””Ginny, a project manager, had a frustrating meeting with her boss, Henry, the VP of Engineering. His exact words to her were, “I don’t care how you get it done, but we need all of these two feature sets done. Now!”

She explained she needed more teams to finish the work. Otherwise, people would be multitasking.

He shook his head. “No way. We can’t hire anyone else and we need everything done. Now!”

She tried several more times, but Henry would not budge. He wanted “all” of it done, and now.

Ginny knew Henry’s request for multitasking would slow everything down. But she had options to finish the most critical work. She started by asking the project team to divide into two independent teams.

Divide a Larger Team Into Two Teams

The original team had twelve people. In my experience, “teams” of twelve people have difficulty jelling, and this team did. They worked as a team of four and a team of eight.

Ginny gathered everyone and explained the problem, that Henry needed “all” of it and now. She said, “If you are willing to divide into two teams, each team can take its own feature set. We might be able to give him more of what he needs.”

The twelve people “discussed” as they usually did—with four people talking and eight people listening. The four people said, “We need one more person to join us.”

No one wanted to.

The loudest person of the four said, “Fine. We’ll just do it without you.” Then they all left the room to organize their work.

The eight remaining people laughed and then discussed what they had to do to finish all the work.

The two teams then worked in parallel on the two feature sets.

After three weeks, the teams realized that the feature sets were not independent. To finish this work, the teams had to collaborate.

It was time to use one-week timeboxes to make progress.

Timebox Some Work With All the People

The twelve people returned to working together. Ginny spent most of the next three weeks on two critical jobs:

  • Supporting all twelve people as they managed their interpersonal relationships.
  • Working with the product manager to refine more of the feature sets and keep the two feature sets independent and discrete.

That’s when Ginny introduced the idea of right-sizing stories. (See How To Right-Size Your Stories For Better Predictability.)

Ginny suspected Henry didn’t really need “all” the features. But the more stories they could finish, the better everyone would be. And she was sure that when she told Henry they needed more time, he would want to know how much work remained. When each team right-sized their stories, she would have more predictability about the remaining work.

By the end of three weeks, each team still had way too much work remaining. It was time to alternate short timeboxes on each feature set.

Alternate Short Timeboxes With Everyone to Progress a Little

By the end of those three weeks, the twelve people had made some peace with each other. And everyone had a much clearer idea about both of the feature sets. But Ginny did not see how to achieve Henry’s demand for “all” of the work.

She met with Henry again, and he continued to demand “all” of it.

She asked all twelve people to work in one-week timeboxes, first on one feature set and then on the other. Here’s what that looked like:

  • Week 1, everyone worked on feature set A.
  • Week 2, everyone worked on feature set B.
  • Week 3, everyone returned to feature set A.
  • Week 4, everyone returned to feature set B.

By the end of four weeks, they had completed enough of the work that the twelve-person team split back into the four- and eight-person teams, who then worked independently.

Secrets to Avoid Multitasking

Ginny was able to avoid multitasking by using each of these secrets:

  • Divide a larger team so smaller teams can focus on one project or feature set at a time.
  • As a team, collaborate on one feature at a time.
  • As a team, alternate work on each feature set.

During this project, no one multitasked. While the people didn’t deliver everything Henry wanted, they did deliver most of it. And, once Henry realized what they did deliver, he realized he didn’t need “all” of it.

Multitasking is the fastest way to slow everything down. (That’s because teams have too much WIP (Work in Progress), which slows throughput of everything.) You can avoid making everything late with at least one of these secrets—even if senior people “insist” on multitasking.

As for the other problems in this newsletter, consider reading Reframe The “How Much” Conversation To “How Little” to address the “all” of it problem. See Practical Ways to Lead and Serve Others for Ginny’s team-based approach to managing this group of people. Manage Your Project Portfolio discusses why Henry had these problems and how managers can stop the need for multitasking. And Manage It! Your Guide to Modern Pragmatic Project Management has more defensive portfolio management for project and program managers.

Learn with Johanna

The consulting book is with my editor. I hope to be able to report on more progress next month.

I posted this request last month, but I’m still looking for one more client willing to discuss and maybe experiment with agile career ladders. A dual-track ladder is insufficient—we need three tracks:

  • A “technical” track focused on how well people influence and coach others about the code and the solutions inside the product.
  • A “product/process” facilitative track focused on how well people influence and coach others about how to work better. (This is the product leadership, project/program management, agile coach, etc. track.
  • A leadership track focused on how well leaders and managers create and reinforce a culture that offers them the business agility they need.

If this interests you, please reply, and we can chat.

New to the Pragmatic Manager?

Are you new to the Pragmatic Manager newsletter? See previous issues.

Here are links you might find helpful:


© 2023 Johanna Rothman

Pragmatic Manager: Vol 20, #1, ISSN: 2164-1196

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *