Just want to say that we are talking about small projects - 2-5 people, lasting about a year or less.
Having encountered several times with the need to become the manager of such a small project, I decided:
a) share your thoughts and
b) ask for advice (after all, I'm probably not the first to encounter the one described below).
So, I first encountered this. One of the main (for me) tasks was to assess the status of the project. It was necessary to understand whether everything was going according to plan, and if not, then timely detect the deviation from the schedule and anticipate a new completion date.
')
To my great surprise, I could not find a simple solution to this problem. The company used Microsoft Project, which, in my opinion, solves this problem very badly.
In short, why I found Microsoft Project and similar tools inappropriate: their key component is the Gantt chart. Which, perhaps, works for “material” projects such as building a bridge or a skyscraper, but for a small IT project, it seemed to me poorly applicable.
- For most tasks, I do not need to build dependencies like "Start-to-finish".
- Moreover, for most tasks, the sequence of their execution can be any and me, as a PM, she is very interested.
- If I assign two tasks with a duration of 10 days to a single person in a project without additional gestures, then I want to immediately see that the project will be completed in 20 days. In Microsoft Project, I will see that the person will be loaded for the first 10 days by 200 percent. Unless, of course, the “handles” move the task.
- I want to be able to transfer tasks between people “on the fly” and change their planned duration. Immediately seeing how to change the estimated completion date of the project.
- And actually, I want to see what I called time tracking.
One more picture in the general heap - for some reason, quite a few people used and used for this task ... a banal sheet in Excel.
Now let's get down to how I see this process, and what I would like to see from the project management system.
General:
1. A tree of tasks.
2. One task - one responsible.
3. “Leaves” tasks can (and should) be given an estimate of the execution time.
4. Task “nodes” automatically summarize the execution time of all subtasks.
5. Tasks are created and distributed among the participants of the PM.
Reporting on the work (the main thing - it should be very simple - and then either will not do, or will do formally!):
1. Each team member contributes daily how many hours were spent on a particular task.
2. At the same time, he sees the “remaining time” for this task and can change it if necessary.
3. When the task is completed, its remaining time becomes zero. (Ideally, if the estimate of the required time was made extremely accurately, it will happen by itself, without changing the remaining time manually).
Reports for PM:
1. Estimated deadline date (initially only on the basis of ratings, then taking into account the amendments to the remaining time made by the participants). Based on the fact that each participant does 8 hours of work every day (ideally, adjusted for efficiency, communication, etc.) Actually, everything is being done for him.
2. The percentage of work done at the moment. Also quite an important indicator.
3. Changing the deadline date in time (reflects, for example, planning accuracy, big changes, etc.).
4. Various sections on tasks (time spent, change of assessment, etc.).
5. Deviations from the schedule.
6. Problem areas (a big difference between the original plan and the actual time spent, abrupt changes in the remaining time, etc.).
7. Custom reports as desired by the PM.
Desirable goodies:
1. Integration with existing bug tracking. So that when creating an error (or a change request) it can be tied to one of the tasks. So that the time in reporting can be tied to an error, but at the same time it automatically added to the time spent on the task.
Note that the start time and the end time of the task do not appear anywhere! All calculations are conducted from the current day (if we look to the future) or according to actual data (if we look back).
All tasks are considered independent. Naturally, in reality this is not always the case, and the assignment of tasks for PM people should try to be done, taking into account possible dependencies. But in the system of some specials. There is no mechanism for this.
Moreover, the end time of one task is not so important here. It is important to evaluate when the entire project will be completed. And this estimate is automatically recalculated after each change (adding a new task or changing the remaining time, creating a change request or an error report, etc.)
What is important is how much time will be spent (or already spent) on a particular task. And it is on the basis of these data that all estimates and calculations are made.
Undoubtedly, the system described here is far from ideal and simplified in many respects, but ... it seems to me that just something like this could greatly simplify the life of a PM from a small project.
Now the actual question is - is there a realization of this view on project management? And does it have a right to life?