TBM 280: 2-6-4-1

A couple of observations:

• Teams with a quarterly goal often wrap things up in the final week or two of the quarter with no time to iterate. It is a mad rush. They finish if they're lucky—sometimes things keep slipping (work expands to fill the available space, and then some).

• Teams waste a lot of time prepping for the next quarter, but they do so in ways that aren't inclusive and don't get everyone on the same page. Two weeks in and they are “back to square one”.

• Six-week blocks are great, but teams frequently feel like they left value on the table when they rush off to the next thing.

• Set sprint lengths become monotonous and an artificial construct. It is fun to switch things up.

If these points sound familiar, here's a fun way to structure a quarter. It's a nice change of pace and helps a team go through the "full cycle" of learning, building, measuring, learning, etc.

• 2w research and discovery

• 6w get something out to customers

• 4w iterate 

• 1w tie up loose ends

Warning Label: This is one of a million experiments you can run to address these issues. The details matter less than the intent. Roll your own!

2w research and discovery

Do ~2w of full-team research. Invite external partners as needed. Run a solid kickoff on Days 1 and 2 centered around a solid opportunity. Plan direct and daily contact with customers. Get the help of an unbiased facilitator to keep things moving. Make research and discovery everyone's full-time job for the full two weeks, but encourage splitting, pairing, and "divide and conquer" where it makes sense. It is OK if there is downtime. Generate whatever docs, drawings, mocks, etc., you need to build shared understanding and increase confidence across the team.

Yes, this may seem like a long time, but when you add up all the ad-hoc 1:1s, planning meetings, meetings to proxy research, and partner syncs, you'll likely spend the same amount of time but do so far more effectively with less telephone.

6w get something out to customers

Timebox ~6w to get something into the hands of real customers (not a prototype—something you're comfortable leaving in production). If you've tried this in the past, you'll note that 6w is long enough to get something substantial done (unlike a 2w sprint) but also short enough to keep you on your toes. You will need to make significant progress by the end of week 1 to have a shot. If this is your first time trying the 6w thing, it pays to be conservative. Do something small with lots of attention to detail over swinging for the fences. Encourage regular informal demos, discussions, brainstorms, and "help unblock me" meetings.

4w iterate

Timebox an additional ~4w to iterate on that thing based on feedback, with the goal of a minimum of 4 updates (weekly). You will likely need to do some analytics instrumentation while maintaining close contact with customers. This block is the team's opportunity to do that precious follow-up they never have time to do. Make it count.

1w tie up loose ends

Finish the quarter with a week of tying up loose ends. This week doesn't need to be structured. There will be a long list of small tasks you'll need to crank out before moving on, including docs, enabling other teams, running a good retro, etc.

Throughout the quarter:

Make this the only thing. For this one quarter, make this effort the only effort. This is it—the only OKR. It is a big block on the roadmap.

Clear the decks of all other work. If reactive and unplanned work occupies more than 20% of your time, consider doing this experiment when you can get that number below 15%. Otherwise, context switching will become your #1 priority, and that's silly. Include in this calculation all of the Slack interrupts, impromptu support for other teams, random questions, and "can you just _ _ _ _ for a couple of minutes" interruptions.

Be very aggressive about limiting unnecessary meetings and interruptions. Start with the premise that each team member needs a minimum of Nhrs (6?) of uninterrupted focus time daily, and triage/cut meetings accordingly. This rule applies to managers and PMs as well. Nothing's worse than a team that can't contact a PM when they are cranking.

Normalize slack time. If you are doing this with your whole team (which you should), you will encounter times when people are waiting. Encourage people to pick up small things that interest them until they can rejoin the "mainline." 

Keep a running history for all to see. Be transparent about what’s happening, and set the expectation for daily progress. Swoop in on blockers immediately.

Remind people that this is an experiment. It has an expiration date (the end of the quarter); at this point, you'll be fine going back the old way.

If you get pushback, challenge people with "What's the worst thing that can happen? We finish something?" Double down on the team's commitment to follow through and not let things slide. Note that six weeks is less than half of a quarter, and quarters typically become a big old sprint. Remind people of all the long-running efforts where a quarter goes by, and nothing changes. Go back through historical data and shine a light on quarters where—with all the pomp and circumstance of big plans, sprints, blah, blah, blah—nothing all that substantial happened.

Final observation. Experiments like this can seem “crazy” on the surface. Somehow the status quo busy-ness and multi-tasking feels so much more productive. But after running one or two, you may find sentiment shifting.