How to Have the Most Successful Culture

We all know the power that culture has in a workplace. Strong, healthy cultures create more than just happy employees. They foster a more successful and adaptable company that…

Wait a minute.

How do we know that?

I mean, it sounds true. It feels true. We’ve all worked in places with a toxic culture that have circled the drain. But how do we know that isn’t selection bias? Do we know that a good culture makes a successful organisation? Is there data to back up this claim?

There is. In the mid-90s, James Baron and Michael Hannan decided to look into it. The consensus was that culture was important but people weren’t sure. It seemed logical that culture matters but they wanted the numbers. They needed a large sample of companies to study, to track the rise of the elite and the demise of the rest. So they headed to Silicon Valley.

Read along with me by grabbing the paper: https://cmr.berkeley.edu/documents/sample_articles/2002_44_3_4776.pdf

The Stanford Project on Emerging Companies (SPEC) analysed start-ups. The goal was to map each company’s culture, plot its trajectory and analyse the results. It won’t surprise you to know they found some interesting trends.

What did SPEC measure?

Culture is an amorphous creature. Each group of people, day by day, has a unique set of rules and principles. Some founders had clear visions on culture, some didn’t. It’s a messy, human thing to boil down into firm numbers. Nonetheless, patterns did emerge. They analysed each start-up along three main dimensions: attachment, coordination/control, and selection.

  • Attachment looked at ways the company enticed employees to join and stay. Monetary compensation was one obvious factor. So was interesting, meaningful and challenging work. Some companies went further, seeking to cultivate a vibrant community. The researchers used the term ‘love’. Employees of start-ups have career mobility, so earning genuine affection makes sense.
  • Coordination/Control looked at how the organisation policed behaviour and achieved results. Some companies had formal processes in place; others relied on group norms to set the standards. Some trusted talented people would work hard; others micromanaged every little detail.
  • Selection looked at how new employees were hired. Were people sought out based on their skills or their compatibility with the company culture? Was it for their long-term value or immediate usefulness?

They found 36 different combinations of these three attributes. However, most companies clustered into one of five common “employment blueprints” (see Figure 2).

The Five Blueprints

No model is perfect. However, they found that most companies fit fairly well into one of five broad blueprints:

  • Engineering companies lure skilled workers with the promise of interesting work. These employees come together as teams that moderate behaviour and value results.
  • Star companies focus on hiring the best people, then staying out of their way. The organisation hopes these elite employees will stick around, bringing up the value of the company for years to come.
  • Commitment companies value the employee above all else. People are hired primarily on being a strong cultural fit. They stick around because of affection and loyalty to the company.
  • Bureaucracies hire people for their skills and offer them challenges. Acceptable behaviour, strategic direction and opportunities for growth are enshrined in policy.
  • Autocracies attract employees by offering high pay. Employment is seen as transactional – an exchange of skills and time for money. Employees are closely observed and tightly controlled. Company direction is dictated from on high.

These categories aren’t comprehensive. Almost a third of companies studied didn’t fit cleanly with any blueprint. Still, these labels make a lot of sense. Engineering companies are a lot like stereotypical start-ups. Bureaucracies are common. I’ve had the misfortune to work in an autocracy before (and the pay wasn’t even that good). Industries that require years of training – think law firms or medical research clinics – often follow the star blueprint. As for commitment companies, they might be the other stereotypical start-up, with baristas, dog-walkers and a strong sense of mutual loyalty.

What causes culture?

The leadership. When the CEO of the company changed, chances are the culture changed too (see Figure 5).

This isn’t surprising. People look upwards and imitate what they see. Sometimes this is a deliberate attempt to appease the powerful. Sometimes subtler psychology is at work. Whatever the reason, the data shows that culture flows downhill. It is up to the leadership to enact the culture. This is one task that can’t be delegated.

You’d think that the employees would shape the culture too. And they do, of course, though less than you’d think. I can point to organisations that transformed their culture by changing management. But perhaps that’s a subject for another blog post.

