Make your project a success ! Q&A

I participated in an in-depth Q&A session with Ivan Dalmet and Yoann Fleury, focused on project success strategies. I shared insights from my experience managing complex projects. This session covered crucial aspects of project management, from technical decision-making to team dynamics and strategic planning. Here are the key insights and discussions from this interactive session.

Project restart vs. code maintenance

Many projects look functional but miss critical parts. Some are only proofs of concept (POC), rarely fit for production. From my experience, half-finished or abandoned projects usually cost more to fix than to rebuild from scratch.

Sometimes you do not have the choice though. And because maintaining inherited code means first studying logic, then adding features, my advice is: write documentation early. Missing documentation, poor design, and stacked quick fixes only make the situation worse.

But not every project requires a restart. A POC can evolve into a stable product, depending on the state of the first version. However, if the foundation is unstable, rebuilding is wiser; if only the walls are weak, maintenance is still possible.

Now when taking back a project from another team or company, here’s what I’ve learned: the real risk isn’t the code, but the loss of business knowledge. Staying connected with original developers preserves context that can’t be rebuilt. Clear communication, documented decisions, and regular knowledge sharing are what truly make teams effective in project management.

Project management strategies

Speaking about project management: from my experience, the best providers are curious and honest. They explain their approach, ask questions, admit limits, and suggest alternatives. If someone comes with overconfidence and no clear plan, that’s a red flag. Profiles and past work can help, but they are never absolute proof.

Concerning the client expectations side, honesty is key. If a project doesn’t match our skills, I redirect the client to someone better suited. I also refuse clients who create more problems than value: ownership disputes or poor communication are signals to walk away, that would be a waste of money and time.

Speaking about money: to convince a bank, preparation matters most. I advise building a solid plan with clear data: development costs, technical debt, team scaling, migrations, and knowledge transfer. Showing you’ve studied the risks and know how to manage them makes funding more likely.

Technical decision making

What about using advanced tools to help you? ChatGPT won’t replace developers, but it shifts their role. From my perspective, it’s like IDE plugins or StackOverflow: a tool that speeds work but requires checks. Legal and reliability issues remain, and sources can be wrong, much like early Wikipedia. Still, it helps fight writer’s block and gives direction when starting a task.

But before adding (plugins or frameworks) or using external tools (AI, APIs), measure first. Without data, new tools don’t solve problems. Though the first tool I’d advise, no matter what, is Git. Even for personal projects: it protects your code and makes CI/CD easier later. Automation and CI/CD save time in the long run, except in short hackathon-style projects where speed matters more.

Plan your progression early: wireframing is an essential step in any project, it avoids mistakes. From my experience, dedicating time upfront prevents costly detours later. A clear design phase keeps projects on track and minimizes errors.

Listen to discover more, and hear how I turned ideas into a flourishing business.