RETROSPECTIVES A Backward Glance at the Future

1
RETROSPECTIVESA Backward Glance at the Future

• Tracy Balkovic
• Hewlett-Packard Company
• tracy.balkovic_at_hp.com

2

• I wanna go backAnd do it all over againBut I
  can't go back I know
  
• - Eddie Money

3
Retrospective - A Definition

• RETRO INTROSPECTIVE
• R E T R O S P E C T I V E
• Looking back on and contemplating the past.

4
WHY RETROSPECTIVES?
5
WHY RETROSPECTIVES?

• Humans are creatures of habit. If we dont do
  something to give us a little shake (or in some
  cases, a kick in the pants!), we will continue to
  do things the way we always have.

6
You go back, Jack, do it again,
Wheels turnin' 'round and 'round
You go back, Jack, do it again
- Steely Dan
7
WHY RETROSPECTIVES?

• If everyone did everything well all the time,
  then this approach would be perfect! ?
  
• BUT, ALAS!...........

8
WHY RETROSPECTIVES?

• Whens the last time .
• you worked on a project
• where everyone did everything well
• AND
• everything went right?

9

• You grieve you learnYou choke you learnYou
  laugh you learnYou choose you learnYou pray you
  learnYou ask you learnYou live you learn You
  do retrospectives you learn
• almost Alanis Morrissette

10
WHY RETROSPECTIVES?

• The retrospective is the little shake you need to
  make changes in how you do things
  
• Its looking back on your project through a
  window of time its almost like looking in from
  the outside

11

• You make it hurt so good
• - John Cougar Mellencamp

12
Its All About the Bad Stuff Right?
13
WHY RETROSPECTIVES?

• Not SO!!
• Retrospectives are about
• What went WELL
• What went WRONG
• What we want to do in the FUTURE

14
WHY RETROSPECTIVES?

• By looking back, or in from the outside, we can
  learn a lot about our project
  
• Areas where things went well
• Areas where we had trouble
• Areas where we can improve
• Areas where we have no control
• Areas where we should leave well enough alone

15

• You gotta learn to listen,
• listen to learnYou gotta learn to listen,
• before you get burned
• - the Ramones

16
If theyre so great, why doesnt everybody do
them?
17
The Good Old Days
ACTIVITY
TIME
18
The 2 Fast 2 Furious Days
Project A
Project B
Project C
ACTIVITY
TIME
19

• Oh, no, gotta learn this some dayWhy, I, gotta
  learn the hard way
  
• - Nickelback

20

• Those who do not learn from the past are
  condemned to repeat it.
  
• - George Santayana
• You can learn it now,
• or you can learn it later.

21

• So, what can we do to gather the information,
  analyze it, learn from it, and still keep pace
  with our projects?

22
A Helpful Process

• We need a method that will
• Gather LOTS of information
• Allow us to analyze what went well and what
  didnt
  
• Help us determine what to do in the future
• Take less than a day

23
Tracys Tried and True Template

• The process has been used for
• More than a dozen retrospectives
• A variety of scopes
• A variety of groups
• A variety of projects

24
Step 1 - The Invitation

• You need to invite the right people.
• Decide the scope/focus of your retrospective
• Entire project top to bottom
• One functional area (like software development,
  test, manufacturing, etc.)
  
• A grouping of like areas (like firmware and
  software)
  
• Invite everyone who was involved with the scope
  you picked.
  
• Give them enough time to contemplate prior to the
  meeting
  
• Give them something to help them think and
  prepare
  
• Give folks who cant attend the opportunity to
  provide input

25
Sample Invite

• Folks,
• It's time for the XYZ Retrospective. The
  retrospective will be for FW AND SW and will be
  on June 28, 2004 from 200 - 500 in Room 1.
  
• I've attached a spreadsheet to help you with
  collecting your thoughts. The overall process is
  also included. If you are unable to attend,
  please use the Planning sheet to capture your
  thoughts and send it to me prior to the
  retrospective. I will include your info in the
  data gathering process.
• If you have any questions, concerns, please let
  me know.
  

