During my early days in the corporate world, one of my jobs included developing the software that synchronized and backed up all the data needed to control the generation and transmission of power for Great Britain. The project manager made the rounds every Friday to update his Gantt chart. He stepped through each activity and asked me for an updated percent complete. I struggled with this exercise and finally, when my tasks hit 67%, I refused to play. Despite my long list of open tasks marked two-thirds complete, I promised him we’d be finished on time.

This tracking ritual continued weekly and every Friday he could hardly contain his anger with me. He told me my group was going to ruin the whole project. I told him not to worry. My promises did nothing to relieve his anxiety.

Meanwhile, all the other leads were marching steadily toward the finish line and keeping him happy. 70%. 75%. 78%. 80%. 82%. To increase the pressure on the laggards, the charts were posted on the wall. Practically everyone was in the 90s and my group still showed 67%.

The day before our deadline, we finished. All our software components worked with each other and passed the system tests with flying colors. Just as I promised. It was time to mark the boxes complete and I did.

He was still unhappy. Why? Because we were the only group that finished on time. The others were trying to squeeze new integers between 98 and 99%.

Why Did I Refuse To Update The Numbers?

I wouldn’t play because it was a meaningless game. We were doing things we had never done before. When you conclude that you are 90% complete, you are assuming that you understand everything left to be done and that all will go exactly as expected. All it takes is one false assumption to throw a wrench in the works and some of those wrenches can be enormous. It doesn’t matter how many hours you’ve put in. It doesn’t matter how many lines of code you’ve written. What matters is whether everything works correctly in the end. This is why the first 90% of coding usually takes 90% of the time and the remaining 10% takes the other 90% of the time (Tom Cargill, Bell Labs).

How Did I Know We Would Finish On Time?

My group finished on time because instead of focusing on:

• Checking things off a list
• Confusing the plan with the goal and
• Trying to look good

We focused on:

• Our software
• The deadline and
• The things that stood between us and our deadline.

To understand the difference better, consider the following chart. The left-hand side represents tasks that are familiar to you. These are things you’ve done before, at least mostly. The right-hand side represents tasks that are unfamiliar. They may be completely new to you or even completely new period. The top half represents simple tasks and the bottom is for complex tasks.

Quadrant 1 covers the simple and familiar parts of the plan. You might be able to do some of these in your sleep. Quadrant 2 tasks make you work and think harder, but they aren’t mysteries. These tasks might require a lot of care and/or coordination with others, but they don’t require a rocket scientist. Quadrant 3 tasks are pretty simple, but since they are unfamiliar to you, you don’t know what you don’t know. Quadrant 4 tasks are where calling in the rocket scientists might be a good idea. You’ve got both complexity and unknowns. These tasks will require some serious learning, heavy duty experimentation, and significant coordination, as well as some luck.

Why Traditional Planning And Tracking Fails

Take a look at traditional planning and tracking practices relative to these four quadrants.

• Which quadrants are most easily planned? 1 is obviously the easier. Next comes 2. Then 3. Last, 4.
• Which are most accurately planned? 1, 2,  3 and 4.
• If you want an accurate schedule, where will you spend the most time? Ah, here is a slight change! 2, 1, 3 and then 4 is the order. Accuracy is possible with quadrant 2 tasks, but it takes work and time.
• If you want to show quick progress, either to get off to a good start or to make up for lost time, where would you focus your attention? 1, 2, and then 3 and 4.

It’s only natural. Traditional planning and tracking practices draw attention to 1, 2, 3, and then 4. In that order. But the consequences are horrible!

• Which quadrant is most likely to blow the schedule out of the water? 4, 3, 2, and then 1. In that order!
• Which quadrant is most likely to blow the budget sky high? 4, 3, 2, and then 1.
• Which quadrant is most likely to be responsible for creating unhappy customers? 4, 3, 2, and then 1.

Most plans simply create the illusion of control and distract from the true priorities.

1. Don’t Let The Plan Become The Objective

Plans are just tools. They are not the goal. Checked off boxes are not results. Don’t let the plan become more important than real progress.

2. Avoid The Questions That Make People Tell You What You Want, But Don’t Need, To Hear

Most project managers think they can get results by holding people accountable with questions such as:

• Are you on schedule?
• Are you on budget?
• Are you going to meet the performance requirements?

How do you think most people answer these questions? They say yes, scramble off, and hope they will never have to reverse their position. The illusion of control is sustained!

3. Make The Unfamiliar Familiar

The familiar tasks are so easy to plan you barely need a plan. If you want to control the schedule, budget, and quality, concentrate on learning so the unfamiliar becomes familiar. Once the unfamiliar is familiar, planning is easy! To speed this transition, adopt a “defense driving” mindset:

• What have we learned?
• What do we still need to learn?
• What are we taking for granted?
• How do you know you will finish on time?
• How do we know we can meet the customer requirements?

These questions take the focus off the plan and put the attention on the only things that matter: the goal, the deadline, and the obstacles that stand between you and your deadline.

This article originally appeared on Forbes.com on October 6th, 2015.