Organizational Culture Drives Down Safety Costs

Organizational Culture Drives Down Safety Costs

A study of 71 manufacturing plants.

Do you ever feel your organization spends more time competing with each other than it does with the competition? That's the organizational culture at many companies. Peter Drucker, who wrote the Theory of the Business, once stated "culture eats strategy for lunch."

This paper outlines the impact that organizational culture had on 71 North American manufacturing plants and the impact it had on costs. It offers an example of how Randy Nicholls, Director of operations of Vari-Form, utilized his plant culture to drive business results around safety, quality and financial performance.

Two organizations with a total of seventy-one plants were examined for this study. Both had applied extensive safety programs within each facility and had also conducted a culture assessment. Yet, even with similar safety initiatives, at the end of the day, the facilities with high performance cultures had better safety records. In essence, Drucker's statement on culture was spot on.

First, let's look at how one leader leveraged culture to boost performance. Vari-Form, a manufacturing plant in Strathroy, Ontario, decided in 2002 they were going to begin actively managing their culture. They began by assessing their culture to get a baseline -- a starting point. They used a culture survey with a benchmark database that compared their scores to 1,000+ other organizations. Their culture scores, on average, place them at the 28th percentile within the database. This rating put them below average in terms of culture performance and with that, a baseline was established.

The change initiative started with the leadership of the plant. Randy Nicholls, the Director of Operations, began having monthly breakfasts with the employees who had birthdays that month. his became an opportunity to get to know the people who worked for him and to get ideas on how the plant could run more efficiently. At first, suggestions were simply what snacks were stocked in vending machines, types of sodas and the location of the smoking areas. But, over time, as the leadership and the hourly began to feel more comfortable with each other and trust began to develop, the suggestions began to turn toward plant efficiency. They also created newsletters and held regular town halls to aid in the communication process. They developed a leadership institute, cross-trained employees and made sure everyone understood how their job impacted the success of the organization.

That was then, this is now. In 2002, they had two plants in one province. They now have five plants in three countries and have doubled revenues and tripled the number of customers. They now either meet or exceed their profit plan. They are 3 million hours without an LTA, ppm is 2, and are best in commodity group for delivery. Their culture scores went from the 28th percentile in 2002 to the 86th percentile by 2010. They now score higher than 85% of the organizations in the benchmark database.

What happened with Vari-Form is consistent with the study where culture surveys were administered to seventy one plants. In addition to the surveys, each plant also supplied Lost Time Accident (LTAs) and recordable cases. The correlation of the culture results to the recordable rate was a -.4225 (p-value of .00012). The figure below shows the plants ordered from low culture scores (left) to the highest culture scores (right). The recordable rate is then plotted out by plant (see figure One).

Figure One: Recordable rate plotted against plants

The data reveals that as the culture scores improve by facility, the recordable rate drops. What is more interesting is when the facilities are divided up into four groups based on where their culture scores fall within the culture benchmark. The benchmark, made up of 1,076 organizations is made up of high, medium and low performing cultures. The seventy one organizations were then divided into four groups: group one were those who were in the bottom quartile of the database (under the 25th percentile), group two was between the 25th and 50th percentile, group three was between the 50th and 75th percentile, and group four was greater than the 75th percentile. To put this into perspective, if an organization scored in the 80th percentile on the survey, it means they scored higher than 79% of the 1,076 organizations -- which is great. If a company scored in the 10th percentile, then 90% of the organizations scored higher then them -- not so great.

So, if you break it down by quartiles or where the cultures fell in the database of high and low performing companies, those in bottom quartile (group one = <25th quartile) in culture performance had an average recordable rate of 2.2 whereas those in the top quartile (group four = >25th percentile) had an average recordable rate of .75. Furthermore, almost 50% of the plants in the upper quartile had a zero recordable rate compared to the bottom quartile who had just 10% with zero recordable cases.

The data also shows a relationship to Lost Time Accidents. Group four, where the facilities scored above the 75th percentile, found 87% of the plants with zero lost time accidents (see Figure Two). Group three had 60% without any lost time accidents. Group two had 40% without any lost time cases. Group one, where the facilities fell below the 25th percentile, only had 33% of the plants with zero accidents.

Figure Two. Percent of Plants with zero LTAs

Finally, what are the cost implications? On the OSHA website, there is a simple cost calculator which calculates the cost of recordable cases and lost time accident cases. It takes into account both the direct and indirect costs of accidents. Group 1, which scored in the bottom quartile of the database had costs estimated at $2,618,000, group 2 was $1,771,000. Group 3 was $518,000, and group 4 in the upper quartile was $259,000. If you simply divide them into above and below average cultures, the above average cultures cost you $777,000 and the below average cultures cost you $4,389,000.

One important component to creating a high performance culture is when the decision is made to manage culture instead of letting it manage you. The Denison Organizational Culture Survey, used in both the Vari-Form case as well as the safety study, quickly identifies deficiencies and areas of leverage with the organization. But, like most surveys, it isn't just the data that creates the action toward a high performance culture; it is the conversations that get started because of the data.

The head of one of the organizations I worked with commented about the conversations that arose after seeing the results of their culture survey. There were things that surfaced within his organization that he had no idea existed. Had it not been for the survey, they would never have been addressed. By understanding the issues and talking it over with his leadership team, he could then take action. This began the first step in opening up a channel of communication with his organization of 800 employees.

Like Vari-Form, the higher performing cultures in the study engaged employees. The communication was not only top down, but bottom up. They found ways to seek out input from employees on safety and production processes. John C. Maxwell, an internationally know expert in leadership, once said, "People don't care how much you know until they know how much you care." A common theme with Vari-Form, as well as the seventy-one plants, was they spent more time engaged with each other whether it be monthly breakfast meetings or town halls. The leaders were visible on the floor. They were communicating with employees. By doing so, they created an environment where both leadership and the hourly were aligned toward a common goal. In essence, they spent less time competing with each other and were now competing -- with the competition.

Jay Richards is a Partner at Denison Culture, a provider of organizational culture and leadership development solutions.

TAGS: Talent
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.