What gets an outcome accomplished? Each principle below is one line you can use on your own work. They come from cases in history, fiction and ordinary life: a moon landing, a mine rescue, a wedding, a website relaunch.
Setting up
Write a goal people can act on without asking. Put the end state, the deadline and the condition that can’t be dropped into one sentence.
Get the honest number, then keep to it. Reset the budget once detailed planning shows the real cost. Then measure everything against that number.
Decide what gives when the date can’t move. If the date is fixed, money or scope has to flex. Agree which one before you need to.
Define done so you can see it. If done means none left, you need a way to find every place that isn’t finished.
Hire the lead, then let them build the team. And train the client’s own people, because they keep what the project leaves behind.
Planning
Take small steps and build in slack. Add a little load at a time and allow for slippage from the start, so one bad week doesn’t end the project.
Plan for the first attempt to fail. Give the second attempt what the first one learns.
Back more than one approach when you can’t tell which will work. Expect some to drop out, and budget for it.
Run plans side by side when waiting is the bigger risk. Plan B doesn’t have to wait for Plan A to fail.
Build a small version in public first. Let people use it before you commit to the full build, and let what they do decide what comes next.
Treat the customer’s data as part of the deliverable. A new system that loses the people who used the old one hasn’t been delivered.
Running
Set a rhythm and don’t break it. When demand keeps rising, fix the interval and work on the turnaround.
Find where the real work is being done. The critical path can run through the smallest team. Check whether the busy, visible work is helping it or getting in its way.
Make someone own the handover between teams. Where one team’s work meets another’s, decide who’s responsible for making it fit.
Give people a reason to join. When nobody has to help, the cause must do the recruiting.
Tell every workstream when the goal changes. A team still working to the old plan pays for it.
Under pressure
Stabilise before you plan. Stop the damage first. Then plan the recovery from where things have settled.
Ready isn’t the same as go. Being ready and choosing the moment are separate decisions. The other side’s plans are part of your schedule.
Know the cost of waiting. When a window sets the date, the choice is go or no-go. Know exactly what the next window costs.
Say so when the plan no longer supports the date. A detailed plan is how you see it happening.
The cases behind each principle will be linked here as each one is published.