Basic rules, learned the hard way.
People are led through challenges. People are managed to perform efficiently. Pick the right technique for the situation. Don't know the difference? If you are going to run a project, you had better find out. There are many books and resources to help you understand the difference between leadership and management. I will list some of them on this site when I get time.
Expect about half of your reports will buy in, and want to improve the process and/or the product to make the overall effort a smash. Half will not. They will want do to it they way they did it last time, just for the paycheck, and will not appreciate your efforts to get them to do otherwise. This is a good mix.
If more than half buy in, their assumed responsibilities and concerns will likely overlap and you will have squabbling and trouble getting things finished. Life is not fun when very serious experts clash every day in these overlap areas.
If more than half accept tasks only, and just want to get through it, you will have a heavy management load and will probably miss some opportunities along the way. It will become a drudge.
A 50-50 split is about right.
Expect to change about 25% of your staffing assignments in the first month.
Staffing is an inexact art, and subject to many pressures that force compromises. If the person doing the staffing is very, very good, the error rate may only be 10%, but like anything, the errors are in there. Look for these changes, and make them when you find them. People who are wrong for the assignment can have their specific duties changed, or moved off the job. Don't leave someone in the wrong position, and expect them to work through it. That usually just makes everyone miserable. There is no reason to expect perfection in selection of the staffing. Working with people is not a precise discipline.
Things will go wrong. This guarantees that if you just run to the original plan, you will end up over budget, schedule, whatever. You need to be constantly looking for methods and techniques to reduce the schedule and scope of work without reducing functionality. Don't wait until you are behind. Build reserves and slack.
Any task that remains high priority for longer than a few weeks is undermanned or misunderstood. Crush your priorities quickly and decisively.
If a person is not performing up to expectations, it is for one of four reasons:
They can't do it. They just don't have the ability to do what you have asked. Make sure it is something the person can actually do. Some people cannot troubleshoot, some cannot work with customers, some cannot take it up a notch in a pinch.
They don't know how to do it. They need training, and example, or just better definition of what they have to do.
They are unmotivated. They may not care about what they are doing. Do they believe they should perform at a high level?
They are evil. Some of the staff may not want the project, certain techniques, or certain individuals to be successful. While not exactly criminals, they don't have the best interest of the project at heart. They may go into passive aggressive mode, or worse, deliberately make decisions that have negative impacts on the program.
Most people's first response to an under-performing individual is to provide training, and it is a positive first step. But that only addresses problem 2) and sometimes address 3). If training does not work, then you have to consider the other possibilities, and realize that you may be dealing with 1), or 4).
Let each person know who they will be letting down if they do not perform. On a good team, the staff will respond more eagerly to each other than you or their manager.
Functional groups in your company are the equivalent of cost plus contractors. They will go down blind alleys if you tell them to, and you will get charged for it. They are also on the hook to improve performance, and you should hold them accountable for this. They may even have formal metrics you can use for leverage.
Keep the number of people in a meeting down. Seven is a maximum for a Roberts Rules meeting. More than that is a briefing, with questions. If there are two people in the meeting from the same function, then there is a problem. They likely are at odds with each other. Get to the bottom of it.
Let people know they should prepare for meetings, even status meetings. If they are not prepared, show no mercy.
Establish your lieutenants, define their responsibilities. See that they interchange with each other. If you are the only conduit between two or more leaders, then you have some headaches coming. If you are a control freak, you like these headaches, because they make you feel important. But every brain has its limits. I have never seen anyone micromanage more than 8 people at a time. If you cannot establish lieutenants, you are severely limiting your career horizons. Or heading to the funny farm.
Don't release anything unless you are sure the maturity of this object will support its next intended purpose. You don't know what it's next intended purpose is? It is not needed immediately? Then why release it?
Some tasks are solely to show progress. These are necessary, in that you will attract unwanted attention if you do not take them seriously. Limit these as much as possible. See the bullet above. Don't get trapped into thinking progress markers are the most important part of the job. In the end, the product has to satisfy the customer, and all those indications of progress will mean nothing if the final product is not acceptable.
During planning, consider the individual. Each of your staff can be thought of as being a combination of two characteristics:
Somewhere between Intelligent and Stupid, and
Somewhere between Industrious and Lazy.
That gives us four types of person you will manage. Some will certainly be at the limits of either characteristic, most will be somewhere in between:
Industrious and Intelligent. Give them lots of responsibility, and listen to what they say. Promote as soon as possible. That is where they are going anyway. Standing in their way may get you trampled.
Lazy and Stupid. Well known, widely available resource, and many techniques for getting good work out of them. Experiment, you will find something. Learn to work with these people, they certainly will be there. And, more importantly, you can get as many of these as you want. See Agile development techniques. This is the only type of engineer I would foist Agile/Scrum developments on. Good explanation here: Why Agile and especially Scrum are terrible. These techniques were developed assuming the developers are in in this category.
Lazy and intelligent. Give these people your toughest problem. They may take some time, but they won't waste resources, and come up with novel solutions. Strangely enough, these people love deadlines, unlike the others in this list!
Industrious and Stupid. Danger! Danger!! Get them Out! They are hazards! If you can't move them out, use them to manage bad subcontractors, or beat quality into something that has clear performance metrics and is well understood. But the best solution is to avoid at all costs.
Figure out who is who before you start your planning, and take the above into account. You will certainly get some of each. And you will not get ALL the best people on your job. Get over it. Work with what you get.
As a note, I have seen plans pitched that will only work if all the best people are pulled from their current assignments and placed on an opportunity. If this is what you need to be successful, then what you are doing better be pretty important, and not just to you.
A reward must be in the currency that the individual will appreciate.
If the individual needs money, a cash bonus will be appreciated.
If a person feels they are unnoticed, an awards ceremony will mean a lot.
If they are overwhelmed by tasks, time off is a good reward.
If they are high profile, opportunities to make an impression on the organization is a good reward.
If they are geeks, the most modern/aggressive tools will be very appreciated.
If they are academic, attendance and presentation at a conference will be appreciated.
If they are a competitive bunch, pick a prize that only one in the group can win. It can be a stuffed animal from the thrift store.
Pick the currency to match the individuals. And don't skimp.
We had a policy for spot awards for a while. These were budgeted to managers, to be issued with just a managers signature, to be immediately issued out of petty cash for a job well done - staying to finish on time, figuring out a tough problem, helping someone uncooperative. $50 or so a pop - handed to the person, by Friday. No announcement, just "Thanks, and here, take someone out tonight".
During discussions, focus on the issue, not the individual.