Retrospectives – Jessie Link | The Lead Developer New York 2017
Articles Blog

Retrospectives – Jessie Link | The Lead Developer New York 2017


>>Hi, everyone, I’m super-excited to be here
as part of this amazing lineup of speakers, although it’s very stressful going after someone
like Lara Hogan earlier in the day because now I’m thinking about the structure of my
thing, I’m thinking about spiders and do I have to go to the bathroom and it’s all very
stressful but I’m going to try to keep it together up here. So this is me. And as Meri said, I’m a Director of Engineering
at Twitter. I helped lead engineering ops up in London. I supervise teams both in London and Boston. I used to be at living social before that. I also used to be a member of the US Air Force,
so I served for six years and so over the course of my career I’ve led lots of different
kinds of teams in lots of different kinds of environments. I’ve led small teams, I’ve led teams of teams,
and along the way I picked up some habits one of them was a scrum master, which was
when I first heard about this thing called retrospectives. So when I thought about the talk I wanted
to give ultimately I wanted to equip you with some practical advice that you could bring
back to your teams right away. So this is a very tactical talk. And most importantly I want you not to just
understand those five types of retros, but when you should use each type of retro and
why you should use each type of retro. And this is the participation part of the
talk. How many folks in here do retrospectives with
your teams at least every two weeks? Pretty good. So some hands up. A lot of hands not up, though, and that’s
something that has sort of troubled me because I really believe in a the power of retrospectives,
they learn about retrospectives, and you talk about what went well and what didn’t go well,
you talk about what should get better and that’s the basis of their understanding for
retros, and so what happens is over time the retros feel very stale, you end up in that
groundhog meeting where it feels like every retro you’re going over the same things. And it’s pretty boring stuff. And it’s because people really don’t understand
what they’re doing when they’re doing retros and so they don’t understand the method behind
the madness, snoop they’re often poorly facilitated and that’s a very painful thing for me to
sort of sit and witness. There is an art to sort of facilitating a
group meeting and hopefully I’ll give you some tips on how to facilitate your retros
better today. And this is how you get into that groundhog
meeting which is we talk about what’s not going well but we actually don’t do anything
about it. And that’s again how you get caught in in
the groundhog day. Next time you do a retro guess what, that
thing is still wrong and so that item keeps coming up again and again in your retro. So a lot of mistakes people make. This first one I think comes up a lot particularly
if you’re a manager or a team lead where you’re also playing the facilitator role. What happens is you go in with your manager
hat and say I’ve got an agenda. But you really need to understand
that retrospectives, are it’s a form to empower the team to fix what’s happening themselves,
so it’s a chance for them to have input into their processes and if you go in there and
you already have predetermined what you think is wrong with your team and you’ve already
identified all of the things that you want to change about your team, then you’re going
to do the retro wrong because you’re not going to be receptive to the feedback that you’re
hearing in those retros. We also introduce bias into the sticky note
process. I’ve been in a lot of retros, and the important
thing about the retro is you want everyone to have a chance to put their input in in
a safe way without being biased by the other people in the room. And so what happens if I know something that
someone else is writing hi may feel like I need to put the same thing. Often when we collect the stickies, we put
up all the stickies up, then you start reading them one by one. That’s not a good thing. Whoever goes first you talk about 50% of the
meeting just about their one sticky and then by the time you get to the rest of the stickies,
you sort of run out of time. And the worst thing that can happen is you
just stop doing retros, or you only do them when something has gone terribly wrong. And I had this happen at a company where we
were sitting around and we had had a crisis with one of the build systems and we thought,
wow, that’s not great. We should probably do a retro. And I was like, ooh, this is not the time
to have that conversation. Waiting until things go terribly wrong to
have a retrospective is like waiting until your house is on fire to install smoke detectors
and I like that metaphor, because it kind of gets to the heart of why I think it’s important
to do retros. It’s not just about seeing what went wrong
yesterday or identifying the problems that are happening today, it’s about then being
able to do something about that smoke before it becomes a raging inferno that engulfs your
team. And I think a lot of people are mistaking
the concept of a post-mortem with a retro. Talking about feelings is actually very challenging
for teams and I think it was Eryn who was talking earlier about you have to build trust
with your team, right? And I think a lot of us having been in the
