This is the second in a series about HOW TO do things in the content world.

The advice below is hard-won, which is another way of saying I haven't always followed it myself. It's less about how you manage a specific project, since that depends entirely on the work, and more about the mechanics that keep any project running well. Here's the process I try to hold to with every client.

Kick off properly

An introductory meeting with everyone who'll ultimately be involved isn't optional. If that means getting the CEO on the call, get the CEO on the call. Everyone needs to know each party's responsibilities up front, and the group needs to align on goals, KPIs, and timeline before the work starts. This is also the moment to question anything that feels misaligned with the core mission, while it still costs nothing to change.

Assign a project manager

If you have a dedicated project manager, the person whose job is making sure everything runs smoothly, count yourself fortunate. If you don't, someone has to take that on, usually as an unpaid extra role. That person sets and enforces the rules, holds the timeline, and has the difficult conversations no one wants to have. They need enough standing on the team that people actually listen.

Assign roles, specifically

Everyone on the project needs a role, and most people will end up holding more than one. Keep a document somewhere visible that names every person and their responsibilities. When two people share a role, drill into the specifics: is this genuinely joint ownership, or does one manage the other?

Simplify the tech stack

An enormous amount of time gets lost searching an inbox for something that's actually sitting in Slack, or wondering whether a file lives in Google Drive or Teams. Unless a client has real discipline here, different departments will default to different tools, and vendors and agencies will bring their own preferences on top of that. Pick the stack early and hold the line.

Cement the approval chain

Start at the end: who has final approval? Then work backward to who else needs input before that. If you're the client, this keeps the number of cooks in the kitchen honest. If you're the vendor, it tells you how many rounds of edits to expect, and it's your last real chance to revisit scope if the pricing assumed something different.

Check in on process at ten days

Projects that have quietly drifted off course tend to stay off course, because it's hard to get off the treadmill once it's moving. A ten-day check-in doesn't need to be a formal call, and it doesn't have to come from the vendor, though it usually does.

Hold the debrief, always

Everyone wants to move straight to the next hundred deliverables, on both sides of the table. Do it anyway, even when the project went sideways and everyone already knows they're not extending the engagement. The debrief is where the actual lesson gets named instead of just felt.

This is one version of a process, not the only one, and it depends on transparency and open dialogue to work at all. But it's a far better bet than the alternative: a project that runs on assumptions no one ever said out loud.