Should you offer staff bonuses for NPS improvements?

I’ll get right to the point: change management is cheaper than bribing employees. It’s more effective, too.

The CEO of a client organisation recently asked us about offering a bonus tied to their Net Promoter Score (NPS). He believed that this would motivate employees to think more about customers and improve their experience.

Without mincing words, I told him I thought that was a terrible idea – one that is universally frowned upon.

He pushed back: “Why? What’s the evidence? Other organisations are doing it. Why is it so bad?”

I’ll admit that these questions caught me by surprise. The answers seem so obvious that I never truly considered the issue. And outside of the “used car salesperson” example, where the argument is made that employees will game survey results, I don’t know that there really is a lot of hard data on the topic.

However, there are a ton of opinions about the practice, and many companies (including most of our clients) do it.

Getting to the heart of the matter

I decided to ask my LinkedIn connections to explain this disconnect, where thought leaders universally say that paying employees for improved survey scores is a bad idea, yet so many companies are doing it.

Are thought leaders right, and those companies are making bad decisions? Or are companies right, and the “damage” caused is minor and offset by the gains of saying a focus on customers (represented by survey scores) warrants investments in bonuses?

Unfortunately, most of the responses didn’t address my core question. Instead, respondents shared their viewpoints as to whether companies should or should not provide these bonuses, as well as their specific advice.

I would love for someone to try to collect the data on whether a company pays these bonuses and what impact doing it or not doing it has on their survey scores. My suspicion is that there won’t be much difference, but we don’t have the data to prove or disprove that at this point.

So until then, here’s my own analysis for why this disconnect exists:

• Intuitive benefits. Behavioural economics shows us that incentives work. When properly incented, whether with pay or recognition, people’s behaviour changes. When I was with Best Buy, any metric put on the scorecard – product warranties, reduced shrinkage, accessory sales, you name it – showed improvement, even without direct payments made to employees. Eventually, the scorecard included so many items that it lost its impact. Regardless, incenting for activity works in many situations.
• Hidden downsides. Gaming – for example, manipulating who receives surveys or what their responses are  – is a common concern. Beyond asking for a top score, teams can delay negative communications until after a survey. Or they can accelerate positive communications so that they hit during the survey. In most cases, this gaming is subtle – so subtle that executives don’t see it, so it doesn’t factor into their decision-making. Also, if incentives lead to bad short-term behaviours, it’s difficult to link that back to the bonus programme. If teams, for example, push out flawed products just before the survey goes out in order to meet a client’s deadline, the connection will be even harder to find.

In giving bonuses based on overall survey results, companies are assuming that 1. teams actually know what to do in order to improve customer satisfaction and 2. they won’t do it unless incented to.

Both points in that statement need to be analysed.

First, understanding how to drive customer satisfaction (I’m using this as a catch-all for NPS, customer effort, etc.) isn’t that easy.

When I led a CX programme in the health savings account (HSA) marketplace, I asked teams what they thought the top drivers of satisfaction were. Product thought it was about features. Marketing thought it was about educating on how to use the HSA. Sales thought it was about pricing.

They were all wrong.

The top driver was the ability to log in to the website. Didn’t see that one coming, did you? Neither did they, so there were no projects that focused on this issue.

What can individual employees control?

The only way to connect individual efforts to the top-level score is to discover what the top drivers of satisfaction are and then share this information across the organisation. Only once you’ve done that discovery work does it make any sense to incent employees on these behaviours – which they can control – rather than on the outcomes, which they can’t.

While the findings are about employee engagement, not customers, Gallup found that when companies incent on employee engagement scores, managers start acting really nicely around the time of the survey, which leads to bad behaviours.

But when the incentive question was switched to, “My manager took action based on the last survey,” the gaming stopped, because they couldn’t fake their way past that one – which is a good outcome!

So that’s my first recommendation: if you’re going to incent on customer survey questions, focus on action-oriented ones – the equivalent of improving the login process. Find items that tie to customer needs and focus your incentives on that.

Second is the assumption that employees won’t act unless incented to do so. In the LinkedIn discussion I began about this issue, Forrester’s Maxie Schmidt, Ph.D, described this as “the myth of the coin-operated employee.” (What a great phrase – I wish I had come up with it!)

Employees do react well to incentives, though not always in the way you expect… or want.

Incentives cost more than you think

Monetary incentives are one possibility, but they come with costs beyond actual dollar amounts, as Maxie lays out: “[W]e are talking hours and hours of the CX team and other people’s time to collect and defend the scores, worse employee motivation, upset customers, etc.”

And when the score comes up .1 short…or .2…or .5? You’ve just demotivated your team! And you wind up with worse outcomes in the long run when employees focus on getting things done before the next survey for an immediate benefit, at the expense of long-term efforts for which they won’t necessarily be “rewarded.”

Instead of using a blunt tool such as paying people extra to do their jobs, let’s focus instead on using change management tools to accomplish the same outcomes without the risk of gaming and demotivation.

To illustrate my point, I’ll use my favorite change management model, ADKAR. As a refresher, ADKAR tells us that if you want people to change, they need:

• Awareness of the need for change

• Desire to support the change

• Knowledge of how to change

• Ability to demonstrate skills and behavior

• Reinforcement to make the change stick

When building a program to incent employees around survey scores (Desire), hopefully you’re also working on Awareness, Knowledge, Ability, and Reinforcement.

If not, then the incentive won’t get the results you’re after. Employees won’t know about it, or know how to change the score, or be able to. Such incentives are just a shortcut to Desire – a shortcut that won’t get you where you want to go.

There is a better way.

Loftier goals are win-win

Instead of taking the easy way out and creating an incentive plan based on the score, focus instead of why your employees should care. If your company has customer-centricity as a focus (and it seems nearly every company at least claims to), then use this as a reason to focus on the outcomes.

In my old role, we were focused on changing healthcare through consumerism, as represented by HSAs. Instead of just paying people to improve customer scores, link the improved customer score to the mission. In that case, it would have been if we don’t improve the login process, our customers won’t use the accounts. That will mean we as a company do worse. More importantly, it means that we won’t change the healthcare system.

That’s the message that should be used, and it’s far more compelling than, “If we don’t fix the login, we won’t get that portion of bonus targeted at customer satisfaction.”

I get it. This is a lot of work, and it’s easier to just base a bonus on customer survey scores, hoping that teams will do what it takes to improve outcomes.

But people don’t work that way.

Unless you take the time to make your employees Aware of the existing experience, Desire to improve it, Know what to do, have the Ability to do it, and Reinforce this need, employees won’t be able to target the activities needed to improve the customer experience.

In the process of doing this hard work, odds are that you will have also created the infrastructure for customer focus…without the risk of your teams sounding like used car salespeople.

This article was adapted from a content series that originally appeared on the Heart of the Customer blog.