Estimating cost projects isn't easy to do, even if you have all the information you need. Bob Artner cautions about estimating "off the cuff" and offers some tips for avoiding this predicament.
It’s late afternoon and you’re trying to get to your next meeting on time. Suddenly, the business partner for one of your projects grabs you in the hallway. “Did you see the upgrade pricing the vendor sent us?” he asks. “It looks like the cost savings could help us get the project approved.”
Anxious to get to your meeting, you offer a quick reply: “I’ve got to dig into the details more, but it looks good.”
Your partner presses: “How much could the new numbers end up saving us?"
“Like I said,” you reply, “I need to look into it some more. Now, I’ve got a meeting—“
The business partner cuts you off. “Give me a ballpark number.”
“A ballpark number?”
“Sure. I just want to know what we could be talking about.”
You are now late for the meeting, so you think for a moment, and then give him a number, based on what you know about the upgrade pricing changes. He smiles and you part.
A couple of weeks later, your boss calls you into her office. It seems that she just sat through a PowerPoint presentation that included a slide showing that your department believes that vendor upgrade pricing can save a lot of money, quoting your ballpark estimate as if it were a complete assessment of the pricing impact.
Another headache I just don’t need, you say to yourself.
In this column, I want to look at the world of ballpark estimates and why they can be dangerous to your professional health. I’ll also present some tips on how to avoid the headaches they cause.
Ballpark estimates: Why they want them, and why you should shun them
Having been on both sides of this process, as an IT manager and as a business partner, I can understand the consternation surrounding ballpark estimates.
From the business partner’s perspective, he or she is trying to get a project approved, outlining its goals, and (usually) working on the feature specifications. Getting this project approved and implemented will help the business partner achieve his or her department goals.
From the IT manager's perspective, however, there is a hitch. Any project coming out of IT, no matter how big or small, needs estimates from technical managers about cost and time to implement. The business partner can’t do those estimates and needs people like you to do them. Having to wait for those estimates to be done can be very frustrating for any manager, especially one who’s trying to solve a problem that affects his or her group every day.
For the business partners, it gets even worse. They often can’t even propose a project until they get a rough idea of the cost vs. benefit ratio. Therefore, they need some guidance from IT before knowing whether a proposed solution makes sense. This means, in effect, that they need two sets of estimates: the first, a set of rough estimates to know whether a project is worth investigating, and another, much more detailed set of estimates, in order to get final project approval.
As you can see, ballpark estimates might be valuable for that first group and completely inadequate for the second.
Now what about you, the IT manager? All your objections to ballpark estimates fall into one category: They aren’t accurate. Anytime you have to access the total cost of a project without full information, you’ll make mistakes.
Accurately doing project cost estimates is hard to do, even if you have all the information you need. That is why doing accurate estimates is one of the biggest topics in our Project Management channel. After all, there are consulting companies whose entire business is built around expertise in estimating things like software licensing and upgrade costs.
So what should you do?
Understanding this dilemma is helpful but doesn’t get you out of the soup. What do you do when asked for a ballpark estimate by a business partner? Here are some ideas:
- Just say no: The simplest solutions are often the best. Have you ever tried just refusing to make a ballpark estimate? You don’t have to be a jerk about it. Try something along the lines of, “I’d love to help, but there are too many unknowns for me to even try. If you want an estimate that is at all useful, I’m going to need to go through all the data and get back to you with an analysis.”
- Explain the difficulties: Business partners aren’t typically unreasonable. If you explain the factors that go into making an estimate for a particular project, they will often understand and respect your reluctance to produce a back-of-the-envelope number.
- Ask what the estimate is for: As I've discussed, a ballpark estimate can be useful for a business partner to decide whether a particular technology solution is worth pursuing. On the other hand, such an estimate is not useful as part of a final recommendation for project approval. Before spitting out a number, ask why the business partner needs it.
- Give a range: Rather than giving a single number, you should consider expressing your estimate as a range. For example, one of my colleagues at another company was asked to give an estimate as how many units we could sell of a particular product. He tried to demur, noting that he had only two data points, suggesting vastly different outcomes, and he needed more time to determine where the truth lay. When the boss insisted on a ballpark estimate, my friend replied (in a phrase that became legendary within the company), “Between 25,000 and 1.25 million.” Of course, it wasn’t very useful as an estimate, but that was the whole point.
- Cover yourself: Regular readers of this column know that I’m uncomfortable with CYA communications, but here is a case where I think caution is warranted. If you must give a ballpark estimate to a business partner or supervisor, I would definitely follow it up with an e-mail reminding the person who requested the estimate that it was provisional and based on limited information. I’d also send a copy to your supervisor.
What do you think?
Have you been burned by a ballpark estimate? What alternatives do you suggest? Drop us a line or post a comment and let us know.