RecognizingWhatsWorking

How do you know when things are working and when things need to change? -- JohannaRothman 2005.06.21

I know when things aren't working for me. It's when I have a feeling of doom that doesn't go away. Sometimes that feeling of doom is, in fact, doom.

It is much more difficult for me to figure out when things are working for me . I have been in positions where everything feels right, people seem to be working well together, the customer appears happy, but the end result doesn't work.

CharlesAdams 2005.06.21

This is probably a classic ISTJ answer, but here goes. If someone will define what a task is and the desired outcome of the task, I can judge if things are working if I meet these defined parameters. Now that all the N's and F's have rolled their eyes around or passed out on the floor...

Really, the problem I have found is that I often have felt good about the situation and how things were going. Then I spoke to someone up a couple of layers in the organization and learned that they felt it was a disaster. They had a different perspective.

I now try to talk to as many people as I can who have different perspectives to learn what they think is working and not working. Sometimes this helps, but sometimes it only confuses me more. In the end, however, the more perspectives I have the better chance I have of doing something that "works"

DwaynePhillips 22 June 2005

How do I know when things aren't working
When I stop listening
When I stop thinking
When I stop changing

SteveSmith 2005.06.22

When I hear a lot of blaming going on around me, I know that things aren't working.
In fact, when I hear a lot of any kind of incongruence around me, I know that things aren't working.
Knowing when things are working is harder, as others have said. Partly that's because we don't know what "working" is--but we do know what not working is, in many cases - JerryWeinberg 2005.05.26

I really like the qualitative information people are describing. Does anyone have quantitative information they specifically look for? -- JohannaRothman 2005.06.28

Most of my "things need to change" markers are subjective, though they might be qualitative feelings about quantitative data (e.g., "gee, the bug count is going up..."). There are some places I/we have set up automated tripwires (e.g., for free disk space on key servers). In other cases, like bug counts, quantitative tripwires are problematic, since it's the qualitative aspects of the bug that might indicate a need for change (as opposed to a need to shift focus). --DaveSmith 2005.06.28

As for quantitative measures, "things aren't working" is a pretty broad place to start. Some simple measures that I find useful are:

- Fault Feedback Ratio (what percentage of the time when you try to fix things do you make additional mistakes).

- Questions answered/questions asked (this is a measure of meetings that aren't working. If it falls below 1/3, your meeting definitely isn't working).

- Number of bullet points per slide in presentations (if this is above three, it's not working).

So, that's three bullet points. Any more and it wouldn't be working. - JerryWeinberg 2005.06.28

The basic quantitative measure I find myself using over and over again is amount of work we've accomplished/amount of work we planned on accomplishing. This ratio is the basis for earned value and several other systems. It may not be a good measure of "what's working," but it's a good measure of "are we doing as well as we planned?"

Sometimes the work is harder than we expected it to be. Seometimes we don't have the people on hand that we were supposed to. There are many other explanations for having a poor score on this metric.

DwaynePhillips 29 June 2005

I haven't yet found a place that formally tracks FaultFeedbackRatio -- or even one that's heard of that terminology!

Ahem. Many of my clients do now :-0 -- Johanna 2005.06.30

You know something is going wrong when the estimated duration of a project or sub-project keeps getter longer... For example: First week, the estimate is 5 weeks. In the second week, the estimate is 10 weeks. In the third week, the estimate is 20 weeks.

KeithRay 2005.06.29

Who are the stakeholders in whatever you are doing? Are they getting value? One way to try to get at that is to ask them.

-- JimBullock 2005.06.29 ("Who do you serve?", "Saruman." - At least he - Lurtz of the Uruk-hai - was clear on who mattered in his world.)

How do you know when things are working...?

You feel it. People on the project are happy and cheerful about their work (rather than happy and cheeful in spite of their work ).

In software - we're confident that we have the right features (more or less), we're implementing and testing those features without too much effort. Our work isn't exceeding estimates (too often).

KeithRay 2005.06.30

With respect to Keith's observation of estimates changing, even more interesting is the timing of the announcment of the changes. If you're changing estimates every week, that's actually healthier than waiting until the day a six-month project is due, then announcing that it's going to slip three months. Try slip-date charts. - JerryWeinberg 2005.07.01

Updated: Friday, July 1, 2005