My take on why goal cascades are harmful and what to do instead

If collective goals matter, you need some way of creating coherence.

Duplicate efforts: “Oh, I didn’t know you were doing that too!”; Conflicting efforts: “We need to make the pegs round.” “We need to make the holes squre.”; Missing efforts: “I assume someone is dealing with the bridge crossing problem?”
Duplicate, conflicting, or missing efforts

The typical way of creating coherence is by cascading goals

The best version of cascading goals works as follows:

  1. The top leadership of the organisation determines an overall vision and strategic intent;
  2. This is delegated to the next level of the organisation who translate it to their local context AND provide feedback;
  3. This continues for each level: convey intent translated to local context, back-and-forth feedback.
“This is our overall vision and strategic intent. Please translate.” Picture of goals cascading to each level of an org tree with feedback loops at each level.
Goal cascade

Cascading goals are a sub-optimal way of creating coherence

There is no reason to assume that the primary driver for a goal should always come from the top of an organisation. The senior-most leaders of an organisation do not have perfect knowledge, nor perfect reasoning and depending on the specific context, they don’t even have the best knowledge or reasoning.

“Wow, that’s brilliant!… unfortunately you’re not the leader so we’ll have to align to this other goal instead.”
Goals don’t always need to come from the formal leaders
“That would have been the right goal to pursue last month but things have changed since then.”
Goal cascades are inherently delayed
“Yes, we probably should have ensured the core service stayed up but we didn’t have any aligned OKRs on that.”
Optimal action doesn’t come from just aligning to top-down goals

Persistent models enable parallel goal setting

Pace layering refers to how different components or “layers” of systems change at different rates (i.e., they have a different pace of change).

  • The overall shared mission;
  • The shared model of reality, that is, the coherent overall strategic narrative based on data and insights;
  • The shared beliefs on how to succeed (aka doctrine)
  • Specific bets / projects;
  • OKRs (annual, quarterly);
  • Delivery plans
Shared persistent model consisting of Mission, Model of reality, and Beliefs allowing teams to plan in parallel.
Persistent models enable parallel planning

Regular synchronization addresses any inevitable drift with parallel goal setting

Even with shared, persistent models, it is inevitable that parallel goal-setting exhibits some kind of drift. Regular synchronization addresses this drift.

“Let’s sync our understanding of direction and strategy” — multiple teams synchronizing at the same point-in-time. Otherwise, “Every level is moving forward simultaneously.”
Point-in-time planning events are more about regular synchronization

Develop persistent models as a network, not just top-down

Instead of a top-down tree where teams are waiting for visions and strategies to come down from on-high, think of every node in a network developing an opinion about vision and strategy in parallel based on past communication and current information they’re seeing in their local context and ongoing synchronisation with other teams.

Multiple teams each with local models synchronising with each other; shared model pulling from and informing local models
Develop models as a network, not just top-down

See also

Strategy deployment: from cascade to translation to synchronization. | by Jason Yip | Medium



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Jason Yip

Jason Yip


Staff Agile Coach at Spotify, ex-ThoughtWorks, ex-CruiseControl