Five Key Recommendations on Managing Project Costs

As regular readers know, I’m all about making project management simpler. That’s not to say easy — projects will always have their complications. One of the key responsibilities of a project manager is to finish the project with a result that delights the customer (or your management team), and that delight is not likely to happen if you run over budget without a very good reason. In this blog, I will outline five key recommendations on managing project costs. None of them have anything to do with earned value management — my recommendations are much simpler.

First, it helps to understand why projects run over budget, and why yours is running over budget. There are several big reasons:

• Requirement unknowns or confusion
• Poor estimating of how long a properly defined activity will take
• Unmanaged scope creep
• Managed scope creep

In today’s world, some projects will simply have more clarity at the beginning than others. When the budget gets blown because someone misjudged how long an activity would take (assuming a clear understanding of the requirements) or because people started doing little extras that weren’t approved by the client, the client has a right to be unhappy. Those are simply unacceptable reasons to exceed budget.

On the other hand, when the budget is blown because there were simply too many unknowns in the beginning, and the scope had to be figured out on the fly, most clients are pretty reasonable (provided your communications are effective). I’ve found once you really get into a project, sometimes it makes good sense to change things, even if it increases costs. You learn things as you go along that allow you to make smarter decisions.

When a project goes overbudget due to excessive re-work, teams need to understand why. Are teams having to re-work deliverables because of unclear expectations, changing requirements, lack of needed skills, or poor communications?

When vendor disputes occur, try to understand why they are occurring. Again, is it unclear expectations, changing requirements, lack of needed skills, or poor communications? There may be other reasons for vendor disputes or re-work. It’s worth your time to understand why these problems are occurring — preferably earlier in the project.

So, what project cost data are important? How do we get it, and what do with do with that data to ensure that we are doing a great job of managing project costs? Here are some ideas:

Have a clearly defined process for creating a project budget.

There are different opinions about how to create a project budget, and how much time to spend on that process. The newest PMBOK guide spends 40 pages talking about that process. At what point does adding time and effort to the activity of creating a budget cease to add value?

I have written before on how to break down a project (decompose) into the essential activities. I’ve talked about the need to distinguish between the essential activities and the individual tasks that need to be done to accomplish those activities. As I have said before, we first need to understand the project activities, including the quality needed and who is involved, before we can determine how much it will cost us to produce them.

For example, take the activity to create a script for a marketing advertising campaign. Is the expected outcome a draft script that will go through more conversations and rewrites later? OR is the expected outcome the final script — which must be approved by a number of executives? The more people who have to sign off on a deliverable, the longer it will take, and the more it will cost.

Once we know what we can about the activities, the next question is how to assign resource costs. Going back to the commercial script example, if you the objective is to get a final script with approval from five executives, being paid different salaries, are we really going to try to decide how much time each executive is going to need to spend on this activity?

I would argue that at some point, precise estimating loses its value. There are simply too many unpredictable factors. Could we predict that in the week prior to completing this deliverable, something would happen in the world that would make the drafted script an unwise choice? Could anyone predict that one of the executives would have a family emergency that resulted in her unavailability — and how that might impact whether a team can declare the script activity completed, or not?

Budgeting is an estimating process and by definition, estimating is not perfect. So the question is what data are you trying to capture? I propose simply having the team estimate the cost of each activity, given the best available data that you have. I don’t usually break out individual labor rates, but I use an aggregate rate, given the people I anticipate being involved.

Develop buy-in from your team on living within the budget.

And then, I work with the team to develop some buy-in on living within the budgeted amount for each activity. Using the script example, if the team developing the script understands how many levels of executive approval are needed, he/she can plan accordingly, and do the best possible creative work in a reasonable amount of time. If the team is unaware of the budgeted amount, and the number of executive approvals, it is working in a vacuum and more likely to run up the costs.

So, if the activity exceeds budget, was it because the client didn’t like the creative work of the first three efforts, or was it because something happened that caused the original work to need to be re-done? The answers to that will drive how you build your team effectiveness and improve your client communications. You may choose to replace a creative who just doesn’t click with a client, or you may decide that the creative needs to have more time with the client. It totally depends on the project — but the focus should be on improvement. You’re only talking about one activity in a much larger project, so you have time to get better.

Ask your client what kind of budget breakdowns it expects.

Not all clients are the same. Some are more cost sensitive than others. Does your client want a breakdown of costs on every single activity identified in the work breakdown structure, or are they happy with a breakdown at the summary level? Should the project be set up in phases, with cost summaries by phase, or does the client want summary cost reporting by functional areas?

On a project of any size, with a client who is cost sensitive, I’d recommend that tracking your costs at the activity level, as opposed to the summary level — even if you only provide summary reports to the client. This will allow you to more quickly identify problems, and the reasons they are occurring.

Consider the project cost data that are needed to improve your team effectiveness.

Earlier, I mentioned the question of whether you should break down the cost of each activity into the labor components for each resource. This is a tough question — and the answer may depend on what you’re trying to accomplish. If you’re trying to build a more effective team, does that data help?

I don’t generally think it does, unless you have a problem person on your team and you need to understand whether he/she has a pattern of underestimating costs. If you’re using a team rather than an individual to create your estimates (which I typically recommend), the voice of one person who typically underestimates shouldn’t create much of a problem. And the open dialogue about the estimates can be a learning process for that individual.

In general, I’d recommend you track the minimum amount of data needed. If you have budget, and actual data for each activity, and your project is moving along steadily, what other cost data do you believe are needed to improve your team effectiveness? I certainly don’t think tracking cost data on individual tasks improves performance.

Create a policy on whether you will track employee labor costs.

Are you going to track in-house labor and overhead? If you’re trying to compare two projects — one being done by independent contractors and one being done by salaried employees, you can’t do that unless you’re factoring in some labor costs for the employees. So, it’s important to have a policy on what costs are being tracked and understand whether you are comparing apples to apples, or not.

Personally, I recommend that organizations track employee labor costs, using an end of day, best guess. By that, I mean that at the end of a day, every employee can write down (or enter) the approximate amount of time devoted to each project activity that he/she is working on. Remember that we aren’t talking about tasks here. If you have a team focused on a small discreet number of activities, there won’t be a lot. But I think something entered into a tracking system is better than nothing. And at the end of the day, that assessment of productivity and where we’ve spent our time is good for everyone. It’s about improving ourselves and our teams — for everyone’s benefit.

I agree that some cost data are essential for properly managing project costs. But, it feels to me like we’ve made it much more complicated than it needs to be. When our data helps us improve, it serves a useful function. When it’s just about creating colorful reports, it doesn’t have much value for me. Why do we need the data? Is it about control? Is it about checking a box? Do your reports really provide any root cause analysis of why problems are occurring? If not, something is wrong.

Simon Sinek, in his book Start with Why: How Great Leaders Inspire Everyone to Take Action says: “Logic dictates that more information and data are key.” (p. 13) And so, the tendency is to create more and more data on which to base decisions. And then, to think that those decisions are based on data, and therefore are better decisions. I would argue that managing project costs takes more common sense and commitment than beautiful reports.

Managing project costs takes more common sense and commitment than beautiful reports. Click To Tweet

I’d love to know your experiences. Share them in the comments field, or schedule a call and we’ll talk.