-3

For the simplification, I am talking here only about the time, invested into the project starting from the task being defined for developers and functionality being passed to testers.

AFAIK, the mostly used tool for support of the project management is JIRA now. Using it, we can find how much time was used for different functionalities. Or, maybe, subtasks connected to functionalities.

But in the real life, a great deal of work of a developer belongs not to the coding, but to the different additional activities: installing and configuring of platforms, libraries, old and new parts of the project, refactoring, functional and unit testing and others. If some activity is a great and well-known time-eater, everybody knows about the problem and we can hope that the problem will be solved sooner or later. But time-eaters can expose themselves only in connection to some specific activities or parts of the product. And different developers are wasting their time again and again, not even knowing that it is a common problem.

The obvious solution is to register time put not only into different functional tasks, but into different organizational stages, necessary for their solutions. Such as reinstallation of plugin A or writing API tests.

If we had such information, we could find the problems and after solution of them greatly increase the effectiveness of the common work. Something as:

When creating a functional test, 
  we spend 40%+-15% of time for 
  the automatization of server launching with the necessary data. 

=> 

We should make some data loader for that server.

The tools as MS Project Manager can show us the structure of the time invested, but cannot measure it. They can be used only for postfactum analysis, if somebody has collected and put in the data.

But HOW can we register and/or measure the time put into different activities connected to the solutions of the concrete tasks? Also, I would like not to pass the measurements into the analysis process manually. JIRA, IMHO, is absolutely unusable for any of these tasks.

I am not so interested in the tool (any table calculator can serve) as in organization of the whole process among tools and people.

  • just use KANBAN and ensure all your stages are present. You'll soon see which takes the longest! – Ewan Feb 5 '18 at 21:14
  • @Ewan Could you put here a reference how the KANBAN solves the problem? I had read about it and I do not remember any its features that could help. – Gangnus Feb 5 '18 at 21:22
  • its the nom de plume of KANBAN to identify choke points. if you put all the stages you want to measure, you'll see them pile up at the slow stage – Ewan Feb 5 '18 at 21:27
  • @Ewan I am terribly sorry, but I haven't understood your comment, after several readings. Could you say more, or differently, or give a reference, please? I am a subscriber of The Agile Manifesto from the year of 2006, and I do not remember a necessary feature in any agile method. – Gangnus Feb 5 '18 at 21:37
  • 1
    well you can always increase the resolution as you go. But you do know the steps, because you do them all the time. You just have to write them down – Ewan Feb 5 '18 at 21:53
1

I used a very simple approach for your problem: An editor (emacs in my case) which I configured such that on a specific key press it added a new timestamp line in a log file. Then, to that time stamp I typed the name of the activity. The result looked like follows (the activity names are certainly in practice abbreviations):

2018-02-05,22:45:17 Read question
2018-02-05,22:47:03 Think about answer
2018-02-05,22:55:12 Formulate answer
2018-02-05,23:05:12 END OF DAY`

This could even be done in a way that allowed me to account the same time to different activities (think you "book" the same time to different accounts):

2018-02-05,22:45:17 Read question; Gangnus' question
2018-02-05,22:47:03 Think about answer; Gangnus' question
2018-02-05,22:55:12 Formulate answer; Gangnus' question
2018-02-05,23:05:12 END OF DAY`

This way I could, for example, figure out how much time I spent on a specific task, but also how much time was spent for telephone calls etc.

The overhead is fairly low. But, in practice I have used it only sparingly, namely when I really was in need of some reliable data for discussions with my boss. For example, once I used it over a period of four weeks to give some data about the true hours spent on a refactoring activity. Once I used it for a prognosis on the expected duration of a thorough document review.

I am sceptical, however, that it can be rolled out to a team. Ensuring everyone is doing it the same way would be fairly costly (training, continuously reviewing the data). The set of "accounts", for example, was never fixed in my use-cases, but evolved depending on what information I needed. For a team, you would have to define common accounts and ensure everyone "books" to them in the same way.

  • I used something as your method on Excel. And I agree with your thought - the main problem is how to integrate this information for the whole team. – Gangnus Feb 6 '18 at 10:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.