As a consultant, I’m often asked to help in the initial stages of a project to estimate “how big is the breadbox”. I completely understand this need. A famous teacher once said, “Don’t begin until you count the cost. For who would begin construction of a building without first getting estimates and then checking to see if there is enough money to pay the bills?”
I want to have these conversations with clients and customers because it’s important that everyone understand what’s at stake. I think it would be better though if instead of talking about an estimate we talk about a budget. Let’s have the exact same conversation but say “budget” every time instead of “estimate”.
Project Estimates versus Story Estimates
What do we even mean when we say “estimate”? We use that word in a lot of different contexts. Up front we try to “estimate” how long it will take to complete the project. Each iteration we try to “estimate” how much we’ll get done that week. While planning we “estimate” how long a single story will take to complete.
Maybe instead, we focus on “estimating” the complexity of a story/project. “Is this story a beagle, doberman, or a great dane?” Maybe we’re focusing on estimating risk.
I think there’s a lot of confusion in our industry from all the different ways we use the word “estimate.”
Today I want to talk about guessing how much time or money it will take to complete a project.
The Fallacy of Estimates
George Dinwiddie asks an interesting question, “What is the right length for a project?”. If we execute the project perfectly, we’ll complete it in the ideal amount of time. Estimating is the ability to guess that ideal amount of time.
George correctly points out that:
It is impossible to predict what set of alternatives will result in the shortest schedule. There is no perfect way to run a project.
Start with Discovery
I think the best way to start a project is to have an extended conversation about the project, it’s scope, risk factors, and time line. Let’s build some mutual understanding about the business context and the technical constraints.
There’s no way to get around it, the longer we work on a project the more we’ll understand it. The inverse is also true, the less time we’ve had to work on a project the less we understand it. What that means is that the day we start the project we understand it the least.
This is as true for the team delivering the project as it is for those who want it delivered. The day a client walks in with a project idea asking for an estimate is the time when we all understand the project the least.
At Gaslight we have a formula we use for a week long discovery to ignite your best ideas. Time and time again this process has produced several “Ah-ha” moments. The goal here is to have mutual, deep understanding of what we’re getting into.
Budgets vs Estimates
But how would the conversation be different if we talked about budgets instead of estimates?
After we’ve done the project discovery, most clients want some kind of an estimate. It seems reasonable to ask for one. We as engineers and consultants, no matter how reluctant, feel obliged to offer this estimate. Put yourself in the shoes of this business owner or manager or entrepreneur. How can they evaluate the project if they don’t know the estimate?
Recently, a client asked for an estimate and I was reluctant to give one. The pressure was on. This client has a multi-million dollar budget and our project was just part of that. How could he do proper planning without an estimate?
I said we should have the conversation but replace the word “estimate” with “budget”. My partner replied, “But we can’t budget for him.” He’s exactly right! We can’t budget for him!
That’s why it’s important to have the conversation. The client often can’t accurately predict the value the project will provide. But the client should have a hunch. If the project isn’t valuable, we shouldn’t do it – no matter the cost. We don’t know how hard or complicated the project really is; but we should have a hunch. The result of the conversation should be something like this:
If we could deliver these big, fuzzy features for this kind of budget, it would be valuable.
Once we have an idea of what the budget is, as engineers and consultants we can make a gut check to see if that’s reasonable. Can we deliver for that budget? One of my heros, Ron Jeffries, recently said pretty much the same thing.
Let’s get started
But Ron went further with two excellent suggestions. Both involve having a highly engaged “product owner” who really understands the business.
The first he calls, “Good, But Not Ideal”. Based on a budget, pick a date and some high priority objectives. Here’s the important part: deliver towards those objectives every week until you hit your date. That bears repeating: ship production quality software every week. You don’t have to wait until the end to reap the rewards.
Here’s the “ideal” solution he presents: build something now. Just pick a budget you’re willing to risk. Take a real product visionary and have them spend a couple weeks one on one on one with a designer and a developer. See what get’s built. Is it valuable? Is it worth continuing? If it’s working, keep going. If it’s not working, stop.