An engineer friend was recently promoted to project manager, and asked me if I could give him some advice on the role, based both on both doing the job and working alongside effective ones. This was my 15-minute brain dump:
First things first: determine what your role actually is. Product, project, and program management all mean different things at different companies, and often have significant overlap. Identify whether you're responsible for the timeline, quality, and communication around a project (what I'd call 'pure' project management), the what and the why (product management), or the meta-strategy across multiple projects (program management).
Next — and this assumes you're a standard project manager— identify what's going wrong. I always like to ask this when I'm interviewing for a role: "Why are you hiring this role at this time? What problem needs solving?" You'll want to talk to your team, your boss, others in the role, and try to come out with a list of pain points. These can be really high level: "Marketing doesn't know when stuff is coming out." "We always miss our deadline by weeks."
One of the biggest mistakes you can make as a project manager, or as any kind of manager really, is spending cycles fixing what's not broken. Time is expensive, including time to research, learn, and support a new process or tool, and unless you're doing it to address a current or clearly upcoming problem, you're bleeding people dollars doing it. Working on a single list of pain points at once, and making that list visible to people, also makes it easier to explain why you're not doing a zillion other things someone might expect a project manager to do.
More assumptions ahead, and YMMV depending on your company, the culture, and the specific stuff you're working on.
If you have deadline problems, your first priority is making the invisible visible, and highlighting the gap between the plan and reality as soon as possible— not right before you're supposed to ship.
You need work items captured somewhere, and a simple dashboard of how much stuff has to be done before the project is done. Google Sheets, Asana, JIRA, Rally— what you use is less important than that you use it consistently and get buy-in across the org. If everybody is happily aligned on a task management tool, your job gets much easier: just make sure stuff is organized reasonably well, and figure out how to create a project-level view of what's done and what's left to do.
The sorts of people that become project, program, and product managers are often perfectionists, and will go off researching the perfect solution for their special snowflake of an organization. Don't. Don't let great be the enemy of good. It's always fine to track new tooling, but implementing anything new burns attention, time, and political capital, and you want to save those resources for the important stuff.
At a really tactical level: take what's there, build a dashboard view on top of it if you can, and send out an update to people on a regular basis (weekly generally works well to start). You'd be amazed at how much wishful thinking and opaque, fuzzy timelines disappear when everybody involved learns what's left to do, what happened the last week, and what the team's looking to do the next week.
You also have to, have to build relationships with the project team and all the stakeholders. It's easy to see a project manager as a second boss or a taskmaster, and your job is to show everyone involved how you're taking work and stress off their plates and making their lives easier. Spend 1:1 time with all the team members if you can— let them vent their frustrations, listen empathetically, look for gaps in information and process you might be able to fill for them.
When you have cross-team or completely external dependencies— and you often will— this is even more important. Realize that people external to your company/team have their own list of priorities and pressures, and until you're a real human to them, you're an abstract 'other' and will frequently be bumped to make a more pressing deadline. If your time and money allows, go all-out with these folks. Visit their office. If they'll be a major and constant partner, fly there. Check in regularly over video, ideally with your team present. Your goal here is that when they look at their list of things to get done and realize they don't have time to do it all, you and your team are in their minds as partners and people, not faceless others.
Project management is a huge career, and it's one I've only scratched the surface of. There's agile, scrum, kanban, lots of different methodologies, plus a world of specific software and tools to manage deadlines and deliverables. I've seen project managers be successful and unsuccessful with basically all of them, but the successful ones, across the board, were those that focused on solving real pain points, using as little change as possible to get what they needed done, and built strong relationships with the people they'd be working with.