Objectives and Key Results (OKRs)

From AgileMe
Jump to navigation Jump to search


Objectives and Key Results (OKRs) as a mechanism have been around since the concept of Management by Objective by Peter Drucker in 1954, and popularised by John Doerr in an OKR form in his book Measure What Matters, OKRs which describes how OKRs were implemented in Intel and at Google among many case studies.

OKRs are cited as providing 10x growth and help to provide alignment to aspirational goals, which should improve teams' focus and help to clarify decision making especially if there is a tendency for teams to get sidetracked or distracted with other work.

OKRs also help teams question what is really valuable and what does the world really want, rather than delivering a random collection of features that may only moderately resonate with customers and end users.

load PDF

Objectives

Objectives are intended to be:

  • significant - that they mean something and answer the why question
  • concrete - straight forward and no nonsense
  • action oriented - cultivate real execution and deliverables
  • inspirational - attract real change and vision

It's recommended to have no more than 3-5 objectives as an OKR.

There tend to be two forms of an objective:

  • committed objectives - which are intended as firm deliverables and help to provide a clear and concise North Star direction for teams
  • aspirational objectives - objectives that are intended as stretch goals or Big Hairy Audacious Goals that are intended to inspire teams to achieve beyond the normal expectation. It is common to expect only 60-70% completion of these goals, but help to foster big thinking in the teams rather than only considering the norms

Cadence

Popular cadences for OKRs are quarterly, and can be monthly for organisations that are quite mature in their agility and have a flexible approach to their working practices

Key Results

Key Results are intended as quantity or quality measures to help track progress towards the Objectives, and are intended to be measurable and quite specific to be useful.

Good key results can be easily measured and are valuable metrics rather than vanity metrics.

Scoring

A popular scoring approach uses a range from 0.0 to 1.0 to indicate the progress which are updated regularly by teams. The following ranges are used with a corresponding red, yellow, green indication:

  • 0.0 to 0.3 - red
  • 0.4 to 0.6 - yellow
  • 0.7 to 1.0 - green

Creating OKRs

Teams are expected to create their OKRs and should be gained by agreement and consensus in the group for the teams to value them and maintain them.

For larger organisations, it may be useful to create tiered OKRs at various levels such as at an organisation level, a program level and team level for example. Here a bottom up approach is recommended, although it may also be useful to do all levels at once and fine tune with feedback from the other levels and teams.

Objective
Improve our subscribers' perspectives of the app
Key Result
* Improve Net Promoter Score by 50%
* Increase customer subscriptions by 10%
* Reduce incident responses by 20%

Self Assessment

Key Results are intended to be measured through out the cadence by the teams and at the end of the cycle teams are expected to reflect on the performance towards the Objectives providing information on why committed objectives were not able to be met and how things can be improved going forwards.

Notes from these reflective conversations may be kept with the OKRs for future reflection and process improvement.

Agility and Flexibility

OKRs may need to be adjusted to reflect the current understanding as the work is done, and so should be maintained to still be in the spirit of how they were originally conceived, but also realistic with what we know now.

Common Pitfalls

Common traps with using OKRs include:

  • failing to differentiate between committed vs aspirational OKRs - committed objectives are intended to be delivered and within the teams' capabilities, but if these are too complex or too large to achieve, then are they really aspirational goals and so should be treated as such. Teams that consistently fail to achieve committed objectives may well suffer from a loss in confidence and a further degradation in performance
  • low value or business as usual OKRs - if there is no specific intent behind an objective other than just doing what we normally do, then there will be no performance improvement, and subsequently the OKRs will cease to be useful or valuable to the teams
  • insufficient KRs for Os - Key Results are needed to help track progress and also inform our understanding of the journey towards the Objectives. A reasonable number of Key Results should help provide a more rounded perspective on the progress towards the Objectives
  • timid aspirational OKRs - not going far enough for the aspirational OKRs will not help the teams to realise big thinking and instead may well encourage them to remain with the status quo
  • sandbagging - setting the teams up to fail won't help them improve their performance, so even aspirational Objectives should be possible with the teams' capabilities, but are a stretch to help them think beyond their norm
  • failing to separate remuneration from OKR performance - if OKRs are strongly linked to individual performance appraisals then this can encourage individualistic behaviour and unethical activities. Conversely, separating OKRs to focus on work performance and the work needed to be done will allow teams to rally around that work
  • lack of safety - if teams do not feel safe and able to objectively fail to meet OKRs then this may encourage them to take short cuts and reduce quality in an effort to game the OKRs and may impact the organisation, customers and users
  • verbose OKRs - OKRs are best to be short and to the point. If they are too long, then it may be interpreted differently by different groups or teams. If they are too short, then there may not have been enough thought gone into the OKR for it to be useful and valuable.
  • measuring infrequently - if OKRs are not measured frequently or not highly visible to the teams, then there is the possibility that the teams lose the sense of alignment, go off track and lose feedback on their progress towards the goals. Frequent measurement allows teams to take corrective action and make good decisions sooner if they know that they are not on track

See Also