Why retro?

Retros are a time for reflection. They’re a key part of how a business gets to learn and grow.

You may not see the value if you’ve never done them, and they can feel like a waste of time if you’re focused on moving forward. Businesses are focused on progress, and building new things is generally how a business creates more value.

It also is hard to do retros right. If your business is going through a rough time, a retro can just feel like a time for reiterating all the ways things are going wrong. But, if you don’t do retros, you don’t have a place for people to talk about how you can fix the business’ problems.

If you do the retro right, you make time to talk about positives as well as the negatives. Even if you don’t have many positives, just allowing people to talk about the negatives in a constructive manner, with a focus on fixes, helps your team move forward.

How do I do it?

Decide on a format

Lots of people have of opinions on format. They’ve written things that can be found via your favorite search engine.

What has worked well in teams I’ve been on in the past:

  • Have prompts for both good things that happened as well as bad things that happened
    • Keep the prompts generic enough that people don’t have to think hard about which category a piece of feedback should go into
  • Allow people to add items on their own. Some people will add a bunch of things, others may add very little.
    • Forcing everyone to go around and think of one thing makes for a painful meeting with little creativity.
  • Pick a specific amount of time for the meeting, and stick to it
    • There will be natural lulls as people think of things, and sometimes people adding things makes other people have ideas. Ending the meeting early can hinder that thought process.

A format that became a new favorite of a team I’ve been on is “Rose, Bud, Thorn”.

  • Rose - “Roses of achievement”, a space to celebrate what went right
  • Bud - “Budding opportunities”, what opportunities have we recognized that are available, should we choose to pursue them
  • Thorn - “Prickly thorns”, what hindered us from meeting our objectives

We set up a Figma document with those categories, set a timer, and then everyone individually adds items to those categories. People add reactions to ones others made that they resonate with during that time. After the timer runs out, we go through items and have people speak on or expand items that they added. There might be some light discussion for items that other people had thoughts on, but we try to timebox those discussions so that we can get through all of the items in the allotted time.

When relevant, we add action items based on the discussed items. Those action items get assigned owners who are responsible for getting them done.

Get everyone in the room

A feature has launched.

Joe wrote the PRD for it, then handed it off to the engineers. Joe moved on to the next project.

Joe comes to the retro.

Lisa, Joe’s boss, approved Joe’s PRD. Lisa comes to the retro.

Dan wrote the feature, and Jennifer QA’ed it. They both come to the retro.

Run the retro

Assign a facilitator, start the timer, add your discussion items, and then discuss!

Making your retros matter

Retros need to be blame free

Everyone involved in the project is working for a business, and that business needs to make money for everyone to continue to be employed. Since the point of the retro is to help the business learn from the past, people need to be able to talk about inefficiencies in the business without personal feelings of guilt or blame being in the way.

Joe and Dan need to be able to have a conversation about how Dan didn’t get enough detail from Joe. Joe might’ve had a million projects on his plate, or a death in his family, but Dan had to rebuild part of the feature twice because Joe didn’t fully explain what Dan needed to build. That’s not good for the business. The team needs to be able to have a conversation about the problem for the business to be able to figure out how Dan can get his questions answered faster. Hurt feelings keep the team focused on emotions instead of solutions.

Jennifer feels bad that she couldn’t get to QA’ing Dan’s work quickly because another team was in line first. She feels that she should have been able to get through that other teams’ QA faster. Another inefficiency that should be discussed, and the team should be able to discuss whether the prioritization of Jennifer’s workload makes sense without her blaming herself. Plus, maybe Dan felt that Jennifer got through QA’ing his work incredibly fast, and Jennifer’s feelings of guilt weren’t warranted.

Follow up and iterate based on what you learned

Lisa thought Joe’s PRD was incredibly well researched, and the feature ended up fitting the users’ needs far beyond anticipated. The company made a zillion dollars off of the feature. Given the correlation, the team agrees that Joe should do that same amount of research in all future projects.

Joe would love to put that same amount of research into every project, but he’s struggling to do so because it’s also his job to respond to all of the user feedback the company gets, and there are hundreds of reports a week.

If the team agrees that the business impact of Joe researching projects outweighs the impact of him answering all of the user feedback, then the team needs to discuss solutions that will allow Joe to do that research without making customers feel like they aren’t heard.

Nothing makes a team lose morale more than giving the same feedback again and again without the business changing.

This is also why it’s so important that everyone involved in a project has to come to the retro. If Joe wasn’t there, the team wouldn’t be able to have the conversation about how much impact that Joe’s research had on the success of the project, and Joe wouldn’t know how important it was to continue to do so. Lisa also wouldn’t get that feedback, and wouldn’t know that there’s a need to move other work off of Joe’s plate.

Maybe there was a new hire coming along that would do research instead of Joe? Only Lisa would know!

Allow conversations to breathe, but don’t allow for them to carry on too long

Joe agrees that Dan needed faster responses from him, and that it’s an important problem to solve for the business to be more efficient. Joe and Lisa start discussing solutions.

There’s definitely value in allowing for some of that solution chatter to happen - if there’s an easy solution right there, then there’s no need for an action item to define the solution, and sometimes discussing a solution can make a problem feel less important to solve. There’s also something to spontaneous conversation creating great ideas!

The retro facilitator needs to allow some space for that discussion without allowing it to overcrowd the meeting and not allow for other discussion items. Jennifer might really want to discuss how she thinks giving others more visibility into her backlog will help the business be more efficient, but Lisa might need to hop to another meeting before Jennifer gets a chance to bring it up, causing the team to miss out on Lisa’s input, which could have included news about a deluge of projects coming Jennifer’s way.

Be creative, and come with an open mind

Let’s say that Lisa was dissatisfied with the amount of time the project took overall. Since she had reviewed Joe’s PRD, she was pretty sure that Dan had everything he needed to start on the project. Therefore, the reason the project took so long was because of something Dan or Jennifer were responsible for.

If Lisa starts with that assumption going in, then Dan and Jennifer are put in a position of having to change Lisa’s mind to get the true problem - the lack of speedy communication between Joe and Dan - solved. Lisa’s assumption was also a blame statement, meaning that Lisa has already put herself in a position where she’s less likely to believe Dan when he points to the real problem.