"Really?? Blast... I'm not looking forward to THIS retrospective!"
The Empire doesn't use Waterfall. Its ruthless efficiency and innovative technological supremacy are clearly the outcome of Scaled Agile with an entire leeegion of its finest scrum masters awaiting every team need.
And, in an organization where you might be Force-choked to death after one mistake (I guess that's one way to "build projects around motivated individuals"), an extreme challenge for a scrum master or coach is to conduct an effective retrospective after a huge disaster has the team throwing blame around like Sith apprentices in a Force-lightning class.
Mistakes still suck, though, regardless of how awesome your company culture and team might be. We don't enjoy making mistakes, and we enjoy talking about them with others far less. Who hasn't dreaded one of "those" retrospectives at some point?
When something has gone badly wrong and thinking is clouded by anger, frustration, or blame, some corporate cultures might find it almost impossible to objectively assess the facts and root causes, then give the team an opportunity to clear the air in a productive, constructive way.
Yet, it's critical to learn from disasters so they don't happen again!
How best to frame up complicated yet vital discussions when morale is low and emotions are high?
Start by building a shared, chronological view of facts to then use as a framework for highlighting significant events and individual emotions.
The Timeline Retrospective and its three-step approach is an ideal format.
the complete Timeline Retrospective template: linear + significant events on top, emotion map beneath.
The Timeline Retrospective works just as well for small teams coming out of wonky sprints or bad deployments as it does for root-cause analysis and broad, many-month project postmortems. (Or, for the purposes of my staged whiteboard photos to follow, disasters on a scale with the destruction of the first Death Star.)
Step 1: Create the timeline of events.
The team was given pens, post-its, and 10 minutes, and asked to create a linear sequence of all the events leading up to the destruction of the Death Star.
Everyone quickly agreed on two events: the Rebels steal the plans, and, some time later, the Death Star is blown up.
But what else happened?
"Hmmm... didn't that wretched freighter have something to do with this?"
"You mean when that smuggler fiend picked off Lord Vader's wingman in the trench?"
"Ah, good point! Write that down. But what about when our Death Star caught it in the tractor beam as it was coming out of the, ahem, remains of the Alderaan system?"
"Oh yes, I remember! Our stormtroopers were tasked to stop it from leaving, because it had the plans on board."
"The plans -- of course! How did they wind up in that troublesome bucket of bolts?"
"You mean we had our hands on them, and let them go? How?"
"Well, Lord Vader convinced Tarkin that we could track the freighter back to the Rebel base. It nearly worked; we very nearly wiped out their entire base and all their fighters. If only our defenses had been tighter... if Vader hadn't been disrupted after he lost his wingman..."
Wow! It's challenging for one person to single-handedly construct a timeline of events. Groups, meanwhile, will spark memories and other cues in each another, bringing their individual puzzle pieces together to form one larger picture.
(Case in point: even though I've watched Star Wars many dozens of times, I didn't remember everything when I was staging these steps on my own. For example, I initially forgot Uncle Owen and Aunt Beru's role in the stolen plans narrative on Tatooine.)
Engaging the team with remembering the facts objectively, and beginning to analyze them critically (rather than "emotionally", which comes later), are the main benefits of this first step. The result is a shared view of events based on multiple pieces of the core truths minus the distortion of false perceptions and hidden agendas. This task of sharing also helps to reunite team-members wedged apart by resentment or mistrust.
Some teams keep retrospectives private in order to speak freely.
I say, train managers and leadership to respect the team and participate impartially.
Regardless, if a manager presses to attend, the team has the right to refuse.
Step 2: Upvote the significant events -- good and bad -- with dot-voting, and discuss.
Once everyone felt comfortable with the events and the general sequence, the timeline was ready for further analysis. The team was asked to vote, as individuals, on any events in the timeline that they personally felt had a significant impact -- positive or negative -- using colored dots and/or symbols. (Red = negative, blue = positive.)
Note: events and post-its are linear, but to maximize space, teams often spread them out over the entire whiteboard. The height has no bearing on or meaning for an event.
Certain events may get positive and negative votes simultaneously. Any technique which uncovers healthy variances in team-members' individual perceptions will allow for critical insights and defuse the likelihood of group-think.
With the dot-voting complete, the team began to ask questions.
"Why would anyone view Kenobi's death as a negative event? Surely that was a high point?"
