Jeff, one of our product advisors at Awell and a mentor of mine once said that the rhythm of a drumbeat is a perfect analogy for the cadence involved in building a product. Similar to the beat of a drum, the product development process needs to have a steady, predictable, and consistent tempo. Without it, the process can become disjointed and inefficient.

The analogy always stuck with me. Not only because I like analogies to express problems or situations in an alternative and sometimes absurd way but also because it helped me visualize and think about the structure of how we should build a product and make better informed decisions on what to build when.

Awell’s Drumbeat

At Awell, we have the following framework that sets our rhythm and cadence:

1. Our mission: what is our purpose and why do we exist?
2. Our vision: where do we want to be 5 years from now?
3. Yearly objectives: what do we want to achieve this year in order to get to our long-term vision?
4. Quarterly objectives: every quarter we set OKRs.
5. Build cycles (Shape Up): the cadence (± 6-week cycles) at which we develop and ship new features

Where the mission and vision are - purposely - more aspirational, objectives and product cycles are where the strategy or plan of action is outlined in order to reach our desired future state. It's where you execute in order to achieve your mid-to long term objectives.

I would also like to highlight that we use this framework to make decisions about what to build and when. There are 100 things we could build into the product and without a proper framework to help you prioritize these efforts it is difficult knowing where to start. We define objectives to help us reach our vision and we build things into the product that help us reach our objectives. On the other hand, the vision can help you set your objectives and objectives can help us prioritize product work. These are communicating vessels.

The framework outlined above also sets the cadence for most of our internal meetings (we are mostly async so we don't have a lot of meetings anyway):

• Every year at the beginning of the year we set yearly objectives
• At the beginning of every quarter, we set OKRs
• At the end of every quarter, we review and close OKRs
• Every 2 weeks we have an “All hands” meeting where we report on the progress of our OKRs
• Every week we have a Show & Tell where we demo progress on new features

This steady and predictable cadence & rhythm is what we call our “Drumbeat”.

Shape Up’s Drumbeat

Although Shape Up is a more on the microscopic level, compared to mission, vision, and yearly objectives which are more on the macroscopic level, it does also have a drumbeat. As described in the Shape Up book you have build cycles of 6 weeks with a 2-week cooldown period.

The cooldown period is a period with limited to no scheduled work where the engineering team can breathe, meet as needed, and consider what to do next. We use it to fix bugs, address technical debt, explore new ideas, do tech discovery for pitches, or try out new technical possibilities.

So basically the cycle time is 8 weeks as between the start of 2 cycles, there are 8 weeks in total. What’s interesting to note here is that it is a 3:4 ratio. Six weeks of building and two weeks of cooldown.

Being Offbeat

One of the first practical problems we experienced when we started implementing Shape Up is that a cycle time of 8 weeks did not match our "drumbeat". We set OKRs every quarter and a quarter has 13 weeks. Neither a single 8-week nor two 8-week cycles fit nicely into that.

That gave us a couple of options:

1. Accept that we are “offbeat” with our product development cycles
2. Adjust the cadence at which we set OKRs to match with cycle times of 8 weeks
3. Adjust the cycle length

It became clear quite quickly that there was not a lot of appetite from the business to have an “offbeat” product development cycle. And for good reasons because as described above these are communicating vessels so it doesn't make a lot of sense to run them out of sync. Neither did we feel that changing the cadence for setting our objectives was something we should do.

Beat Sync

That means only one option was left to sync the tempo of the different drumbeats: adjust the cycle length so it fits nicely into a quarter.

We still wanted to have cycles that are long enough to build something meaningful start-to-finish and short enough that everyone can feel the deadline looming from the start. Which meant that a single 13-week cycle (10+3) was not an option.

Quarters have 13 weeks, which unfortunately is an odd number so we accepted the fact we would have 2 cycles of different lengths within a quarter. Additionally, we wanted to adhere to the 3:4 ratio described above (building features vs. cooldown). This means that in a single quarter of 13 weeks, we would like to reserve at least 3 weeks for cooldown.

We put those parameters into a formula at this is what we ended up with:

• A 6-week cycle (5 weeks + 1 week cooldown)
• A 7-week cycle (5 weeks + 2 weeks cooldown)

Although this is merely a purely practical problem that can easily be resolved. It took us over 2 years to get to this setup where Awell’s drumbeat is in sync and at a rhythm and pace that works for the whole company. A seemingly easy problem to solve, but figuring it out and try different setups is probably what took most of our time.

I am curious about what local adaptions you have made to Shape Up’s cycle length and how it fits into your organization’s drumbeat!