In 2011, Google launched Google+, an ambitious social networking platform aimed at the mushrooming Facebook user base. At its height, more than 1,000 engineers were contributing to the effort. After “Plus” famously failed to meet goals for engagement and market share, Google pulled the plug.
But what would have been seen as a colossally costly flop at other firms was classified as a win inside Google—from the perspective of compensation.
To understand why, you have to understand the system of OKRs, for objectives and key results, that is in place to guide strategy and power performance at Google. Those working on Google+ did indeed have goals and metrics for achieving user numbers, but even though they fell way short, those workers were not penalized. At the same time, since bonuses at Google are calculated primarily using a multiplier tied to profit levels, “you cannot give less than 100% multiplier because you’d demoralize the people who had nothing to do with it,” says Laszlo Bock, Senior Vice President of People Operations at Google from 2006 to 2016.
Either way, whether you worked on Google+ or not, your compensation didn’t take a hit. This was a recognition that Google+ may have failed for reasons out of the control of the individuals who created it.
A counterintuitive law of management
Bock is emphatic about the theory behind what happened at Google. “Objectives and key results should be totally divorced from compensation,” he says.
This law might seem counterintuitive. After all, isn’t that the point of setting goals and performance targets? If you work hard and make your numbers, shouldn’t you be rewarded for it? Why would you want to separate the two things? Is that ideal even possible once your goals and numbers are declared?
Google actually learned this lesson the hard way, when the company was still in its infancy. “One year,” Bock says, “Google tied OKRs for usage of a product directly to people’s compensation. People started gaming the system to get their bonuses.
The very idea of tying monetary incentives to hitting key results was thus deemed detrimental to both the product and the broader culture,” Bock says.
Avoiding perverse incentives
For an even broader-based illustration of why personal goals and bonuses should not be linked, take the common example of sales quotas. If you’re a salesman with a quota for selling $1 million worth of widgets for the year, for instance, you might work harder to earn a $10,000 bonus if you make your quota. If there are 100 other salespeople who do so, the company will hit revenue targets and be successful.
That is all well and good, says Bock, but if you do that you must be aware of the side effects. “You can absolutely pay a sales organization on sales quotas. You want their sales numbers to be part of their OKRs. But just remember, anytime there’s a sales incentive plan it creates perverse incentives, whether it’s tied to OKRs or not.”
There are many ways unintended consequences could crop up. If the goal is pure revenue achievement, people might push for volume even if there are no profits. In the software world, salespeople could prioritize enterprise contracts that may look like they’re putting lots of dollars on the board, but they might lose money over the long run. Or maybe the team will focus on financial services as a vertical instead of retail because financial services spend more.
Clarifying what your objectives really are
That’s where the simple levers of OKRs come in. Objectives are rarely just raw revenue, and they can be fine-tuned using key results that complement or clarify what the individual and the organization is trying to achieve.
For instance, instead of setting a key result of average dollars per seat, it could also be the total number of seats. Instead of achieving certain sales overall, the key result could be achieving sales across a range of five different industries.
Whatever the objectives and key results are, they should be lofty and ambitious, enabling people to stretch—without the expectation that all targets will be reached. At Google, in a typical year, 60% to 70% achievement of goals is the norm.
Creating more purpose-driven goals
More to the point, OKRs for both individuals and the organization should be a mix of performance metrics and higher order aims that shape and fortify the company culture.
At Google, salespeople would have OKRs where one goal would be the sales quota, says Bock. But there would also be other kinds of goals.
The broader OKRs should include demonstrating strong organizational citizenship behaviors. For instance, the objective of building a world-class team of talent could be linked to key results such as conducting at least 5 interviews or improving diversity on our team by at least 2 percentage points over the year. Another set of key results could be measuring nonprofit work, or measuring being a good coach and leader.
“That’s how salespeople who missed their quotas might still get the bonus,” Bock says. “Because there’s always other things besides hitting a sales target that also needs attention.”
It’s all in the mix
That’s where the right mix of OKRs comes into play, as bringing citizenship behaviors into the set of 4 or 5 objectives for each employee creates the right kind organization-shaping conversations.
“If you’re introducing OKRs to your organization,” advises Bock, “have one as the most pressing financial need and use slots 3 to 5 for other pressing needs. Keep the financial goal simple initially and have something that offsets it, such as complementary key results underneath it.”
In general, he adds, you want one OKR that’s economic, one that’s cultural and one that’s a kind of efficiency-oriented. You could have a personal one, too. But the idea is you care about the culture and all that’s underneath it.
Precision is also central to what Bock calls “the magic of OKRs.” With other performance-setting systems, like S.M.A.R.T. (Specific, Measurable, Attainable, Realistic/Relevant and Time-Bound), the goals can be either too generic or too focused on dinging people for not achieving certain numbers, by denying them their full bonus.
The early growth of Google is a classic case in point. If Google, for instance, had an OKR to drive lots and lots of search traffic, it could have a generated a lot of clicks by putting bad results on the page, so the user would just keep clicking to the next page. “But that’s a bad incentive, so we would add a key result around minimizing the amount of time people spent on the page.”
Eventually, this evolved into precise metrics balancing quality and volume. If people say, it’s too confusing to have several goals, what should we focus on, quality or quantity, Bock has an answer: “We’ve hired really smart and talented people. Let’s have both.”
Intrinsic beats extrinsic
It was Napoleon Bonaparte who said, “a soldier will fight long and hard for a bit of colored ribbon.”
The insight is that people are in it not just for the money or even the greater glory but for the things that drive them internally—autonomy, mastery, and purpose, as outlined in Drive by Daniel Pink.
At Google, an example comes from the 2016 Oscar-nominated movie Lion, in which the main character is seen using Google Maps to find his way back to his birth parents after 25 years of separation. For employees inside the company, this was seen as a superbly human story that fortified the reasons why they work at Google in the first place.
Notice the stark difference between how Google mainly makes it money—through advertising—and the narrative of what Google is about—connecting people and making human knowledge accessible. That’s why the Lion story was so energizing for people.
Whither the wisdom on bonuses
Along these lines, Bock points to research showing that the performance bonus is the kind of extrinsic reward that often falls short of motivating people in a sustainable way. “When you add incentives/bonuses, you have a short-term increase in productivity, but if the bonus goes down, people check out.”
A set of empirical experiments tested the responses of people promised cash awards to solve puzzles. The ones with the incentives soon drop below their initial efforts—and their performance also drops below the control group, suggesting that bonuses sometimes might be disincentivizing.
“When you tie performance to dollars,” he says, “people calculate their dollars: $100K per year plus $10k bonus. I could try and work longer for $10 extra per hour. But you don’t even want employees to make that calculation.”
The conventional wisdom used is that highly valued employees need bonuses. At Google, though, it’s not just elite workers; everyone is eligible for a bonus tied to company performance. Yet it’s been shown that people appreciate salary more than a bonus. “People in their minds discount the value of a future bonus by 80%,” Bock says.
Connecting through narrative
In the end, that’s the way you connect people to purpose—through narratives such as hero and villain stories, by illustrating the impact on human beings. It could be about saving a life or how you inspire people.
And that narrative needs to be built into leadership and management through OKRs that set the right priorities. Amy Wrzesniewski, a professor at Yale University, has shown that only about a third of people find meaning in their work. But when you can connect more people to meaning, on average, they’re about 21 percent more productive; they’re happier, and they lead more fulfilled lives.
Or as Laszlo Bock puts it: “There’s all this goodness that can happen.”
If you’re interested in starting our OKRs 101 course, click here.