The dark side of estimations
TL;DR
Yet another ranting article against the all-powerful estimations in time management. Estimation is your enemy, not your ally. Don’t trust them
You asked me how much time this task will take. Maybe one day. Maybe two. I don’t know. Actually, I don’t even want to give you an answer, just tell you a placeholder answer that won’t please you. But believe me, it’s for your own good.
Well, I know you want to know how much time this will take, for real. it will determine your budget. Your planning. Your charge. Your priorities. Well, almost everything. This would be wonderful to be able to tell that that task will take exactly 137 minutes. Really, I do agree with you on this point. But the fact is, this will never happen.
Estimations are the obsession of managers. Do not blame them for that. That’s perfectly understandable. This tool is a wonderful way of organizing, planning, budgeting… Without an estimation, the future is just a terrible darkness where you can’t be sure not to fall from the cliff when you put your foot on the ground.
However, what’s worst than working on estimations for days, do all-nighters to have that perfect plan for the next three months, and finally see after three days that nothing goes like hoped?
Let’s get back to basics. What is the true nature of an estimation? Well, just try to estimate simple tasks you do every day. Empty the dishwasher? 5 minutes. For real, run a timer and see the reality. I guess you will be easily 20% far from your estimation, and that’s probably a task you’ve already completed hundred of times in the past. Get out to buy a sandwich for your lunch? 10 minutes. Arf, there is more people than usual, it took you 20 minutes. Doing the shopping? 30 minutes. May have been so if that article you absolutely need hadn’t been out of stock. Hmm, it took you one hour, finally, after visiting another shop to find what you needed. Estimate a project with 92 tasks? Good luck. Impossible with a good precision.
If you can’t estimate a trivial (short and known) task with an absolute precision, how would you be confident in an estimation on a set of tasks that would take days and that you have never accomplished at least one time in the past? Human brain absolutely sucks at estimating. Even for well-known tasks. The reasons are we don’t have a good notion of the time, and we can’t foresee the unpredictable.
Ok, an estimation can’t be precise. But that’s better than nothing, uh?
What if I ask you 10 dollars/euros/whatever to give you an answer you know will be false? Ok, give the money. Well, I can tell you the Earth is flat.
Disappointed? You have lost 10 bucks for something totally useless. So are estimations. When you produce an estimation, you spend time to collect data and think about the tasks. All that time is lost, while it could have been used for something really valuable. Like… doing the task, for example.
But you give voluntarily a wrong answer to my question!
Yes, and this will happen with the people that will estimate for you. Yeah, they will lie to you. Why? They may not do it consciously. But what happens in their mind? They have to produce an estimation. They know there could be something going wrong happening to them if they estimate badly. Well, you may not punish them but their credibility would be engaged. They said 2 hours, but it took them 4. So the next time, they will tell you 4, even if they feel the task will for real take only 2. Estimations is an engagement form. A moral engagement.
But I really need this estimation!
Ok, let’s do a step further. I give you an estimation. What you ask me will take 20 days. I start my chronometer. I start my pressure-meter too. The race against the time has started.
20 days later, you come back to me. Uh Oh. I’m only at the half of what I “promised” you. You’re in a bad situation. You don’t have budget anymore. Or you planned the publishing of your project to your customer. Anyway, you’re in trouble.
Now, let’s take the opposite example. I tell you the project will take 40 days. Immediately, you say this won’t be possible. You don’t have the budget, or whatever. Yeah, but how are you sure this won’t be faster, in the reality? Maybe I’m just far away from the truth, and just imagine difficulties where it will be easy. You’ll stop a promising project for a wrong reason. Doh.
Ok, let’s sum up all this mess.
Here is what you hope:
Here is what you get:
Are you sure you still want me to estimate? Are you sure you want to pay for that? Maybe not really…
