PEST5 Workshop – Release planning

This is one of the assignments used in PEST5, the post contains all the info for the all the roles. If you plan to use it, make sure you don’t spoil yourself.

Roles:

Product Owner – Business and Market Quality representative
Project Manager – Development side manager, knows the team and development problems
Test Lead – Testing side, knows the team and testing problems

Assignment:

Analyze your info, talk to instructor regarding details. Think through what you want to achieve and how.

Your assignment is to agree on a plan – so that development are reasonably sure they can finish, testers can be happy about the release (have enough time with it) and business gets the stuff when they need it.

General details:

Product name:   Gazebo
Current software on market is OK but there are some bad issues.
Current plan was to have new feature launch in w14 and then second release in w24 (both with some market fixes as well)

w14 Scope, (already communicated to the market that these will be coming):
Whistle (not communicated, planned to do as market event on w20)
RT44
Hippies
Goodies
Market error fixes

w22 Scope (release communicated, scope detailed are not):
Trailers
Market error fixes
Sunshine

Situation:
Today is w14 and thus that w14 release is not possible, due to quality of Goodies. You need a new plan.

Role specific details, others should not see these:

Product Owner:

  • A new release is plan is needed ASAP, to solve the critical Market Error fixes which had been promised already for w12 (and were postponed for 14). People are returning products due to these issues.
  • Goodies is a very requested feature – the sooner we get it out the better. We are getting bad reviews without it.
  • Whistle has been in development for a while now and must surely be ready for market before w20 – a big marketing event is planned (and communicated).
  • And Trailers and Sunshine must be shipped before end of June – as July is vacation time and we can’t postpone after that.

Project manager, development:

All tasks estimation given with full team effort – so it takes 3 calendar weeks to do both Trailers and Sunshine, for example. No parallel work. Overtime is not counted in here. Development time does include testing while development – thus the time testers ask for is after “feature complete”, to investigate integration issues and fixes to those, etc.

  • RT44 – done, very difficult to merge into previous release (1 week to do, high regression risk)
  • Hippies – done, can be merged into previous release (2 days, low regression risk)
  • Whistle – not done, difficult to handle (technical+business changes all the time), getting it ready for w14 was anyway a problem which was planned to be solved with a hotfix at w18. Correct development would take 4 weeks.
  • Goodies – done, but few important but seemingly isolated issues to fix. Doable with 2 weeks.
  • Trailers – doable, 2 weeks, low regression,
  • Sunshine – doable, 1 weeks, low regression,
  • General Market issues – not doable as much as needed, regression risk high, not enough info from market about the issues. Only lot of complaints. Must be prioritized.
  • Critical Market issues – can be merged into previous release, medium regression risk, 1 week

 

Test Lead

All tasks estimation given with full team effort – so it takes 2 calendar weeks to do both Trailers and Sunshine, for example.  Although tested during development the stated testing time used after developers are “feature complete”, to have time to test the full solution. Bigger the regression risk, the bigger the uncertainty of testers being happy about the release.

  • Original w14 release was planned for 5 weeks
  • Original w22 release was planned for 4 weeks.
  • RT44 – easy to test, low regression risk, 0.5 weeks (already tested, only quick look on final release version needed (0.5 days))
  • Hippies – easy to test, medium regression risk, 0.5 weeks (already tested, only quick look on final release version needed (0.5 days))
  • Whistle – time-consuming (2 weeks) to test for deeply enough from test viewpoint (without any changes during that time), high regression risk with any change
  • Goodies – Regression risk high, time-consuming to test. Not much used by customers, but still important feature demo wise. Developers are almost always not able to keep their estimations with this component. 2 weeks
  • Trailers – low regression, medium to test, 1 week
  • Sunshine – low regression, medium to test, 1 week
  • Market issues – regression risk high, not enough info from market about the issues. Only lot of complaints. At least a week, depending on how many developers actually solve.
  • Critical Market issues – medium regression risk, 1 week of testing.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s