"Because I've since spoken with Lord Vader and he concedes that killing the Jedi Master may have, in his words, turned him into a 'Force spirit' -- yes, I know, utter nonsense -- who was then able to possess one of their X-Wing pilots and successfully guide him to launch his proton torpedo into the thermal exhaust port."
"Interesting. Well, it looks as though we all agree that the one event having the greatest negative impact was when that freighter --"
"The Millennium Falcon."
"Ah, yes. Gold star for knowing the name. The most negative moment of all was when the Millennium Falcon escaped."
"Quite. Had we managed to prevent it from returning with the plans to the Rebel base, there's no way the Rebel scum would have been able to analyze the plans and launch their attack."
"And, if we go even further back, it should be noted that we failed to find the plans after they were jettisoned from Princess Leia's consular ship in that escape pod. Shocking security lapses at Mos Eisley."
"Oh, I see! We'll just completely ignore the fact that the Death Star had inherit design flaws in both its defense systems and that bloody thermal exhaust port, shall we?"
"Well, clearly that's another area upon which to improve in the next go-round."
"What? We're building another one??"
"Hmm. 'Jettisoned. Jettisoned.' You know, I rather love saying that."
Inevitably, the team will start to critically examine the sequence of events, looking for patterns and tracing root causes.
Step 3: Map team emotions alongside the timeline of events + draw up conclusions.
How did everyone feel about these events?
Using separate colors, each team member was asked to draw a parallel wavelength or "seismograph" of their emotions as they related to the events above, then describe their emotional journey to the team in turn.
Here's a fictional narrative to accompany one of the emotion lines:
"I'm the blue line... As you know, I'm a security analyst. Well, the Death Star thermal exhaust port design flaws were upsetting, but not nearly as bad as the fact that we just didn't do a damn thing about it. That was incredibly depressing. Our bureaucracy can certainly slow things down at times.
"My lowest point, other than the time leading up to the destruction itself, was when we couldn't find the plans on Tatooine. We had a few chances and we blew them all. I can't control the design flaws, but the security lapses made me feel helpless. I was getting no information in the field, nothing tangible. If only there were a way I could have rendered the plans themselves unreadable... This was a real low point.
"Then, I heard reports that Lord Vader had executed his Jedi nemesis, Kenobi -- these Jedi being such spiritual leaders for the primitive Rebels -- and I felt we were regaining the upper hand. Crushing their spirits like that gives me a reason to live! Then, when we lost the freighter, and when the Rebels launched their attack, I was despairing again until we launched our TIE fighters, where I had a brief uptick. But ultimately we ran out of time. We had the Rebel base in strike range. Who knows what we could have done with just a few more seconds!
"My takeaway from all of this is that, design flaws aside, we needed more time. And we might have had it if the plans themselves been encrypted. I suggest we begin immediately investigating file-level security procedures and encryption methods as one of our actions from this exercise."
Looking at the plotline of Star Wars: A New Hope, the Imperial team assembled here determined that the design flaws of the Death Star were not as significant in the long-run -- or as easy to be prevented -- as the multiple security lapses and lack of file encryption which allowed the Rebels to return the plans to their team on Yavin IV, then proceed to quickly analyze (and exploit) them.
Using this incredibly nerdy theoretical example to draw correlations to real-life situations while explaining a particularly useful disaster analysis technique was a ton of fun, and I hope you enjoyed it, too. Poor Governor Tarkin.
May the Fourth be with you!
All content, character names, script references, designs, plot elements, and anything else pertaining to Star Wars are the registered trademarks of Lucasfilm Ltd. and the Walt Disney Corporation!
Additional Examples and Resources
a SoundCloud iOS and Android joint team timeline retrospective. Used with permission.
Above is a timeline retrospective that I facilitated as a root-cause analysis exercise for the SoundCloud mobile engineering team following a new feature deployment that had some hiccups. (We used large post-its so that remote viewers on our teleconference could follow along.)
- As we collaborated on the events, the team decided to split an iOS-specific path away from the Android path, then brought the two paths back together at the end. (We adapted the format to the context and realities at hand.)
- As they worked, I noted down broader themes alongside particular events to fuel discussion and analysis.
- The team opted not to do an emotion map, but the dot-voting of positive and negative events gave them an opportunity to touch on how they felt at the time.
A staged analysis of the Revolutionary War
Another description with detailed instructions for the Timeline Retrospective, using the far more stuffy example of the American Revolutionary War, can be found at the bottom of last month's blog post.)