Objectives and Key Results (OKRs)

Updated on September 27th, 2021
Objectives and Key Results (OKRs)

As product managers, we have the responsibility of moving the business forward, and one of the most effective means for ensuring we do so is through the use of objectives and key results (OKRs).

I admit that when I first started as an associate product manager, I had no idea what the term “OKR” stood for, and it took me a while to fully understand how to use them! So, to help other newly-minted product managers onboard quickly, I’ve written this article as a deep dive into OKRs.

The OKR framework enables us to identify what’s important to the business, to set targets for what we should achieve, to track our progress towards our targets, and to diagnose the root cause for any unexpected results.

Below, I’ll discuss the difference between objectives and key results, and I’ll also share best practices for leveraging each one. So let’s dive in!

Objectives and Key Results (OKRs)

Using OKRs is a valuable way to frame your decision-making, and to determine whether you’re faithfully moving the product forward towards the company’s long-term goals. Let’s break down both terms separately.

An objective is a qualitative goal that isn’t itself measurable. For example, an objective might be “demonstrate the viability of a particular market” or “empower users to work more efficiently.”

On the other hand, key results are quantitative measures that are easily measured. You’ll want to measure the baseline, and then define the key result relative to that baseline. For example, a key result of “empowering users to work more efficiently” might be something like “improve time-to-complete for a particular task from 25 minutes down to 10 minutes.”

There’s a starting point (25 minutes to complete a particular task) and an ending point (10 minutes to complete a particular task), and both are clearly measurable.

Why do we pair objectives and key results together?

Well, on the one hand, if we only have objectives but no key results, we wouldn’t be able to determine whether we were making progress towards our objectives.

And on the other hand, if we had key results but no objectives, we might wind up optimizing metrics that don’t actually drive the product or the company forward. Too often, I see product managers make the mistake of optimizing metrics without first asking “which objectives do these key results drive?”

So, when we pair both qualitative aspirations and quantitative goals together, we can meaningfully move towards a product vision that unlocks positive impact in the world.

Let’s talk a little bit more about how we should use OKRs.

Best practices for setting objectives

Generally speaking, each product manager should be focusing on just having one or two overarching objectives. Since each product manager is in charge of a given problem area or a given scope, that scope should be what’s identified in the objective as well.

For example, I used to own a real estate agent CRM mobile app, with the problem space of “how do we enable real estate agents to drive more home sales?” 

My objective was therefore “empower real estate agents to effortlessly drive more home sales.” Notice how my objective very closely parallels the problem area that I own.

Could I have broken down the objective down into multiple ones? Sure - for example, I could have broken it down into three components like “enable real estate agents to collaborate together,” “strengthen the online presence of our agents”, and “eliminate repetitive processes.” 

But, the problem with doing so is that I’m then forced to pursue overly-granular objectives that may not actually drive the business forward. What my employer actually cared about was “enabling agents to drive more home sales.” Our bottom line wouldn’t actually have improved if I had solely pursued the subcomponent of “enabled real estate agents to collaborate together.” 

So, make sure you only have 1-2 objectives as a product manager. Remember that every time you’re simultaneously prioritizing multiple things, you’re actually prioritizing nothing at all. You have to make tough decisions on the single thing that matters most to your product and to the company.

Of course, if you’re a director of product management or higher, then you’ll have multiple objectives at the product group level. That’s fine since you’ll have multiple individual contributor product managers driving each objective. But, at the individual contributor level, you shouldn’t have more than 1-2 objectives at a time.

Within an objective, you should similarly have only 3-5 key results for each objective. Any more than 5 key results, and you’ll wind up chasing too many goals at once. Be ruthless in your prioritization!

Now we know what OKRs are, and how they work. Let’s dig a little deeper into how we might measure “key results.”

Key Performance Indicators (KPIs)

Key results are also known as key performance indicators (KPIs). Don’t be surprised if people mention KPIs as part of OKRs, even though KPI isn’t represented in the acronym for OKR! Based on my experience, “KPI” is used much more frequently than “key result” is, so I’ll refer to them as KPIs below.

Your KPI needs to reflect the objective that it’s associated with. That is, you shouldn’t try to use a KPI that’s “easy to measure” but doesn’t align with the business objective. As an example, I frequently see product managers measure “number of active users”, but the business may not be focusing on acquiring users; they might be focused instead on driving task efficiency, which is a much more difficult metric to measure.

Again, make sure that you only have 3-5 KPIs associated with any given objective. Aim to capture each of the different facets of your objective. For example, let’s look at my objective “empower real estate agents to effortlessly drive more home sales.” I should probably capture KPIs like “increase number of sales per user,” “decrease sales cycle time per user,” and “improve lead-to-sale conversion rates.”

