Design Key Results for OKRs that work

man using tools laid on ground to increase his putting ability in golf

Key concept: What is an OKR?

If you haven’t come across them already OKR is a goal-setting approach pioneered in Intel, based on Peter Drucker’s work on Management by Objectives in the 1950s, which is now used widely across the corporate world.

OKR stands for ‘Objective Key Results’.
The Objective element describes a substantial, often long-term outcome we want to achieve.

Key concept: What is a Key Result?

The Key Result is a smaller outcome, a stepping stone goal, that will help us reach that big ‘Objective’.

An example personal OKR

Objective: Give up smoking
Key Result: Go 10 days without a cigarette

An example corporate OKR

Objective: Quadruple online sales
Key Result: Grow email list to 50k subscribers

Design effective targets and key results using the ROKET-DS approach

Self-pace course available from KPI Academy

Here's a quick introduction and some resources on each of those ten key result design steps...

Play Video

Step 0: Identify existing issues

If you are struggling with targets that are not driving the right results or are fuelling poor behaviour, it’s important to have a full understanding of the issues. This will give us important input when we come to test our revised targets. In the in this step we research and record:

Existing targets and incentives?
Side-effects, symptoms & problems?

Play Video

Step 1: Plan outcome

Without a clear picture of what outcomes we are looking for, we are doomed to failure. We need to clearly and precisely define the outcome we are looking for, avoiding vague ‘woolly’ words, that can lead to misunderstanding and confusion. In this step we ask:

Intended high-level outcome?
Why is this outcome important?

Play Video

Step 2: Match KPIs

It can be very dangerous to target a single metric or KPI, as human nature is to focus intensely on measures which are targeted and rewarded, often to the detriment of other, important, outcomes.

An example of this risk is the common situation of a sales professional selling extra unwanted products to customers in exchange for a discount on their intended purchase, on the agreement that they can return the extra products after month end (and sales bonus payment based on sales revenue).

The solution to this challenge is to use a KPI Tree to create a balanced set of KPIs to target.

Which KPIs show if you are moving towards 'Planned outcome'?

Play Video

Step 3: Identify and engage target owners

It is critical that we identify the owners and managers of our planned targets or key results, so we can…

Understand who has control or influence over the KPIs identified in Step 2
Assess owner ‘agency’ in the next step
Involve a sample of that group in both testing steps
Build full engagement and buy-in for the planned targets and key results

Play Video

Step 4: Check owner agency

The word ‘agency’ describes whether a target owner is fully equipped to attempt to hit, or exceed, the planned target. There is little point in setting a target that is completely unobtainable. Doing so, particularly if you then add strong incentives, can lead to dangerous behaviours ranging from cheating through to blatant criminality. It can also lead to team stress, anxiety and attrition. The elements of ‘target owner agency’ are:

Skills
Time
Authority
Resources

Play Video

Step 5: Draft target values and rules

Once we have the ‘basked’ of KPIs that we need to develop target/key results for, we need to decide on the draft values and timescales. This discussion will involve considering:

The meaning and impact of the planned outcomes from achieving the targets
The practicalities of achieving those results
The relationship of those targets with incentives (if we intend to link these targets/key results to rewards/punishment)

Our outputs from this stage are:

Target values
The 'rules' for achieving targets

Play Video

Step 6: What hat test targets

With any system of targets or key results, there is an ‘intended’ way in which they should operate.

For example, a sales target may be intended to work in the following way:

We set a stretch target for each sales team member
The sales team increase their activity and/or conversion rate
We see and uplift in sales margin in line (or exceeding) the stretch target
The sales, margin and profit of our business increases in line with the stretch targets

At this stage we just think about how things should work, not about the ways in which people might game the system or do unexpected things - that comes in the next step.

Step 6 can be summarise by the following question: 'What should happen?'

Play Video

Step 7: Black hat test targets

Humans are highly sophisticated and maximising the personal benefit from any situation. Anyone who has played a board game with loved family members will have rapidly discovered just how devious and ingenious those family members can be. Many targets and key results are launched without proper testing.

In Step 7 - Black-hat testing we gather a sample of the target-owning stakeholders and ‘reverse brainstorm’ how the target/key result could be ‘achieved’ but with the worst-possible real-world outcome. Doing this enables us to fix the design issues before any failings lead to operational or PR catastrophe.

Reverse brainstorm how to hit targets in worst-possible way
Apply ROKET-DS Diagnostic

Play Video

Step 8: Fix problems and re-test

Black-hat testing is likely to flush out a number of unanticipated issues. To fix those, head back to the appropriate step in the process.

For example, if we think we need some extra KPIs to get a balanced outcome, then we need to head back to Step 2 and briskly work our way back through the following steps.

Once we can clear the ‘black-hat testing’ stage with no significant problems being identified we are ready to move on to the final target setting step.