What’s the best culture?

All cultures have value. It’s like apples and oranges. Sometimes you need the firm hand of an autocracy. Who can really say which is the “best”?

Haha, only kidding. Have some data.

Of course, it does matter how you define success. In the cutthroat world of business, merely surviving is an impressive achievement (see Figure 6).

There’s a clear loser here and it’s not surprising. Autocracies are only as good as the person at the top. Commitment and star blueprints, though, are harness every employee’s strength. It’s not surprising that a company endures when it invests in people or hires superheroes.

It’s interesting how poorly Engineering did, especially when compared with Bureaucratic. One could speculate that bureaucracies are stable without being oppressive. Maybe that extra stability helps them weather storms.

Another benchmark of start-up success – the all-important initial public offering (IPO) (see Figure 7).

Wow, go Commitment. You’re killing it. And, huh, Bureaucracy wasn’t terrible again.

Speaking of IPOs, there’s an interesting quote talking about the risks of focusing on the wrong thing:

I automatically ding anyone who comes in here pitching their business plan if they tell me that their goal is the IPO. If that’s their goal, there are going to be huge organizational problems down the line. An IPO might be a means to an end, but it shouldn’t be an end in itself.

How about growth rate (see Figure 8)?

Another loss for autocracy. Heavy is the growth rate that wears the crown. Interestingly, the Commitment model falters here. It’s hard to pin down why. My WAG would be that Commitment organisations invest in their employees at the expense of growth. It’s a long-term play with short-term costs, which explains how well it does by other measures.

There are a lot of conclusions to draw from the data. The researchers conclude that the Star and Commitment blueprints are the best. Star blueprints are a gamble – as the graphs show, they are less likely to succeed but the payoffs are better.

meme
I don’t always succeed but when I do, I make billions.

The Commitment model offers a safer bet with long-term payoffs.

Can the culture change?

We’ve seen above that culture can change. SPEC shows some interesting results from organisations that change their blueprint.

Fill in the blank: “Over time, organisations tend to become more _______”. Then check Figure 4.

If you guessed ‘bureaucratic’, congratulations, you are not (necessarily) cynical.

Should cultures change? Now, that’s an interesting question. The researchers note that change is disruptive. Changing your blueprint can drive away (or at least alienate) employees that liked the old style. Losing corporate wisdom is a painful outcome. They conclude that:

  • Changing blueprints is at least a little bad. Change leads to at least some turnover. Turnover correlates with slowed growth and higher risk of failure (even when adjusting for other variables).
  • Changing from a blueprint outside the five main types has low turnover. So does changing from an Engineering blueprint.
  • Changing to a Bureaucracy or Autocracy has a lot of turnover. Changing to a blueprint outside the five main types is even worse.
  • Change is smoothest when new leadership drives it.

What does this mean for your organisation?

Let’s assume that these findings apply to you. That’s not something we can take for granted – you might not be a start-up, for example. Still, the more you learn about organisations, the more you realise the work on similar principles.

Reading through all the data, the Commitment blueprint stands out as the strongest. It has its flaws, including some weird stuff around gender diversity. (The researchers note that Commitment culture values people that share the company’s values. This can emphasis ‘fitting in’, which can become ‘hire people like me’). But for long-term stability, retention and agility, Commitment is tough to beat.

If you are starting a new organisation, build it along Commitment or Star lines. If you are taking over an organisation, now’s your chance to reform it.

Let’s say that you’re not in a leadership position. What then? Then the SPEC findings can’t help you. Well, you can share these observations with your leadership. I have. But beyond that, you can’t apply them. One of the results is that culture flows downwards.

Luckily, there are other models.

If you aren’t in a leadership position but you are hungry to drive change, have a look at Rebels at Work. The handbook is for people like you because it assumes you have limited direct power. As you will see, though, your actual power is significant.

All models are wrong. The key is to see where they stop working and find a different one that will help you.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s