Coping with a Fear of Inaccuracy

September 24, 2014

“Even imperfect answers can improve decision making.” – Donald Reinertson

When I read this from Reinertson’s book on flow, I realized that I had found the reason that people have so much trouble with story points. It’s a matter of overcoming their fear of inaccuracy. They are under the misguided belief in the accuracy of using hours or days to estimate work on projects. They’re basically afraid of being wrong (aren’t we all?) and that is the source of a lot of resistance to change. Being wrong sucks. I get that. Nevertheless, I’m wrong a lot.

Fortunately, wrong isn’t always boolean (unless you happen to step in front of a swiftly moving bus). There are shades of wrong. You can be just a little wrong, your aim just a little off, and still be headed in the right direction. Or you can be a lot wrong (the bus). That’s where frequently re-examining your decisions can help you catch the stuff that’s a lot wrong and fix it. What about the stuff that’s a little wrong? Don’t sweat it.

In the software world, a little wrong is still pretty useful. There is a tremendous amount of error and missing information. Compared to that, being slightly wrong isn’t so bad. Being slightly wrong gets you started – at least in mostly the right direction. You’re going to fine tune it anyway, so there’s really no need for decision making precision. That will come later, when you know more.

To me, the ability to overcome our fear of being wrong stems from an all-or-nothing mindset. We can’t allow ourselves to be even a little wrong for fear of failure. As Reinertson rightly points out, there is a time and a place for precision in decision making, but it’s not ALL the time.

Planning Feedback: Don’t Panic!

September 9, 2014

Ring the Alarm

So the other day a VP asked our team for an estimate on a project. Now, putting aside whatever feelings you may have about the usefulness of estimates, we did a little planning, a little investigating, a little debating, and came up with an estimate for the project. I brought the estimate back to the VP and then the fireworks began. Apparently he had been thinking of a different number than we had given him. He wanted it done in half the time that we had forecast.


Now at this point in the story a lot of teams will panic and come back with one of two reactions:

  1. Fight – “We can’t do that! That’s not Agile” (or some variation on that tired theme)
  2. Cave – “We have no choice…”

But wait a second, there is a middle path. You can agree, but ask what can be compromised in terms of scope or other project constraints. You see, a new project is not just a learning process for you. Its also  learning process for the customer too. When  you get that first feedback, DON’T PANIC!

There is one important part of the planning process that I often see get lost: iteration. Doing a single round of planning and then presenting it to your customer as “Take it or leave it” isn’t what I’d call much of a dialog.

What we should be doing is some lightweight planning then review with the customer. “That’s horrible!” They cry, and then you say, “So what number were you thinking of?” And they return with something totally preposterous. OK, that’s cool. “Is there functionality we can drop to hit that date?”


Of course not. So you go back, you scratch your head, cut out all the fluff you can find…and you still are way off the desired estimate. So then you take it back to them and say, “Hey look, I know this isn’t what you wanted, but here is the ABSOLUTE minimum we can get away with.” At which point, the customer looks at you with a tear in their eye and says these magic words, “What CAN you give me?”

Now you have a real negotiation! It may not always play out this way, but hopefully you get the point. You can’t freak out when they give you that first reaction. Stay cool – this is the first time they’ve had to test their fantasies against your capabilities. They need to learn too. So give your stakeholders a chance to work together with you on figuring out just what is possible. Negotiate. The more iterations you go through, the better your chances of coming to an agreement that everyone agrees is the best.