Stepping Back for a Broader View

As much as it pains you, working with the reality in front of you is more productive than working with your hopes and wishes. That means you’ve got to face the reality of your situation and deal with it. Blaming the estimates, estimator, or implementor for not meeting your plan is a destructive form of crying over spilled milk.

When things aren’t coming together as you planned, it’s time to change the plan. Nothing else is likely to work. Let’s hope that you make the decision to replan early enough to accomplish the things that matter to you. There are some ways to detect problem earlier, which gives you more flexibility in your replanning.

The foremost way to get early warning is to welcome bad news. Celebrate when you receive it. Almost always there is someone who knows things are headed for a train wreck long before that wreck happens.

Another way to spot problems early is to integrate systems early. The most pernicious problems in system development are disagreeing assumptions between components. Don’t leave that until last. Connect your systems first and then make them work correctly together.

Work in a fashion where progress is unambiguous. See things working to indicate progress. Estimating completion percentages is a great way to hide missed assumptions. Use functional slices, and measure them in a Boolean fashion; they’re either done or not.

For safety, reestimate portions of the system from time to time. Are the new estimates in line with the old? Or have you learned something that changes your assumptions?

When the actuals don’t match your plans, you’ve surely found something that violates your original assumptions. Dig in and figure out what it is. Don’t be too easily satisfied, as there are likely multiple assumptions that need rethinking.

The questions you might ask yourself include:

• When we made this estimate, what did we assume that turned out to not be true?

• What has changed since we made this estimate?

• Have we seen similar issues with prior missed estimates?

• When could we have noticed this if we’d been checking our estimates periodically?

• If we account for these changes in estimates of future work, how does that change our plans?

• Does our new plan have enough safety margin for severe risks?

Do all of these things, and you can improve your future prospects. That’s something to celebrate, isn’t it? Does that make it easier for you to welcome bad news? It’s a blessing in disguise.

In the next chapter, we’ll look at ways to test our assumptions even earlier and more deliberately.