> ## Content Index
> Fetch the complete content index at: https://www.mastersinpublicaffairs.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Strategy lies in the difficult choice
- URL: https://www.mastersinpublicaffairs.com/essays/strategy-lies-in-the-difficult-choice/
- Published: 2026-05-09T08:00:00.000Z
- Updated: 2026-08-11T18:48:51.000Z
- Description: Choice creates losers in the room, and only the leader has the political authority to absorb that cost.
- Author: Joseph Lavoie
- Tags: Good Strategy Bad Strategy, Richard Rumelt, #essay, #ai-3

> “Bad strategy is the active avoidance of the hard work of crafting a good strategy. One common reason for choosing avoidance is the pain or the difficulty of choice.”  
>  
> — Richard Rumelt, *Good Strategy Bad Strategy*

I have produced the dog’s dinner more often than I’d like to admit.

*Dog’s dinner* is [Richard Rumelt](https://www.mastersinpublicaffairs.com/episodes/good-strategy-bad-strategy/)’s term for a specific kind of failure. It usually starts with the desire to demonstrate that the team has been thoughtful. The file is complex, the client is sophisticated, and we want them to see that we’ve done the work. So we hit them with appendices, stakeholder maps, decision trees, tables that cross-reference jurisdictions and influence vectors and timing windows. The deck swells from twelve slides to fifty-six. By the time we’re done, the document is a monument to how complicated the situation is.

> Although, increasingly, I have noticed clients are put off by the thud of big strategy documents. They don't have time, and if we can't save them time while also designing good strategy, why did they hire us in the first place?

The client signs off. They thank us for the rigour. We feel like we’ve delivered.

We haven’t. We’ve delivered a wall of complexity that proves the team has been busy and produces no decision the client can actually act on.

For a long time I told myself this was just the cost of doing complex work. In truth, the wall of appendices isn’t what the client required. It’s what I produced when I avoided the actual work.

The actual work is the kernel: diagnose what’s going on, decide what’s pivotal, refuse the rest, concentrate the result. That often takes a *single* page, not fifty-six slides.

Rumelt spent forty years inside boardrooms as a consultant and a UCLA Anderson School professor, watching organizations produce documents called “strategy” that he’d argue weren’t strategies at all. [*Good Strategy Bad Strategy*](https://www.mastersinpublicaffairs.com/episodes/good-strategy-bad-strategy/), published in 2011, sits at the centre of a good strategy reading list because it takes a structural question seriously: what makes something a strategy at all.

Most strategies fail, Rumelt argues, because senior leaders make small political accommodations that, in aggregate, replace the strategy work with something else. Something that satisfies more people in the room, hurts no one’s interests, something everyone can sign.

> “Universal buy-in usually signals the absence of choice.”

Put another way: death by consensus.

Strategy, in Rumelt’s framing, is scarcity’s child. Without the constraint of having to choose, there is no strategy — only a list of things the organisation hopes will happen.

The diagnostic test Rumelt offers is the kernel. Any real strategy has three irreducible parts:

- A diagnosis of what’s actually going on
- A guiding policy for how to deal with it
- A set of coherent actions that carry the policy forward

Pull one out and the strategy stops being a strategy. The piece most often missing is the first, because diagnosing means choosing — saying that one factor matters more than another, knowing the second factor has a constituency in the room who’ll be disappointed. Diagnosing creates losers. The room is full of people who don’t want to lose.

If your organisation isn’t producing good strategy, the first place to look isn’t the team or the framework. It’s the leader. Strategy work can be delegated. Choice cannot. Choice creates losers in the room, and only the leader has the political authority to absorb that cost. The leader’s job is to refuse the document everyone can sign and send it back. To inflict the pain the team will not inflict on itself.