software engineering field we like to talk about data and complex things. Talking about feelings is awkward, it feels
weird about — I always joke that I have all the feels. That’s my defense mechanism because I find
it uncomfortable to talk about feelings. Sometimes it’s really difficult to understand. Well you’re angry, what does that mean? Why are you angry and most important, it’s
very scary to talk about feelings because to talk about feelings is to make yourself
vulnerable and some teams might not be there yet where they have that trust built up where
it’s OK to be vulnerable and what’s good about retrospectives is it gives you some guidelines
to surface all of those feelings that feels organic but the person also feels protected. So
the key takeaway is you can accomplish a lot of things with retrospectives, like I said
it’s not just about saying what went wrong, although it certainly is good to surface,
you know, sort of what happened, but more importantly it’s not just what happened, but
why did it happen? What’s the root cause? What are the hidden blockers beneath the surface. I’m not going to read the whole list to you,
but again because people have a very basic understanding of retrospectives, they think
all retrospectives are created equal. And so you end up in this situation where
you might not be using the right retro at the right time with your teams. And there’s a whole sort of different retrospectives
that you can do with your teams. And so you should think of your retros as
a complete set of tools that you should pull out the right time for the right job. This is me on my way to run a retro when I’m
having a good day. I’ve got my white board there, you know, I’ve
got my candy cigarette and I’m ready to go. So before I get into the five types of retrospectives,
I want to talk really quickly about the clustering pattern and why you should use it with your
retros. Like I said before, one of the key mistakes
people make are sort of reading the sticky notes one by one and then you’re end up in
this thing where you’re fixated on the content of the individual sticky. But what you’re really trying to do is draw
a pattern. Some persistent, some bad process, some code
smell that’s sort of holding your team back. So what I ask people to do is once the stickies
are all posted we pick team members and I say for the first category, I want you to
group them into two like clusters, so things that seem to be talking about the same general
thing, put them together and I want you to summarize what that cluster is about. And why is this powerful? Why do we do it this way? There’s Ann effect in psychology called the
echo effect and the simple act of mirroring back someone’s words to them builds empathy
hand builds understanding. So you will read the stickies as you’re grouping
them but I like to take it a step further and the reason I ask teammates to do it instead
of myself. I want them to put it in their own words. And that’s sort of what you’re doing, you
say can you summarize what this cluster is all about and again it lets you address the
larger issue. It’s not OK, this system failed last night
and the other thing. What you might see in this cluster is oh,
we’ve got an alerting problem. So you want to focus on so larger sort of
issues. The other thing is it shows your teammates
that they’re not alone. Whatever they’re feeling, whatever is bothering
them. I can’t tell you how many teams I’ve done
this with and they are shocked that someone else is feeling the same way as them. So that’s a really powerful sensation, it
really shows you that yes, we all do care and we all care about these same things. And I guess what if there are outliers? That’s OK, too. One thing you want to do is really want to
be sensitive to a team that’s not aligned. If you see someone who’s an outlier consistently,
it means that they are not thinking about the same priorities or things that the rest
of the team is thinking about and you can address that honestly and openly with them. If you don’t have any clusters, it may or
may not be a bad thing. You have to interpret, read the tea leaves
in the moment. But if you don’t see a lot of clusters, it
might be a sign that your team isn’t fully aligned. So let’s get right into it. The first retro I’m going to recommend you
all give a try with all of your teams is the one called the four Ls. So I’m going to tell you how to run it and
why you should use it and when you should use it. So for the four Ls, instead of talking about
what went well, what didn’t go well, what should we change, you ask the team to fill
out the four categories. I liked, I lacked, I longed for and I learned. There is a bit.ly link, I’ll post these tonight. So don’t feel like you have to write everything
down here. So why do I like the four Ls? I’m always comfortable with the what went
wrong category. I think a lot of times that feels very judgey,
right, and I think people are sometimes reluctant to dive into that. So I think wrong is a heavy word. But like lacked, I longed for, those are meaty
words and they’re very evocative and then you can sort of respond to that. I do get a lot of questions about what is
the difference between lacked and longed for. So generally lacked is things that you put
thing that weren’t really called for. I lacked a powerful-enough laptop. I can fix that with the IT team, I longed
for a modern and progressive front end framework. Those are persistent things that are sort