26
Step 1 - The Invitation

• What Ive found
• People feel comfortable writing things down
  less intimidating than speaking up in a crowd
  
• More participation no one person can hog the
  conversation
  
• It doesnt turn into a bitch session they
  have to focus on the positive as well as the
  negative

27
Step 1 - The Invitation

• What Ive found (contd)
• Sending out the prep worksheet lets folks who
  cant attend participate I can add their info
  to the PROs/Cons
  
• I often will not invite managers to
  retrospectives that focus on more tactical things
  people feel less intimidated when their manager
  isnt in the room
• We get LOTS of data!

28
How Does it Work?

• Lets try an experiment!

29
STEP 2 - Gathering

• Your goal is to gather as much information as
  possible
  
• Use headings that work for your projects
• Keep people moving if they sit too soon,
  encourage them to go around one more time

30
STEP 2 - Gathering

• What Ive found
• People feel comfortable writing things down
  less intimidating than speaking up in a crowd
  
• More participation no one person can hog the
  conversation
  
• It doesnt turn into a bitch session they
  have to focus on the positive as well as the
  negative
  
• Sending out the prep worksheet lets folks who
  cant attend participate I can add their info
  to the PROs/Cons
  
• We get LOTS of data!

31
How Important Is It?

• The experiment continues!

32
STEP 3 - Prioritize

• Once you have gathered the information, you
  need to prioritize it
  
• Give everyone a number of votes
• 3-5 cons, 3-5 pros
• Square root of cons, Square root of pros
• Gut feel
• Tally the votes and move to the next step

33
STEP 3 - Prioritize

• What Ive found
• Sometimes the number of people are skewed (e.g.,
  too many test folks and not enough developers or
  vice versa). In this case, I may give the FOCUS
  group the opportunity to modify the priorities
  a little in the next step
• Too many votes takes a LONG time
• In the past I have let them put as many votes on
  one thing as they want as long as they dont go
  over the total allowed for the category. This
  often causes some strange priorities, so you
  might want to avoid this.

34
STEP 3 - Prioritize

• What Ive found (contd)
• If you have more time, you can consolidate things
  before you vote. This is time consuming, so I
  usually dont worry about it and consolidate it
  after the votes.
• You can also take the top 2 or 3 items from each
  PRO/CON list and put them on separate sheets for
  voting purposes. If you do this, it might save
  you a little time, but you also might not have
  the HOT item on the list for votes (it does
  happen sometimes).

35
Were prioritized, now what?

• Back to the experiment!

36
STEP 4 - Analyze

• Start with the highest and do the top 3-5
  issues.
  
• Once youve completed the list, you can ask if
  anyone has a burning desire to add anything else
  from Pros/Cons sheets. This gives a sense of
  fairness especially when the group mix is
  skewed away from the FOCUS of the retropsective

37
Step 4 - Analyze

• For PROs
• List the benefits to the project/team/etc.
• List the effect it had on the project
• List what might happen if the PRO didnt occur

38
Step 4 - Analyze

• For CONs
• Make sure to understand the issue. This may take
  some discussion just because everyone voted for
  it doesnt mean they read it the same way
  
• Discuss and list the consequences/issues that
  occurred because of it
  
• Discuss and list what the group can do sense of
  ownership
  
• Discuss and list what management can do

39
Step 4 - Analyze

• What Ive found
• PROs feel harder than CONs
• For CONs, focusing on what can we do gives the
  team a sense of ownership and they feel less like
  victims of bad decisions. If they say, we cant
  do anything, ask them what they could do to
  prevent from being in a bad situation again if
  the same thing occurred

40
What do we do with the data?
41
Step 5 Presenting the Data

• After youve documented everything, send it out
  for review to the attendees. I usually include
  
• Overall process
• ALL PROs/CONs by category
• ALL PROs/CONs by votes
• What Went Well
• Issues Addressed

42
Step 5 Presenting the Data

• Set a time to meet with the group thats
  interested in the feedback
  
• Current project management
• Project teams for similar future projects
• Invite the retrospective attendees as optional
• Go through the process with them dont just
  jump into the issues!
  
• Cover the PROs as well as you cover the CONs
  theyre both important