Connecting Strategy to Action - Anti-Patterns in OKRs

Objectives anWhat not to do when deploying OKRs in your organisation.

November 5, 2021

Anti-Patterns in OKRs 

Objectives and Key Results (OKRs) are exploding in popularity as a mechanism for taking strategy into delivery, in fact they form part of the model that we recommend that organisations use to take Strategy into Execution in remote:af teams.

how remote:af organisations take strategy into action with OKRs

OKRs done right are a brilliant way to implement what Lt David Marquet calls ‘Intent Based Leadership’ in organisations through a simple mechanism which enables focus and alignment. But... there’s a few things that you need to have in place before they will be effective. Here’s some of the key anti-patterns we are seeing in the use of OKRs with remote or hybrid remote teams.

OKRs with matrix management models

If... you’re trying to use OKRs with a matrix management model

Then… Go directly to management jail. Do not pass go, do not collect your bonus.

OKRs work really well in value-aligned operating models that are built around cross-functional teams who can deliver in a largely autonomous fashion. They are designed to cascade bi-directionally through an organisation, with leaders setting intent and teams telling leaders how they will get there.

They don’t work particularly well in organisations that have operating models built around functional hierarchies that use the project or business process metaphor for orchestration of value. Use of OKRs in these environments will likely lead to local optimisation of the functions at the expense of the system.

You could develop a workaround which creates temporary team structures that span your business functions and use Impact Mapping or something similar, but OKRs won’t work out of the box. Alternatively explore our approach to remote:af Operating Model Design to design operating models that are optimised for remote or hybrid environments.

OKRs as a control mechanism

If... you’re using OKRs as a mechanism to get more control

Then... you’ll get a fraction of the potential value and a lot of sleepless nights

OKRs are not meant to be a control mechanism for leaders to have traceability / visibility of all the work that is going on in an organisation and they definitely shouldn’t be the sole source of the work backlog for teams (this will vary by team type, arguably Mission Teams should be completely driven by OKRs).

OKRs are designed to capture an organisation’s aspirational goals, a ruthlessly prioritised set of the things that will create the greatest organisational value, rather than Business as Usual or Operational work. They’re designed to allow leaders to take their hands off the puppet strings, safe in the knowledge that their people know they can be most effective at adding value to the organisation. Autonomy is a key motivator for knowledge workers and a well designed OKRs framework sets boundaries 

If you’re craving that sort of control of the system as a leader it’s a good idea to firstly reflect on ‘Why?’, and explore the difference between leadership and management.

OKRs as a linear process

If... there’s only one route to get from intent to action

Then... you’re going to have a bad time.

When you look at a diagram of the OKR process it may resemble a linear decomposition of intent into value but in reality a good OKR process recognises complexity. Key Results can be quite objective / numeric for items that teams feel fit into the clear domain of Dave Snowden’s Cynefin framework, but as the complexity of the objective changes, so must the decomposition process.

More complex objectives might decompose into multiple parallel probes with subjective key results that value learning, analysis or planning. The overall purpose of your key results is to demonstrate progress towards your objectives. And, the best way to do this changes with the nature of the work.

OKRs without collaborative spaces

If... you don’t have a space for teams to asynchronously collaborate on OKRs

Then… your people are going to go beyond zoom fatigue and into zoom narcolepsy

A critical element for solving the problem of taking strategy into delivery with remote teams is to create digital spaces that don’t suck, where people can collaborate around content rather than just talking on VC. You’ll need access to a decent digital whiteboard (e.g. Miro or Mural) and you’ll need to create spaces that enable people to collaborate together on common goals.

OKR tools can be useful for documenting and scoring OKRs, but they’re not particularly good at the design and decomposition process.

Our remote:af guides are enabled with a toolkit that supports the key processes you’ll require: Strategy Design, Strategic Planning and OKR Development, Strategy Scoring, Inception / Product Discovery, All Hands Planning and the Virtual Obeya.

In the absence of that, spend some time thinking about how to create spaces where teams can collaborate on the setting of OKRs, on exploratory work for objectives that aren’t straight forward, and on analysis and planning work for Objectives that require expertise or span multiple teams, and on making progress visible for scoring.

OKRs without the back-brief

If… you set OKRs and don’t ask your people how they are going to deliver them

Then… you’ll need to invest in therapy to bridge the gap between your expectations and the outcomes that you get

In his brilliant book “The Art of Action, How Leaders Close the Gaps between Plans, Actions and Results”, Stephen Bungay gives a detailed history of Mission Command (Augfragstaktik), describing how leaders can lead autonomous teams using the Intent > Plan > Action > Outcome cycle.

In this book he highlights three key gaps in intent based leadership: The knowledge gap, the alignment gap which is our focus of interest here, and the outcome gap. It is not enough, he says, for leaders to simply set intent - they must understand how the team intends to get there and ensure it is aligned with both the leader’s intent and not in contention with other parts of the organisation.

OKRs need to cascade bi-directionally through the organisation, decomposing to work and then recomposing in the form of back-briefs to leadership. Where teams assess an objective or key result as unachievable rather than aspirational, there should be an opportunity for realignment. OKRs that create dependencies between teams need to be planned before they can be objectively back-briefed, which means that the strategy and planning cycles need to be aligned.

OKRs with misaligned cadence

If… you’re running a three months strategy cadence in an organisation that struggles to go from idea to value within six months

Then… your people will feel like they are playing whack-a-mole

The strategy cycle is inexorably linked to the delivery cycle - you need to be able to reflect on the outcomes of strategy before you can adjust strategy for your next cycle, which means you need to see intent turned into plans, plans turned into actions, and actions leading to outcomes.

One of the key anti-patterns we often see is organisations trying to run a short (e.g. quarterly) strategy cadence when delivery lead times are fairly bloated. If it usually takes you four to six months to release a feature and you’re trying to do strategy quarterly, you’re not going to have a great time.

Use a tool like flomatika to get visibility of feature lead times and make sure that your strategy cadence is longer than the 85th percentile of your delivery cadence. Note: while some industries (e.g. Utilities and Pharmaceuticals) necessarily have longer lead times, if that strategy cycle feels too long, your delivery cycle is probably too long as well.

Also remember that some aspects of strategy, particularly the more capital intensive ones, necessarily have multi-year lead times. OKRs are primarily built for digital projects and would need a significant rethink for goals of this nature.