Using KPIs/Getting a data-driven org.

From CitconWiki
Revision as of 14:06, 7 November 2010 by 93.96.137.193 (talk) (New page: 2.5 years of dev got a release out every fortnight 27 financial projects 5 made money, 2 made most of it this didn't change anyone's mind Maybe was too aggressive Maybe it was embarrassing...)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

2.5 years of dev got a release out every fortnight 27 financial projects 5 made money, 2 made most of it this didn't change anyone's mind Maybe was too aggressive Maybe it was embarrassing

Don't see a correlation between the quality of the project and the value it brings Building the wrong thing

Accountability is missing? Arguments for value wanted from one side, but not provided by the other side

Accountability doesn't change things PM a project that doesn't have market If he said there was no market then he has no job

Maybe it has to happen early. During the initial phases decide what needs to happen

What to do "Agile" to obfuscate accountability "Agile" separates accountability into team and product owner Team is only accountable for implementing what they are told Product owner

A/B testing: 1/3 of the time the change makes not diff. 1/3 they make a negative diff. 1/3 of the time it works

What are you accountable for In an organization you cannot make mistakes Everything is success! Do things that are safe! Or I can blame you!

Accountability: things can be traced back to the root cause. Making the problem yours.

Role play: product owner

Why not have an information radiator about how you are doing (products selling, etc.)

Management doesn't want to take responsibility for stopping a project or continuing it. Having some measurements even if they are gamed is probably better than nothing. Maybe gut feeling is better than nothing. Design companies have more of that. Microsoft version of design: user groups and testing vs. Apple: gut feeling and "good" designers

Do we make decisions rationaly. Is there a rationale for ever decision. Decision models: "garbage can" model:

   Problems
   Participants
   Solutions
   Vs.
   Problem connected to causes

The moment someone feels responsible is where you start improving. The data is a placeholder for conversation. Maybe an uncomfortable conversation, but it is a starting point.

Defensive reaction to being told about responsibility.

Responsibility Virus: google for it, you'll find a pdf

Culture where people can talk about this. Talk with teammates. Don't worry about getting shot down. Collect data that one thinks should be measured. Maybe just voice concerns. Suggest improvements.

Being silent and not raising an issue is maybe complicit. Some people stay under-responsible because they don't know how to communicate well. There is no external encouragement to share reservations. Responsibility of the management to elicit these things?

Look for ways to minimize defensiveness by looking for problems and communicating

Great question for retrospectives: What is un-discussable right now?