of bothering the team that they don’t have, and so you want to get it there. I also really liked the learned category. I think it’s really important, yeah, I think
another speaker talked a lot about how we learn so much from failure, right? And we don’t learn from success, so instead
of just fixating on what went wrong, but just saying what did you learn from it, it sort
of takes that failure and you’re trying to make a positive sort of learning out of that. Even if it’s I learned what color your face
turned when I push code that doesn’t have tests. Someone told me that one time. And it’s red. So don’t do that. I also like that it puts a first-person perspective
on it. So it’s really like how it affected me personally. And starting to build those things of trust,
because’ talking about this is how I feel about that, I liked that, I didn’t like that,
I’m lacking something to get my job done and that stresses me out and you can have really
good conversations aren’t those words. This is a great one. You can use this in your regular sprints if
you’re doing sprints or whatever your cadence is, weekly, biweekly, monthly. So you can use them early and often to go
in and diagnose root cause. Lacked is a ripe category to find process
improvements. If you’re running conferences, training, events,
it’s pa good one to sort of run with your participants, because again you can say what
did you like about the conference, what did you lack, what did you long for that I didn’t
cover so you can get some really meaningful outputs from your team with this one. Retro No. 2, draw with the sprint, I love
this one. You hand out blank sheets of paper and pens
of multiple colors and you say draw the sprint or draw the release or whatever it is. And that’s it and it’s intentionally broad. If you team really pushes back, I say whatever
it means to you, what happened in the sprint, how did you feel about it and afterwards we
all reveal the pictures and you talk about what’s going on in your picture. And it really is true that a picture is worth
a thousand words and these are words you won’t necessarily get folks to say out loud normally. There may be things that people struggle to
articulate using their words but by expressing things in a picture, they can express things
that they never would have been able to do in another form if it’s verbal. And. No er we a just not very talented drawing,
but we were learning, so this is good, right? What I loose like about it is by having people
to think in pictures, you get a lot of insight into how your team thinks about things and
it sort of gives you a lot of insight into their personality. One of the things we’re really trying to do
as leads, is we want to make sure we’re facilitating good conversation and communication in our
team and understanding how someone conceptualizes ideas, how they think about things, how they
organize their thoughts can really help you when you’re saying why is there a communication
problem? Oh, maybe it’s a left brain, right brain thing. This person likes to communicate in a very
linear fashion and they’re very detail oriented and they’re struggling to connect with someone
who’s thinking in more abstract concepts or they jump around before they come to a conclusion. And again you’ll see this, you’ll have drawings
that are express in a very linear fashion, they tell this very linear story, like point
to point. Some people may have drawings that are sort
of flowy and come back to a central core and have another set of drawings that spin off
on a new idea. Some people are all over the place. Remember before I said talking about feelings
is difficult. It’s very, very challenging, right and to
say I’m sad, that’s probably the most vulnerable thing you could ask someone to say in front
of their coworkers, most people will not say that. But you’d be surprised how many people will
draw this, right? They’ll draw an unhappy face, they’ll draw
somebody crying, they’ll draw a storm cloud. You might see someone who draws, that is pretty
good drawer, but maybe this person is stressed out, right? So it’s really interesting how people express
themselves when you give them that opportunity to think outside the box a little bit and
do it in picture form. I had someone draw something very similar
to this. So again, some people are quite linear thinkers
and they literally drew a timeline. Now, the interesting thing, it is fun to see
who’s talented right? So you might get this and you might get this. I don’t know. But what’s interesting about about this is
it’s still a slutter, right? These are still two people talking about the
same thing. Two people have cash on the brain. Some express it better than others, obviously. Maybe one is happy about it, one isn’t. So we have a team where independent of each
other, four of the ten developers, drew rocket ships and it was so interesting and I was
like, what is that metaphor, and oh, the sprint started slow and in the last few weeks we
were just cranking and it felt like we were on a rocket ship to the moon. Now, the quality of the rocket ship drawings
vary from person to person like this but you see these really magical moments that manifest
them selves in pictures. So when would you use this in? You use this to break up the monotony when
things feel boring, if you have a team that’s listless, maybe they’re a little disengaged,
this is a really fun one, because most people have never done a retrospective like this,
