One of the exciting things about employing OKRs is that you are not certain to achieve them. As you might recall from my last article, good OKRs are challenging. Christina Wodtke, in her book Radical Focus, provides guidance that a team should be no more than 50% confident, or 5 out 10 on a scale, that they will achieve the outcome. We need clear proof, in the form of key results, that let us know if we are progressing towards the objective and when the objective has been achieved. Clear proof also helps establish the credibility of these objectives with stakeholders who depend on your performance to achieve their own goals.

In the remainder of this article, we’ll cover:

How to provide proof
Key results that provide credit, but not proof, and reasons behind it
What you can do to support proof over credit

How to Provide Proof

For me, providing proof means:

Find the one metric that best represents the objective
Ensure the metric is one you can correlate to your efforts
Make the result clear
Provide additional evidence

Finding the one metric to best represent your objective, sometimes called an impact metric, may be obvious or it can be a bit tricky. Imagine a runner who wants to become one of the top runners in their area. You’d have a few options available to you: percentage of races won, average finish per race, fastest time compared to competition, etc. Which of these would best support your claim on being one of the top in your area?

Whatever metric you choose, it needs to be one that correlates to your efforts. Picking metrics that are hard to attribute to your efforts, is weak proof that you achieved the objective. For instance, what if you picked the average finish per race to support your running objective. Now imagine, you finished top 3 in all of your races, but most of your competition in these races was either from another area or not the top competition. With this knowledge it would be hard to support that you met your objective.

Quantitative data provides objective evidence that the target was achieved or it wasn’t. By having already chosen a metric that you can correlate to your efforts, you strengthen your belief in the data. Qualitative data can still be useful to understand why you did or did not achieve the results, but the outcome is clear without it.

Finally, consider having a couple of additional key results which support your objective, around three in total. These will provide more insight and allow you to impact a few more metrics that you believe are important to move.

Key Results that Offer Credit

Proof is important because achieving the objective is uncertain. There are a range of outcomes from badly missing or abandoning the objective to achieving or exceeding it on the high end. Organizations new to OKRs often have not yet undergone the shift necessary to create effective key results. As a result, these are a few common mistakes they are likely to make:

Activities – they record activities like “build a database” or “migrate users to the new system”
Vanity Metrics – they use metrics that make them look or feel good, but don’t really make the impact intended by the objective; these are often contrasted with actionable metrics — those that provide real insights to performance
Ground Shots – they set the bar too low on the magnitude of the change they would like to see; consider this term in contrast to roof and moon shots from the previous article

With any of these, you would expect them to be above, instead of at, the 50% confidence level. In other words, these types of results are more certain and potentially safer for the people that have to deliver them.

Some or all of the following reasons could influence the selection of results that provide credit more than proof:

Not safe to take chances; people fear the consequences of missing their targets
Expectations are low; people don’t think they can deliver much or that a bigger impact may not be needed or warranted
Not used to this form of goal setting and achievement; feels unnatural; stay with what you know and understand
Lack of awareness of what matters or how you can measure it

Moving Forward with Proof

The reasons for identifying key results that provide credit may be intentional or unintentional. Regardless, they fall short of the impact you can deliver using OKRs. Here are some recommendations on how to avoid key results that provide credit:

Provide education on OKRs including relevant examples of good ones and a list of mistakes to avoid
Make sure everyone’s clear on higher level goals and strategies and their importance (rally them around a higher purpose)
Bring an external facilitator to run your first OKR workshop; this is equivalent to hiring a personal trainer or coach to move you beyond your limitations
Ensure you address any limitations around safety or expectations so that you get the most out of your OKRs

Here is a recap of how to create proof with some additional tips:

Find the one impact metric to prove the objective
Shoot for about 3 key results that you can use to gain insights and achieve your objective
Make sure you use clear results that you can measure quantitatively
1. Avoid binary results that are ‘all’ or ‘none’. These don’t lend themselves well to evaluation until after the fact and don’t support decision making
2. Good results often start as absolute or relative increases or decreases (e.g. increase from x to y; or increase x by z%)
3. Provide a range that includes your baseline, or current measure, and your target measure
Constrain the key results by time; as in by or before x date we want to achieve these results. Provide enough time to gain insights from the data and make decisions to better support the work.

Summary

By creating objectives with key results that are believable and measurable you can provide evidence whether you have really achieved your goal. The evidence will be understood and accepted by both those who worked to achieve it and those who benefit from the work. Further, creating these types of key results sets up the kind of flexibility to explore different options to achieve them and pivot as you obtain the result data.

While it sounds easy, there are some pitfalls to consider where people naturally find metrics that provide more credit than proof and insights. Despite being natural, don’t settle. Work to educate, challenge, and support more meaningful and challenging key results that show a real impact.