Avoid the planning fallacy to stay on schedule and on budget
How many times have you, under pressure and hoping to impress, given a wildly unrealistic estimate on a project and then found yourself saddled to a task you can’t possibly accomplish? Estimates are tricky beasts at the best of times. But when we’re delivering them on the spot we often fall victim to the planning fallacy.
The planning fallacy happens when you create subjective estimates based on best-case scenarios and with zero reference points. You look at your team, you look at the project at hand, and you think “we’re talented and we know how to do this. I bet we can get it done in X hours.”
These estimates fail, time and again, because we base them on positive thinking rather than evidence. Instead of thinking in best-case terms, look for an evidence-based reality check. Here’s where to start.
Look at previous projects.
The best cure for subjective estimates is to compare your current project against previous projects, or other teams who have accomplished similar tasks. How long do your projects usually take you? Have you found that you’ve over- or under-estimated in the past? How can that inform the estimate you’re making today?
Don’t assume you’ll do better this time.
It’s tempting to look at past projects and say “well, we messed up on that project because of these reasons, but we’re sure that won’t happen this time.” And you may be right: maybe that specific thing won’t happen. But something else will.
Estimates usually fail because of our inability to predict the future. Somewhere along the way, the scope of the project will shift. Requirements will change. A core team member will get sick, or a client will fail to meet approval deadlines. We don’t know which of those things will happen, but we can be sure that at least one or two of them will. Plan on it.
Make short-term predictions.
Programmer Ron Jeffries, in his article “Estimation is Evil” argues entirely against the idea of making estimates. Instead, he recommends working in short cycles which force the prioritization of the most essential things and allows for quick learning and more iterations. It isn’t always possible to scope out projects this way, and it’s even less likely you’ll sell a client on this model. However, there is some wisdom to the idea that you should focus on short-term predictions.
It’s hard to predict who an entire project will go. However, you’re likely to produce much more accurate estimates for shorter amounts of work over a shorter period of time. Break your project down into pieces, and tackle them as you go.
Be flexible, and don’t expect miracles.
Subjective estimates, because they are disconnected from reality, can leave you and your team stressed. They can also result in a lower-quality project than you’d ideally like to deliver. We offer them up because we feel pressured to do so in order to please a client or to meet expectations of a manager. The best way to avoid them, particularly when we feel squeezed to do better, is to counter with a reality check.