every time I’ve done this there’s been tons of smiles, people are laughing. Having said that it’s not one you can use
a lot. I would use it very sparingly, maybe once
a quarter because eventually you get tired of thinking of the drawings or you start gaming
the system. But so toss this one as sort of an ice breaker
and surprise the team with something interesting and new. Retro No. 3, a timeline retro. This one is pretty easy. You draw a timeline with an arrow at the right
end. And the stickies represent events. Good events, problematic events, significant
events. And you get a visualization that looks a little
bit like this. So you have two ways to sort of visualize
those concepts as good, bad, and insignificant. You can use colors. If you’re super-prepared, I have to scrounge
stickies up from wherever I can get them and I don’t have a lot of color choice, so instead
you have them go above and below the line to show what’s good and not good. This is like retros 101 because it’s easy
to understand. This one is very easy. Think about everything that’s happened in
the last two weeks and if you think about it put it on the sticky note and put it on
the timeline where it happened. So the team really understands how to put
these inputs. The thing is because it’s time based, not
significance-based people want to come up with something that’s
really powerful and significant, but a lot of things what’s really gone on with your
team is what happened the day before, somebody got sick, maybe somebody had a bad day, but
because people are thinking about it in time, they conjure those events up. So you’re going to surface things that might
not show up before. And this is where the clusters are quite interesting. The clusters will build themselves by the
nature of the event, right? So you can look at those spikes and say are
there any small things that might have tipped us off to that big spike, whether it’s positive
or negative. It tells a really crisp story about what happened. You can use this all the time. You can use it for sprint retrospectives,
and it’s good for post mortems to say, oh, what just happened the last two weeks and
so you can really get to the root cause analysis. What’s really interesting is when there’s
disconnects and half the team will put the stickies about the same event above the line
and half the team will put it below the line and so some people might look at an event
and say it’s a really good thing, some people might say it’s a really bad thing and that’s
a goods chance to say OK, team, let’s talk about that. Why did you think it’s a good thing, why did
you think it wasn’t? Let’s put our perspectives on the table as
it were. The fourth type of retro you can run with
your team is the one called the gratitude retro. Each person writes down two to three thank
you notes for other people on the team. You will read every sticky to each other. Why is why is this a good retro it feels like
the more you give the more you get something back, it feels really good to be nice to someone. It’s very easy to fixate on all the things
that are going wrong and it’s nice to say let’s just talk about all the things that
are great about our team and our teammates. And then that last bullet is really important. We take for granted that people know that
we recognize them, that we see, you know, what they’re doing, and we had one instance
where I had a developer and multiple people said, hey, thanks for how fast you do code
reviews and he was almost moved to tears, he’s like I take so much pride in trying to
n responsive as a teammate but no one has ever said thank you before. So it feels sort of awkward to say, hey, good
job. I came from the military, I’ve sports background
so I’m a high fiver, and especially I live in London now and the British are very resistance
to my rah-rah, so I sort of wander forlornly, with my hand up or my fist out and I think
someone is going to high five me or first bump me, right? But it never happens. And the gratitude retro is nice way to introduce
that practice into the team. It’s a really good one to do, just to break
things up. Use it as a change of pace retro though. You should not overuse this because the most
important thing is your mom is say thank you to the nice lady, it becomes mechanical, it’s
not going to have a lot of impact with your team. So it’s very important that it’s an organic
sort of sentiment. If your team is low energy, it’s a good time
to bring it in. I’ll give you a few caveats. Think very carefully if you’re going to use
this with a team that has a lot of open conflict. If you have a team that has open conflict,
they are probably not ready to give sincere thanks, right? You may get something like thanks for not
being a jerk face today. I know you’re all laughing but that’s a very
damaging sentiment, right so you don’t want to feel like you’re forcing the thanks. So I would use a different type of retro and
you’d have some coaching to do sort of as a person. The other thing you want to be a little sensitive
to, you don’t want to sort of the Valentine’s day syndrome where someone doesn’t get any
thank you stickies, so I always try to make sure, that I don’t know if anyone is seeing
the good work that person is seeing, I make sure I go out of my way to have a couple of
thank you stickies for them. But it’s a huge risk but a huge high reward. I really encourage you to give it a go. The final retro that I’m going to talk to
