KPMG is dumping engagement surveys, and so should you

Varun 7 months and 2 weeks ago
share share

A side-note before we begin: Disqovery handles all of the issues outlined in this post. Say hello and we’ll be in touch.

We are on the verge of something big. Managers are beginning to rethink what people data really means and how it is and isn’t useful. You don’t believe me?

Even KPMG is abandoning staff surveys, that old annual corporate ritual. (link, PDF)

This news comes via Robert Bolton, head of KPMG’s HR practice. I highly recommend you read the writeup from AFR. It has some of the most progressive HR thinking I’ve seen from a company of KPMG’s size. I’m going to give you some extra insights into Mr. Bolton’s quotes, and why it’s so important and relevant to managers today.

Why are you measuring data?

“We are entering an era of evidence-based people management.” - Robert Bolton, KPMG

Indeed, why? Too often managers and HR collect data without a clear plan for how that data will be used, and to what end purpose. On the flip side, managers also try their best to make improvements, but how do they know that their strategy is working? Gut feeling? Investments in people are becoming more important, and without hard evidence to demonstrate their effectiveness, executives will balk at anything that disrupts the status quo.

How are you measuring data?

“In reality, engagement is an ill-defined term. And measuring it once or twice a year with some static survey is not very scientific, no matter how much it’s dressed up to appear so.” - Robert Bolton, KPMG

I’m sure you’ve all seen these surveys before. Your company does them once or twice a year, and they typically take 10 to 30 minutes to complete. Have you ever thought about how flawed this approach is?

Is it timely?

The workplace moves fast. Now it moves faster than ever. If you collect data via surveys once or twice a year, that information is already out of date. Don’t go making decisions and measuring effectiveness using old data. You don’t ask someone for their Amazon wishlist 6 months after their birthday, do you?

Of course if you’ve ever designed, deployed, and analyzed a survey, you know well how much time it takes to run through that cycle. That’s why traditional employee surveys cause so much trouble.

Is it accurate?

How many questions on a survey before you stop paying attention to it? How many times can you answer strong agree or disagree before you’ve zoned out? Research has gone as far as to show that respondents had 30% lower answer quality once a survey went past one minute.

Is it scientific?

Let’s pretend for a moment that you get the data you need when you need it. We still face the problem making sure your survey means something.

See, this is the problem that Robert Bolton refers to when he says it is “not very scientific.” And unfortunately, this is where everything falls apart.

For example, if you’re tracking a specific metric, like employee sentiment, over time using semi-annual surveys:

Default caption test

Hmm, according to my surveys this team looks to be consistently good, and gradually getting better. Oh wait, when I measure continuously I see that it’s not that simple.

One of your teams may be going through a particularly tough time at the moment that they receive your survey. Whatever data you get from them is focused only on their experiences in the last week or two. That’s how you lose meaning when you get data once or twice a year.

If you visit New York on a rainy day, you don’t assume that it rains constantly. You must get enough data for your analysis to make it statistically accurate.

What to do?

“It’s a revolution.” - Robert Bolton, KPMG

This is a tricky problem. How much time can you devote to surveying, compiling, and analyzing data? Surveys are boring, and getting good data out of your team is hard. If we truly are in the era of evidence-based people management, then the quality of that data is more important than ever.

Disqovery is a tool that uses experience sampling to get a small set of data at a high frequency and high accuracy. We ask a few lightweight questions that take no more than 30 seconds, and we’re focused on exactly what you need to understand your people better and get feedback to improve the workplace. You just have to say hello to get started.

Banner image is credited to Seattle Municipal Archives.