How benchmarking your survey helps contextualize results

Your recent employee survey revealed that only 80% of your staff is actively engaged in their work. You’re ready to sound the alarm bells with your leadership team to figure out why a whopping 20% of your workforce is just punching the clock and biding their time until they can move on.

Our data shows that, worldwide, companies see an average engagement rate of just 70%. When you have that as your baseline, it’s easier to see that you’re actually outperforming others – and maybe it isn’t time to call a fire drill quite yet. 

That anecdote is proof that at its most basic, data is just numbers. When it stands by itself, it lacks meaning. That’s why, with any number or statistic, you need context. 

Context provides additional detail and helps paint the picture of what a number communicates. One way to find context is through comparison, and benchmarking is one specific method of comparison often used to understand data. 

When it comes to employee engagement surveys, benchmarking can help you avoid panicking about numbers that seem low. On the other hand, benchmarks can also help highlight where you might want to strategize for improvement. Below, we’ll explore what benchmarking is and why it’s key to understanding survey results. 

What is benchmarking?

Simply put, benchmarking is using a data metric as a point of comparison. The phrase comes from woodworkers and cobblers who placed marks on benches to indicate a shoe size or board length. 

The American Productivity & Quality Center (APQC) describes benchmarking as the process of measuring your key business metrics against competitors or industry peers. It can show you the industry average for a given metric, so you can see how your company stacks up. 

The process of benchmarking can be found in many areas, such as competitor research and price setting. Regardless of the specific scenario, knowing the industry average for what you’re trying to measure helps you compare whether you’re above or below your peers and adjust accordingly. 

Benchmarking examples

One example of benchmarking is the participation rate for staff surveys. If you had to guess, you might think a 100% participation rate in a staff survey indicates an engaged company. However, that isn’t the case.

In reality, we found that a company that had a 100% participation rate actually had one of the lowest engagement scores. When we drilled down, we found that 100% participation can indicate mandatory participation – which often equals bad survey results. When the survey becomes something employees have to check off their to-do lists, they’re more likely to select all “strongly agrees” or “strongly disagrees” just to get the task done. 

The actual benchmarks for survey participation range from 65% to 90%, depending on your company’s size.

Another example is Culture Amp itself. One survey we posed to our employees asked them to rate their agreement with this statement: “When it is clear that someone is not performing in their role we do something about it.” 

Our score was 34% for the question. Taken as a raw number, 34% seems low. However, the benchmark is 48% for companies comparable to us and 62% for the top 10% of all companies. Without those comparisons, it’s hard to make sense of what the data is telling us about the response. 

Why is benchmarking key when discussing engagement survey results?

The famous quote by Andrew Lang, “Most people use statistics like a drunk man uses a lamppost; more for support than illumination,” comes to mind when thinking about why benchmarking is necessary for understanding engagement survey results. 

With the right benchmarks, you can discover where you stand within an industry, compared to your competitors. As you learned from the example above, numbers don’t mean much out of context — but you can provide some much-needed illumination by using benchmarks. 

These benchmarks give you a point of comparison and data about what might need improving. Your results compared to benchmarks can also indicate how hard a problem will be to change. While benchmarks won’t give you a strategy in and of themselves, they do help inform you of what’s going well and what may need improvement. 

How to set employee engagement benchmarks for your company

When setting benchmarks for your company, you’ll want to think about relevancy and stability. In other words, you’ll need to consider your sample size and choose the right companies to include in that sample. 

For sample size, we recommend looking at 20 to 25 companies, with a minimum of 20,000 total people. Large sample sizes help avoid data skewing in a certain direction, and this is the number where we’ve found that the spread of scores stabilizes.

Besides sample size, you’ll need to consider types of companies. In general, you’ll want to use companies that are similar to yours in order to have relevant results. For example, if your company has less than 50 employees and is in tech, you likely wouldn’t want to include an enterprise-level company that is in consumer packaged goods. With companies that are too different from yours, you risk creating a benchmark that isn’t relevant to you. And as the old saying goes, that’s like comparing apples to oranges. 

That said, it can be helpful to broaden your focus a little bit. That means moving outside of identical companies to ones that may have fewer elements in common with yours, but enough overlap that including them in your benchmark makes sense to help you get where you want to be. 

Along those lines are competing companies. For example, do your exit interviews show that a good portion of your employees is hitting the road to work for your competitors? Including competitors that your company admires (or is losing out to) is another way to create a robust data set for your benchmark. 

Lastly, if there’s a company that you consider an exemplar of what you hope your company will achieve, including them in your data is also a smart idea as it can give you a more ambitious benchmark to work toward.  

A quick word of warning: Assembling benchmarks can be time-consuming, difficult, and often expensive. A way to offset this is to use a shared database or collective intelligence.

Chasing the perfect benchmark is impossible

The peril of analysis paralysis is very likely when looking for ideal benchmarks. After all, you might spend hours researching other companies and still not find enough that are perfectly comparable to yours. That’s normal. You won’t be able to find perfect matches as every company is unique. 

The important thing is to use a variety of benchmarks and companies to create your benchmarking data set and recognize that your ultimate goal is to use your engagement numbers to inform actionable goals. 

Remember, at the end of the day, you’re trying to improve your employee experience. Numbers only go so far, so listen to your employees. Their insights and opinions should inform your strategy and are often even richer than data points alone.