture that bad can be rebuilt. The mechanism runs both ways. And people feel the loss, which means the appetite to fix it is already sitting in the building. An HR leader at a sports-media agency described the same slide (their old culture was "10 percent soft accountability and 90 percent just good vibes") and then said the line that gives every turnaround its shot: "the entire company is waiting for it to come back." They're not waiting for new personalities. They're waiting for someone to rebuild the practice.
You install a system, not a slogan. Enthusiasm fades by the second week; a rhythm doesn't. The Culture by Design methodology runs the turnaround in four moves, and every one of them is about behavior, not personality:
1. Define the behaviors. Name the specific, observable Fundamentals that drive how your best people already act. Not values. Behaviors, written so a new hire could copy them.
2. Build rituals to practice them. Put one Fundamental in front of the whole company each week. Repetition is what turns a poster into a habit.
3. Roll it out to everyone. Deskless crews, remote offices, every site. A minute a day, not a workshop nobody remembers.
4. Reinforce through coaching. Managers catch people doing it right and name it, in the moment. This is the step that makes it stick, and the step DIY attempts always skip.
None of that requires you to reach into anyone's soul. It's a set of repeatable actions, run on a cadence. That's what a culture operating system is: the mechanism that keeps the behaviors alive after the kickoff energy is gone.
Only two things close the door, and both are choices, not fate.
One: no one at the top will hold the line. A culture rebuild is 90 percent accountability. If leadership won't reinforce the behaviors (or worse, keeps handing out the "longer leash" because confrontation is uncomfortable), no system will save it. The behaviors have to matter, and someone has to make them matter. That's the difference between a bad culture you can fix and one you can't.
Two: you treat it as an event instead of a rhythm. This is the quieter killer. One leader we spoke with described a sticky note on his monitor that read "get yourFundamental of the week out." It "lasted all of a couple of weeks." Another, mid-turnaround, put his finger on exactly what was missing: "there is no structured, defined process where this all fits." A kickoff is not a culture. The rhythm is the culture. Skip the rhythm and you'll be asking "is it too late" again next year.
Notice that neither of these is about how bad things have gotten or how long they've been that way. We talked to a new owner of a sign manufacturing company where some employees had been there 40-plus years and, in his words, the culture "changed based on the way the wind blows." Four decades of drift. Not too late. It was a blank page, not a locked door, because behaviors can be defined at any age of a company.
You change the inputs, not the humans. Define the behaviors you want, make them impossible to ignore by practicing one every week, and coach to them until they're the default. The same people who feel like the problem today become the ones running the new normal, once you give them a standard to hit and a reason to hit it. To improve company culture that has gone bad, work in this order: behaviors first, rhythm second, accountability throughout.
It won't feel different overnight, and anyone promising that is selling you the pep talk. But it moves faster than people expect, because the first defined behavior gives a lost team something concrete to do on Monday. No one drifts to a better culture, the same way no one drifts to better health. The drift took years. The rebuild is deliberate, and you feel the first turn in weeks.
The trap in "is it too late" is that it can talk you into doing nothing, and doing nothing is the one option that keeps costing you. Gallup puts the price of replacing a single employee at one-half to two times their annual salary. Every quarter you wait, you pay for a weak culture in turnover, slow onboarding, and the good people who leave citing "the culture."
One line from that same Gallup research should end the debate: 52 percent of employees who quit say their manager or organization could have done something to prevent it. More than half of the people you're losing were savable. That is not the math of "too late." It's the math of "we haven't started yet."
Usually, yes. Most toxic cultures aren't a people problem, they're a standards problem: no one ever defined how the team is supposed to act, and no one reinforced it. Occasionally one person has to leave, but the fix is installing and coaching new behaviors, not replacing the roster.
It's a rhythm, not an overnight switch. You'll feel the first shift within weeks because a single, defined behavior gives people something concrete to do immediately, but the lasting change comes from practicing one behavior a week and reinforcing it over months. Drift takes years; the deliberate rebuild is faster, but it never happens by accident.
The honest answer to the CEO in the newsroom, and to you if you're asking the same thing: you're probably not too late. You're late. Those are different things. Late means the drift has a head start and the rebuild takes intention. Too late means the door is closed, and it almost never is.
Culture is behaviors, and behaviors can be redefined at any point in a company's life. The only real deadline is the one you set by deciding it can't be done. See the four steps of the turnaround on our how it works page, or have a short conversation with us and we'll tell you where your culture stands and what the rebuild would take.