Your KPIs should have a starting threshold and an end target, and both should be measurable.  On top of that, your KPI should be time-bound, such as “within the next quarter.”

Once we have our KPI, we then turn it into a 0 - 100% scale that shows our progress towards the KPI. As an example, if one of my KPIs was “move real estate agent NPS from 25 to 35 this quarter”, then an NPS of 25 would be 0%, an NPS of 30 would be 50%, and an NPS of 35 would be 100%. 

When we set goals for ourselves, we should try to stretch a bit. Stretching is what enables us to grow as product managers! So, that means we shouldn’t set an end target that we’d hit 100% of the time. The goal loses its ability to motivate if it’s too easy to meet.

On the flip side, we also shouldn’t overextend by setting massively ambitious goals that are completely unachievable. When we regularly miss our targets, we demotivate our teams and our stakeholders, and that loss of faith can cause burnout, misalignment, and conflict.

So, as a rule of thumb, try to set a goal where you’re pretty confident you’ll get to 60-80% of the end target. As an example, if I’m pretty confident that I’d be able to increase the number of real estate agent users on my mobile app from 10k to 15k users by the end of the quarter, then I should set a target of 18k users by the end of the quarter instead.

How does that math work? Well, I started at 10k users, and I’m confident I can end at 15k users, so that means that the distance is 5k users. I should set 5k users to be the 60% mark, so to get the 100% mark, I should do this math:

5k / 0.6 = 8.3k growth in users

Or, if I’m a bit less confident, I could set 5k users to be the 80% mark, so to get to the 100% mark, I should do this math:

5k / 0.8 = 6.3k growth in users

Another way to look at KPIs is to assess whether they’re SMART goals:

  • Specific: the goal is clearly defined, easily understood, and is not ambiguous
  • Measurable: the goal is quantitative and can be measured
  • Achievable: the goal is realistic
  • Relevant: the goal actually matters to the business
  • Time-bound: the goal cannot be in “the indefinite future”

All of our guidance above aligns well with the SMART framework! The great thing about using KPIs is that everyone’s aligned with the same view of reality.

Rather than arguing qualitatively over something like “real estate agents love our app” vs. “real estate agents hate our app,” we all see the same view of the numbers and can determine what the best next action is to take.

Let’s say I decide to stick with the 8.3k user growth target. Now let’s talk about how I’d take action on the results at the end of the quarter.

How to use KPIs to grow the business

There are only 3 outcomes that can come from a KPI:

  • You meet the 60-80% target threshold
  • You exceed the 80% maximum threshold
  • You don’t achieve the 60% minimum threshold

For each kind of outcome, we should run a targeted retrospective so that we can strengthen our product. Let’s discuss what each outcome means for our product and our business.

If we meet the target, then we should take the time to understand why and how we met the threshold. What was effective for us? What did we do to minimize risk and maximize upside? What could we do differently next time to achieve even better results?

If we overshoot the target, then we should ask ourselves honest questions. Did we set the goal too low, and if so, how do we ensure that we set it higher next time so that we aspire to greater heights? Or, was it that we didn’t expect a favorable market? If so, how do we ensure that we update our understanding of the market so that we can capture as much of it as possible?

If we don’t meet the target, there’s no need to panic yet. We need to ask ourselves questions that are both compassionate and probing. The overarching goal is to answer this question: “What could we do better next time?”

Here are some questions to ask when we don’t meet the target. Was this miss due to a failure that we could have mitigated or foreseen? Was it a feature that significantly underperformed, and if so, why? Did we run into a lot of bugs, and if so, is it time to invest in infrastructure and resiliency?

Or, if this was due to external factors, was there competitive action in the market? Did the entire market experience a downturn, or was it just us?

By taking the time to ask ourselves these questions for the 3 different kinds of KPI outcomes, we can continue to evolve our product development teams, the product itself, and the culture of our company.

To learn more about how to diagnose a metric change, see our article on analyzing metric changes. While it’s written from a perspective of a product manager candidate interview, the same framework works well for your day-to-day product management responsibilities too!

Closing Thoughts

Objectives and key results, commonly known as OKRs, are a powerful way to ensure that you’re moving your product forward towards the right long-term vision.

OKRs combine qualitative objectives with quantitative KPIs. By using both in tandem, you motivate your team to excel, and you reveal insights into how to continually improve your processes and judgment by looking at the outcome of each KPI.

You no longer need to fear the acronyms “OKRs” and “KPIs”, congratulations! In our next article, we’ll pair OKRs with another powerful concept called “the directly responsible individual (DRI)”, which will further refine your ability to drive success as a product manager!

Want to learn more about product synergies? Chat with other product managers around the world in our PMHQ Community!