How To Set The Right Performance Targets for Your Team
The art and science of motivating, aligning, and focusing people
Performance targets can make or break your team. Done right, they will motivate, align toward common goals, and drive focused work. Make a mistake and you will get a demotivated team focused on gaming the system for their personal benefit.
Wells Fargo bank was long considered the most successful cross-seller. A client with a checking account would be encouraged to take out a mortgage, set up credit card or online banking account. Performance of the branches and employees was measured by how many products a client had on average.
In 2016, a story broke that Wells Fargo managers opened as many as 1.5 million checking and savings accounts, and more than 500,000 credit cards, without clients’ consent.
The company faced civil and criminal suits reaching an estimated $2.7 billion by the end of 2018. The creation of these fake accounts continues to have legal and financial ramifications for Wells Fargo and former bank executives as of early 2021.
— Wikipedia
Over the last few years, I participated in dozens of target-setting cycles across many company functions on every level of the organization. This article is a set of non-obvious rules I learned from successes and failures.
1. Don’t Be Lazy
The famous quote by Peter Drucker says “You can’t manage what you can’t measure”. While Drucker never actually said that, the idea seems reasonable. Managers often use it as a guiding light, forgetting about an equally important aphorism “what you measure is what you get” that ends up sinking their ship. This is what happened with Wells Fargo in the story above.
Measuring value is hard. It is much easier to measure what people do.
Software engineer writing a line of code is easy to track, but the value added by this line of code is not obvious, so let’s measure performance as the number of lines of code written. It is scary how many people still seriously suggest this approach in one form or another.
This is an exaggerated example, but more subtle versions of it happen all the time. Activities — things people do — are almost never good for measuring performance. You have to think harder and figure out the ultimate value you want to get from their actions, and measure that. You will almost never reach the “ultimate value”, but try to get as close as you can. Don’t be lazy.
That said, sometimes you can use activities for measuring performance — more on that later.
2. Create The Source of Truth
The last thing you want to happen is this: you set great goals at the beginning of a quarter, the team works hard for three months, and at the end of the quarter, when it is time to celebrate, no one knows how to measure the results, or people cannot agree on which measurements are the right ones. Oops!
On November 7, 2000, United States television networks, first announced that Al Gore won Florida in the presidential election in the hour after polls closed in the peninsula but about ten minutes before they closed in the heavily Republican counties of the panhandle. Later in the evening, the networks reversed their call, moving to “too close to call”, then later giving it to Bush; then they retracted that call as well, finally indicating the state was “too close to call”.
Gore phoned Bush the night of the election to concede, then retracted his concession.
In the end, the whole election was decided by the famous Bush vs. Gore ruling by the Supreme Court.
To prevent this unfortunate outcome, all you need to do is to agree on the source of truth for performance. It could be a dashboard, report, screen in an application, or a spreadsheet that everyone agrees to trust when it comes to performance indicators. Create it as early as possible and don’t allow anyone to create alternative measurements that don’t match the source of truth.
3. Know The Limits of Attention
I didn’t have time to write you a short letter, so I wrote you a long one.
— Mark Twain
Our mind cannot hold more than 3–5 things at the same time. If someone on your team needs to think about more than 5 performance indicators, they likely won’t pay attention to any of them.
Don’t set more than 3–5 performance targets.
But there are more than 5 important things! Of course, but don’t forget that the goal of performance indicators is to focus on what is the mostimportant. Just for this quarter, are there 3 things that you want your team to focus on? Next quarter it could be other 3 things.
Many companies use term Key Performance Indicators. You would’t have more than 3 different keys from your house, so don’t create more than 3 Key Performance Indicators.
It does not mean you should not measure anything besides these 3–5 things. Measure as much as you can — you’ll thank yourself later. Just don’t make them targets for your team.
4. Use Leading Indicators
To be effective, targets need to be on people’s mind at all times. But many performance indicators — like customer revenue or customer satisfaction — take too long to realize and while important, not useful to look at daily or weekly.
The solution is to create Leading Indicators that have a strong causal link with Key Performance Indicators and are faster to measure.
Ask yourself: what happens daily or weekly that indicates future performance?
For customer satisfaction it could be speed of reacting to support tickets, for revenue — number and quality of meetings with prospects, for number of new hires — number of interviews. Pick ones that meet the following criteria:
You can reliably measure them with the frequency you want to review performance (typically weekly).
There is provable connection between leading indicators and the actual performance.
