The art of project planning

Part of my job is to help team members form their project plans. Engineers or even managers often find it hard to decide what should be done first, and what next in a project. They spend a long time working on something, but in sprint demos, they have nothing meaningful to show - don’t demo your code reviews please. Eventually when they get to the end of project, tons of unknown obstacles start to manifest, they sadly claim the project will be late. An awkward moment to face the shock of stakeholders! Many long lasting projects aren’t even that lucky. Because of the lack of progress perceived by the stakeholders, priority starts to change, stakeholders get impatient - the projects get canceled! So, it is an art to form a project plan that can: 1. Give fast feedbacks to team members that they are on the right track towards the goal. 2. Identify unknown unknowns as early as possible. 3. Deliver incremental values/confidence to stakeholders. The general guideline of good project planning is the same old, but gold Amazon leadership principle: Customer Obsession. Think from stakeholders’ point of view, and plan the project milestone and date to give them the value and visibility as early as possible, as frequent as possible. Suppose you are a contractor building or renovating a large house for your stakeholders. There are two ways to plan the project: 1. Plan the work horizontally, layer and layer: you first pour the foundation, ask stakeholders and their inspectors to validate the work is conforming to the original architectural design and local build code. Then you do the framing, again ask your stakeholders to check the quality and progress, then the roofing… 2. Plan the work vertically, room by room. Maybe the stakeholders want to use a room as early as possible. So your plan should prioritize the work to make the room livable as early as possible, allow stakeholders to start using it, then you move the work to the next room. In software projects, the horizontal way is similar to the CI/CD movement. You want to get your software changes into a continuous deployment pipeline to production environment from day 1, even you can only ship a “hello world” feature at the beginning. You get your infrastructure done first then incrementally add more and more content that is shipped to production continuously, maybe under feature flags, until the project is ready for customers. The vertical way is similar to the MVP (Minimum Viable Product) approach. You deliver the thinnest possible feature end to end to customers, get their feedbacks, then move on to work on the next vertical cut feature, one at a time. The essence is the same, no matter what project management methodologies you choose, you need: 1. Fast feedback to team members 2 Early and incremental values to stakeholders But don’t over think, after due diligence, get on the journey. Enjoy the ride!