Managing a complex engineering plan from idea to delivery
Image by PIRO from Pixabay

Managing a complex engineering plan from idea to delivery

Organizations nowadays are complex and really autonomous teams are hard to find. Even the Agile Principles founders have come to the same conclusion and came up with the Agile at Scale frameworks such as SAFe and DaD and LeSS. In these frameworks, the desired organizational structure is based on “Trains”, each train to be a set of teams that are cross dependent, where the workplan is managed by Scrum of Scrums ceremony. This means that epics are usually spread across teams, and have cross teams dependency which we cannot avoid anymore, though always should be minimized and simplified as possible.

So how can we set a unified view of our engineering work? How can we create visibility and predictably as much as possible to the Management and teams? How can we measure from when an idea is born till it's delivered to customers?

I'll represent the framework we work with. We use the Jira Advanced Roadmaps, provided by Atlassian but I know there are multiple other addons providing the same ability.

Laying the Jira infrastructure for a unified view:

We’ve created one jira project called Engineering Workplan where we add our quarterly goals, initiatives and epics. So all our epics are numbered ENG-XXX.

For each epic, all teams involved add tasks and stories according to their part of the teams (we have one Jira project per team so it's easy to see which task belong to each team, but we also use the new Team field to differ). This means that each epic has children tasks on multiple teams (RPS-123, AMT-456 and so on) connected to the same epic. See picture:

No alt text provided for this image
ENG Epic with children in differnt projects

Also in order to unify our language, we’ve moved to unified sprints, 2 weeks each, named in a format that’s easy to understand the timeline from (Nov-1, Nov-2, rather than Sprint 156).

Providing visibility to overall product plan

We’ve set a workflow of an epic to start when an idea comes up ("New" Status) until it’s Done: We use "Requirement" Status for when the epic’s requirement are being documented and "Definition" Status to the stage where the full PRD and Design are written with the help of the architecture team. After that the epic moves to "Tech Design and test plan" Status where the focus transfers to the architecture, the main team leads involved and the QA leads. Only after that the epic is moved to "Ready for Dev" status. This provides very clear statistics on how many epics we have on each stage, and for each delivery stakeholder (Eng Team Lead, Product, Designer, Architect..) to know what’s on their plate and what comes next.

No alt text provided for this image


Providing visibility to our Development plan

As an epic reaches Ready for Dev status, the teams break the epics into tasks and stories in their boards and assign the tasks on all the sprints till the end of the quarter to provide visibility on the planned work of the epic. We also open integration and E2E QA tasks to make sure the plan includes all the relevant phases. This creates visibility into the epic development plan and if something gets delayed or changed, the impact is visible and known.

Providing visibility to our teams Sprints

We use for time estimations 1 story point equals 1 work day (don’t judge) and we provide rough estimates for our tasks in order to provide an estimated development effort and timeline. Then, we assign the tasks to sprints, and use extensively the Group by Sprint and Group by Team to see what’s each team’s plan till the end of the quarter and where are we exceeding our capacity. This way we can understand what’s the focus for each sprint and with the right buffer, provide predictability into when we’ll be able to deliver a feature.

No alt text provided for this image
Bottleneck and over-allocated teams visible and easy to analisys

To summarize, the ability to provide a high level plan that allows in a click to go further down the details and then back up to the big picture is super crucial especially in big companies, where details might overflow but are needed for precise delivery. Using the latest Gantt view provided by Advanced Roadmap and aligning on a single method of Jira usage as shown above provides us the ability to manage our ship into delivery.

great work! thanks for sharing

Like
Reply

To view or add a comment, sign in

Others also viewed

Explore content categories