Because leading indicators are not the actual performance, they are tricky and if used incorrectly can cause a disaster. The next few rules will help you avoid that.
5. Counter-Balance Leading Indicators
Remember the Wells Fargo story from earlier? This is what happens when you allow people to over-optimize for leading indicators. The number of open accounts is not valuable on its own, but it was a convenient proxy taken too far.
To avoid making the same mistakes, do the following:
Think what could suffer if your team focuses on the leading indicator too much. It is usually quality and customer experience, but sometimes less obvious things.
Pair your leading indicator with another one that measures whatever can be affected negatively.
For example, if you measure the number of deals sales close, pair it with customer success and retention (unless immediate revenue is your end-goal), and if you measure how many bugs an engineer fixes, pair it with the number of bugs re-appearing in the same area of the application.
6. Beware of Zone of Control
Setting expectations outside of what your team can control will be useless at best and demotivate at worst.
In the 1960s, two University of Pennsylvania psychology graduate students discovered that when dogs received electrical shocks that they could not control, they later showed signs of anxiety and depression, but when dogs could end the shocks by pressing a lever, they didn’t.
What’s more, the dogs that received the uncontrollable shocks in the first experiment didn’t even try to escape shocks in a later experiment, even though all they needed to do was jump a low barrier.
If the output of your team depends on the input they receive or on work done by another team, don’t make your team fully responsible for the output — don’t set the output targets. Instead, set targets for converting inputs into outputs.
For example, if your team’s job is to make customer successful, the result will depend on what kind of customers they get. It is easy to demotivate such team by demanding to make successful a customer who would not even return calls. The solution could be either allowing the team to control what they get as an input (qualify customers in this case) or set targets that take quality of the input in account.
On the other extreme of this problem is setting targets that are fully within control. Such targets are very comfortable for your team, but often not as meaningful as something that can be influenced, but not controlled.
Sometimes members of your team will say something like “we can deliver good work, but we can’t be responsible for customer using it and be happy with it — it’s outside of our control”. Don’t buy it. This argument sounds similar to “sales should not be responsible for revenue since they can’t force customer to buy — it’s outside of their control” . Your business would not get far with this mindset.
Team should feel enough control over the outcomes to not feel helpless, but not much more. Finding the right balance is an art that good leaders and managers should master.
7. Don’t Ask Your Team To Lie
Sometimes you have to rely on self-reported performance. For example, customers success managers may assess the health of their accounts, or engineers may estimate the size of their tasks.
Never use self-reported data for measuring performance.
Sometimes it’s the only data you have, and you feel pressure to say “well, what else should I use?”. Using “nothing” is infinitely better than using self-reported metrics.
If you use self-reported performance, you give your team a choice between lying and looking bad. If you compare people or teams based on self-reported performance, you motivate them to lie better than the next person.
I spoke with a Scrum Master recently who told me his team had nearly doubled their velocity in only two months. Rather than be happy about this, though, he was concerned.
He knew the team had not suddenly become twice as productive. In fact, he doubted they’d actually sped up at all. Yet their velocity showed they had.
Mike Kohn — How to Prevent Estimate Inflation
Watch out for self-reported metrics and their derivatives. Looks where the raw data for your performance targets is coming from and if at least some of it is self-reported — fix it.
8. Don’t Mix Aspirational Goals and Performance Targets
Objectives and Key Results is an amazing management tool created by a legendary CEO Andy Grove. It transformed Intel, Google, and countless other companies.
A key quality of OKR is ambitiousness. They motivate people to achieve more than they thought possible, think big, and push their limits. The problem arise when people try to use OKR as performance targets.
It should be obvious that saying “set aspirational goals” and “your salary will depend on achieving these goals” at the same time is s̶t̶u̶p̶i̶d̶ sending mixed signals, but having OKR in between throws people off and more than a few times I’ve seen people trying to use OKR to assess people’s performance.
You can have both bottom-up aspirational goals and top-down performance targets, but they cannot be the same goals. Google uses a notion of committed vs aspirational OKR and it is a good approach if you want to go this path, but be super clear which goals are which and never mix them.
Conclusion
Performance targets is one of the most powerful tools leads and managers have at their disposal.
Use this power right and you can create a star team. Use it wrong and get a disaster that could destroy your company.
Put some time and effort into getting performance targets right.
Like this post?
Want to receive posts like this in your inbox (1-2 per week)?