you about is what’s called the sailboat exercise. Forgive me, I draw this sailboat and it’s
terrible, I have my father’s drawing ability, which means none. The wind in your sails, what drives us, what’s
powering or engine as it were. What are the anchors holding us back. That’s supposed to be a hidden shoal or a
coral reef, that blob there. What are the hidden dangers that are going
to shipwreck us, and then most importantly and sometimes they leave this off in some
version of the sailboat, but I like to draw it on, what is the promised land? What does success look like to us? What are we trying to achieve. So you want a big piece of paper to draw this
out. If you want a less terrible picture than mine,
practice or look at pictures of sailboats until you can draw them with confidence. Like this one a lot. It really helps you center in hon the big
things that are impacting your team. When you say things like anchor that’s a pretty
serious word. And so you can kind of start to root out or
sniff out the things that have been plaguing your team over time. The shoals metaphor is actually quite powerful. Very interesting to me that that’s the most
vulnerable I’ve ever seen teams I manage get is putting stickies on that thing. So these aren’t things that are kind of bothering
me, these are fundamental threats. You can use it for sprint retrospectives,
and I have. I like to use it more strategically. It’s also very helpful when you want to reset
a team. I did this exercise, I had just inherited
a team who I didn’t know very well and we had just had a reorg, which is how they came
to me so I flew out and I did this with them. And the reason why I did this is
I want us to get on the same page with we’ve just had a reorg, what does success look like. They put their stickies up, and almost all
of them wrote the word attrition and this was a company that had a lot of turnovers
and one of the devs piped up, you know, we all wrote down that attrition is something
we’re scared of and we’re the only department in the company that hasn’t had attrition. Almost all of you wrote something in along
the lines of the wind in my sails is the amazing teammates I have to work with every day and
so that’s a very natural thing. If that’s what powers you every day that’s
that was a huge signal to me as a team lead that I needed to be very, very careful about
team dynamics because any sort of disruption to that really did run the risk of fundamentally
derailing that team and it was almost a mini gratitude retro for them. And oh, yeah, we do like each other, oh, yeah,
group hug. Again it was very good. Those sometimes you get in nit-picking things,
and sometimes you want to draw really big issues and the sailboat can really draw this
out. Retros only work if you do them and do it
regularly. It’s basic hygiene, it’s smoke detectors,
it’s brushing your teeth, whatever metaphor works for you there. You only want to fix one or two things per
sprint, like you can’t solve all of the world’s ills, you may do a process change but that
might not be the right process change for you, right? So it’s very important that you can continue
to do these so you can assess these process changes that you’re making as a group. Don’t wait until the house is on fire to do
them. Do them all the time. You want to make sure you’re following up
on your action items. Like I always tell my team, I have no sacred
cows, I have almost no sacred cows, if you tell me you don’t want to write any tests,
I’m going to fight you. Anything is on the table. All I ask is that you try it for a few weeks
and then we can change it. Like the real-world tools, there are many,
many more that you can do, there’s a link here. I’m going to post these slides tonight on
the bit.ly link. But I really want you to take away, think
very carefully about what you’re trying to accomplish before you go into that retro,
know what you want to get out of it, know what your team wants to get out of it and
pick the right team for the job you want to do. There’s a bunch I didn’t cover, remote teams. There are techniques you can do if you’re
remote. What if your team is super-large. Oh, I’m sorry, I’m out of time. But I do love talking about this stuff. My handle is at the bottom. So if you want to tweet at me and talk more,
that’s great. I hope you liked the talk, you can tweet at
me if you did, and if you didn’t like it, you can still tweet at me and I will try out
the new anti-abuse tools on Twitter. So go ahead and hit me. I’m ready. [applause]
And one final, I always like to put this in here, because it makes me embarrassed, it’s
a ringing endorsement on why you should do retros from my old teammate. That was my key architect and he said this
to me after the sailboat exercise because he gave me the hinky eye when it started and
he said, you know, it wasn’t terrible. Which was like, I’m going to carry that memory
to my grave, because it felt so good. Thank you so much. [applause]
MERI: Thank you very much, Jessie. So if you need to stand up and stretch, stand
up and stretch right now. Come on, you look like your fidgeting, don’t
be like toddlers. Best superhero pose! Right, everybody back down. We got one last talk before the next injection
of caffeine will arrive. We’re all going to pay awesome attention,
be a great audience.

Leave a Reply

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

Back To Top