8

I work on an very small software development team of 3 (remote 'full' timer, a full-time developer on-site, and myself as 'lead' on-site) in a "small" aerospace manufacturing plant of about 300 employees & 150 machines.

Our Situation:

We have been given an incredibly large chunk of projects to research, design, and develop:

  1. a live WIP-Quality-Inspection-Hub for our machine operators;
  2. designing and creating a few automated inspection machines;
  3. a reporting view within the hub for our inspectors and quality supervisors;
  4. an auto-blueprint-reader/marker;
  5. a program that digitizes our contract review process
  6. a "request module" for our machine operators and eventually for any department that has a request of another department;
  7. a 3D "shop floor metrics" display for our high end customers...
  8. and about a dozen other smaller projects

I was tempted to say that one of the bad things is that our team is small and inexperienced, but we work well together and are all very quick learners and have already tackled some projects that engineers before us couldn't (the automated machine is 'half-automated' currently and has been a huge boon to some of our 'value-stream' parts, a project engineer before us was never able to get that off the ground).

One of the 'difficulties' (not necessarily bad or good in my eyes) is our shop is in high demand. So our manager, who is also the engineering manager, is super busy and sometimes won't even check in on us for 1 or 2 days (on rare occasions it can feel like 3 or 4), and usually when he checks in it is just to see what we are working on, usually not to change our direction and never micromanages.

However, because our work is starting to bear fruit, and so many departments & processes need to all be brought into the 21st century in parallel to avoid bottle-necks, our priorities will sometimes change suddenly (usually after a big development 'release'-cycle).

The nice thing is that because our company is large and profitable enough (and our team is small enough), and has been around for 50 years... we sort of don't have deadlines, as long as we continue to show improvements because the company has been quite successful before us anyways, we are in good standing.

But of course my manager has multiple people above him who want more accurate reporting on our tasks completed, and projects in progress.

Summary of needs:

  1. Priority viewing/changing across all projects in one view, so that our manager can say "this is everything they have on their list to eventually tackle", and also change priorities of tasks & projects on the fly (i.e. They have completed a few milestones on this project and we can start testing it out on the shop floor... in the meantime they should switch to 'Project X' for now)
  2. Standard project management things such as tasks, sub-tasks, tags, assignees, status (open, in progress, closed), deadlines, etc.
  3. Time recording
  4. Reporting of tasks complete and time spent on a task.
  5. The ideal package would allow us to design whatever hierarchy we want, but as long as the hierarchy can go from "all" to "granular" easily... well that would be sweet.
  6. Time estimates, and 'milestones/sprints' are not our focus currently.

What we have tried so far, or I have done short testing on:

  • We have been running off priorities given directly from our manager in person. It has worked well for about 12 months but we need to upgrade for reasons listed above.

  • Running a shared file on Excel; but Excel often feels too clunky for this, and shared excel files seem to be ticking time-bombs.

  • EDIT: - A handful of project management software including many pure Kanban sites/software such as Trello and Tiaga.io (which always seem to lack hierarchical-prioritizing views and often lack time recording). We have also tried software built upon more unconventional or 'less popular' ideologies such as priority matrices. It seems that our best software solution (so far) focuses on a hierarchy of containers and lists that can be prioritized on a granular scale or at the macro level, and levels in between those two extremes (ClickUp).

  • We recently hired a project manager to work under the engineering manager and help us balance our priorities. But she is brand new this week, and has a lot to learn before I think she will start being able to really help us on this front.

  • We have also discussed designing our own priority manager within our "request module" (project item 6.) but of course this will take time and we need this functionality to help us even get the request module done. I imagine we would fashion it to work more or less exactly how ClickUp works, but with some minor design changes.

  • My manager has also discussed the idea of outsourcing any projects we want expedited... but unless he is getting impatient I would like for us to eventually tackle everything ourselves. We are kind of building a reputation as a small powerhouse team and I would like that reputation to grow.

Where we stand now: Ultimately my manager has given me the power over this decision, and whatever I feel works best he is willing to work with. My current plan is to suggest we use ClickUp until we can create our own internal version... but I am eager to find a better solution.

  • You may want to focus less on the software in your Question and more on the problem - Questions asking for software recommendations get flagged to close. – Sarov Jul 12 '18 at 18:01
  • Hmm... I get your point. But I am not specifically asking for software, I am just asking for a "better solution". The list of software is my attempt to solve the problem, if you notice I also mentioned that we brought on a project engineer and we may just also make what we need in house. This would definitely get removed from stackoverflow... and if I didn't list the software I tried I wouldn't be providing the solutions I tried. – Paniom Jul 12 '18 at 18:11
  • Fair enough. Hence my use of 'may want to', rather than 'should'. – Sarov Jul 12 '18 at 18:35
  • I cut down the segment that focuses on software to just what types of ideologies we have tried with software. Hopefully this helps to avoid leading this to seem purely like a software request. – Paniom Jul 12 '18 at 18:53
5

So, it looks like, you have two problems:

  1. Clear priorities.
  2. Delivering various projects in parallel.

First one can be addressed by creating a clear list of all features (not tasks!) from different projects you work on -- so called Backlog. And this Backlog is common for all projects. And you set policy that you work on this list from top to bottom no matter what is in it. And it's not your problem what is in it - you don't care. If any stakeholder wants you to work on his project, he should put features from it to the list, shifting other ones.

This list must be shared across company, and have names of those who "orders" every feature. So, if another stakeholder finds that his features are too low in this list, he should communicate it to others. In such case it becomes not yours, but problem of the stakeholders as group, to set right priorities, and they can set any process they like for this: weekly meeting where they meet and groom it, I think, one of the best choices.

This can actually help with second problem too, because it will make them split their requests to smallest features, so that their projects would move "in parallel". Ofcourse, you will need to setup Continuous Integration system for it and write unit tests to be able to release small chunks quickly. Also, in our company we do our best to help our customers to determine MVP for every feature/epic, so that they get 80% value for 20% of our efforts. It is always difficult, but pays back, when your customer see that you are his partner, not just pair of hands.

From methodology and software point of view, Kanban should be perfect: it allows you to have clear queue of work, also limiting Work In Progress visually, to let everyone know, that you can not do everything simultaneously. LeanKit or Kaiten are very good for it. JIRA is also ok.

  • +1. I’ve been in OP’s shoes and doing what you’ve recommended here worked well for us. – RubberDuck Jul 14 '18 at 23:13
0

It’s already answered, but you need to just use normal RoI or cost of delay analysis to identify the most valuable project. Throw the rest away for now. Split that into the most valuable 20%, and throw the other 80% away. Split that 20% into chunks that can be done in a week. Take the most valuable chunk and split it into 5 „daily chunks“. Develop and put into production the first daily chunk. Then thexsecovd, then the third, then the fourth and on Friday put the fifth most valuable chunk in production. On Monday reasses. Should you start the next chunk in that project? Or start another one?

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.