Coffee with OKRs · 6 min read
Companies have mastered the mechanics of goals. Alignment loses to bonuses, tools and habits
More than ninety quarterly objectives. That is how many there were in an organisation of over a thousand people, where the structure of objectives took the natural route and copied the organisational structure: strategy and annual goals, product objectives for the quarter, and under them an objective for every team. It was our own implementation, and we called that state alignment, which it wasn't. With that many objectives, agreeing on a direction stops being a value and becomes an administrative cost.
From the episode · in Polish
Four areas of alignment you can check one after another
In this episode we break alignment down into four things you can check separately: a direction linked to daily work, shared objectives set around the product instead of cascaded down the structure, real collaboration of working groups, and transparent communication of conclusions. Along the way there is the story of our own implementation in an organisation of over a thousand people and the moment when an entire level of team objectives was abolished there. The part about bonus systems that reward silos is in the second half of the conversation. from 00:33:15
The recording is being edited. Leave your address and you'll get the link when it's ready.Let's start with what the word means, because in Polish it is sometimes rendered as levelling, and that is misleading. Alignment doesn't mean doing what the CEO said, because that is simply obedience. Nor does it mean standardisation or centralisation: teams are meant to keep their identity and their potential, just pointed in one direction. It also isn't about every team contributing to the strategy, because that is physically impossible and some departments will always be in a supporting role. It is about a state in which every team knows where the company is going, can judge whether it has a chance to contribute, and can at least check whether its daily work contradicts that direction. Below roughly fifty people this can be handled with ordinary communication, because everyone can be gathered in one place. Above that, structure, the number of channels and inertia all grow, and with remote work the problem appears as early as twenty.
The order of work on this is fairly concrete. First, a direction that can be linked to anything at all: a strategy reduced to a sentence about wanting to grow doesn't help you make any decision, because then the answer to every client who increases revenue is yes, even if serving them forces a change to the product, the plans and the cost structure. A good strategy answers the question of what we will not do. Then the links have to be shown openly, not just held in people's heads: the strategic goal, the annual goal, the quarterly objective and the backlogs drawn in one place, even on a shared board, because the drawing itself builds awareness of what follows from what. Next come shared objectives, the ones set at the level of the product or service, jointly by representatives of the teams, with people chosen only afterwards to fit the objectives. A working group is formed by skills, for one cycle, which isn't always a quarter, and nobody is seconded to it full-time, because that never works anyway.
The hardest is the layer nobody implements deliberately. The most systemic block to collaboration is a bonus system that rewards individual or departmental results: it is hard to expect a person drawn into a working group to act against their own financial interest. One floor down sits a conflict of tools, and sometimes even of ways of working, because one department works in an agile way, another in a classic way, and a third on sticky notes. So the group has to agree on one tool, as simple as possible, for the duration of the cycle, and plan in it together, review progress every week, and draw conclusions about whether these actions are actually getting it closer to the result. None of this can be replaced by an announcement that we now need everyone's engagement, because people's first thought after a sentence like that is: what does that mean in practice, and what am I supposed to do. And one last thing, which only shows after a while: the cost of agreeing rises with the number of objectives. With several dozen objectives, alignment becomes a separate event lasting several hours, as in an organisation that, halfway through a six-month cycle, set out several dozen laptops with the objectives so that everyone involved could read them and consult one another.
Takeaways
Four things worth looking at before you decide OKRs don't work
- 01
Alignment is not obedience
Nor is it lining everyone up with a ruler. It is a state in which a team knows where the company is going, can judge whether it can contribute, and can check whether its daily work contradicts that.
- 02
Objectives around the product, not down the structure
Objectives are set jointly by representatives of the teams, at the level of the product or service, and people are chosen only afterwards to fit them. A cascade down the organisational structure gives you as many objectives as there are teams, and blurs the company's direction.
- 03
Check the bonuses before you ask for collaboration
A bonus system that rewards individual or departmental results is a systemic block, not a question of goodwill. A resource problem is usually a priority problem too: as long as it isn't clear what matters most to the organisation, the more important projects will always win.
- 04
The cost of agreeing rises with the number of objectives
With several dozen objectives, alignment is a huge effort and costs a great deal of time. Simplifying the structure and reducing the number of objectives lowers that cost, and that is one of the stronger reasons to do it.
The whole topic in one place
How to talk about OKRs in your company?
None of these layers can be fixed with a presentation, but without a well-told direction there is nothing to hold on to. How to build that story, slide by slide, whom to show what from it and what not to say in it, we have set out in a separate text. There you will also find a list of the other episodes on communicating goals in a company.
Author

Lena Zhukova
founder; strategy workshops, quarterly sessions
Founder of OKR Poland. For twenty years she has worked with boards of companies from 30 to a few thousand people; she wrote dozens of strategies with them and helped implement them in more than 40 organisations.
Next texts by e-mail
The OKR Poland newsletter
We write when we have something to say. To start: eight hours of recording on preparing a company for strategy workshops, plus the OnePager.
