The Dash for Dashboards

As long as I’ve been in the project management business (well over 20 years), in every company I visit, there has always been an executive who asks when he can get the one-page project management status report. This report would include the full range of “real-time” indicators letting him or her know what the state of all projects.

The state of technology at the moment makes it very possible to create dynamic displays that can reach into a variety of corporate data sources (including your enterprise project management system) and produce these visually stunning screens.

Lately project management system sales people have been showing these kinds of examples to management types. You can easily imagine the reaction. It’s like showing candy to a baby. Management gets very excited.

What is it about dashboards or even that elusive one-page report that gets management excited? The reason turns out to be the simplest of all. When I speak with senior executives, the requests I receive are very common. There is no shortage of data for executives to sift through. There never has been. What is lacking is information.

Executives are placed in positions of responsibility and are held to account for the results of that area of the business. What many executives lack is the knowledge of when they should take action and when not.

Let’s take some data with two perspectives. A project was planned to last 12 months. It’s budget, $100,000 per month or $1.2M for the year. It’s six months into the project and the spending to date is calculated by Finance at $900,000. Ignore for the moment, the notions of what “spending” means. We’ll leave the compexities of “actual”, “committed” and “requisitioned” for another article.
Now, you’re the CEO and you’re at an emergency meeting which has been called by the CFO to discuss the “runaway project”.

The CFO tells you that the project is 50% over budget and is going to cripple the company. “It must be cancelled immediately,” he says.

The Project Manager swears the project isn’t in trouble. It’s doing very well, he indicates and shows a series of barcharts showing how the project is running ahead of schedule. “The project should not only not be cancelled, it should be held up as a showcase of job-well-done,” he says.

You’ve hired these people because they’re very good at what they do but both are asking you to make opposing decisions. What do you do? Well, likely nothing. You send the two managers off to argue between each other and return with a consensus. You, as the senior manager are paralyzed. The data on both sides of the argument seems valid but you are left unable to take action.

This type of scenario plays out in organizations every day. In the end, the two managers may find that the project is, in fact, under budget but spending money faster than projected because it is running 30% faster than expected and will ultimately conclude ahead of schedule. Had the project been cancelled, as the CFO had recommended, it might well have crippled the company.

Executives tend to ask me the simplest questions:

“Are we ok?”

“Should I take action?”

“Is this problem serious?”

An executive I spoke to several years ago pointed to a pile of paper about 6” high on his desk. “That’s this month’s project reports,” he said with a sigh. “I can tell you that project #5 has task #215 that is 6 weeks late. What I can’t tell you is whether that’s a bad thing. I truly have no idea. Can you help?”

This executive had 23 project managers who used no less than 14 different tools and methods for creating those reports. When we were done with our project there, we at least had the data all in one flavor and a summary of the reports with some variance analysis. It took a while, but it was a breakthrough for the organization and made a huge difference for that executive for sure.

Now we troop in with a colorful dashboard with indicators like a traffic light system and say “If the ball is red, then it’s a problem and you should do something”. You can imagine how executives react. Oh, they want the dashboard right now.

There are a few reasons why dashboards can’t be the first thing you deploy. There are some presumptions when we look at project data that may not be completely obvious just by looking at our picture.

First of all, we presume that the data has been approved or vetted in some way. We expect that there is someone who is prepared to stand up if asked and say they are responsible for any element of data we care to ask over.

Next ,we presume that the data is complete. When we look at a resource capacity planning report showing our project resource availability, we assume that all work is accounted for.

We also presume that the data actually does belong together. If we have a short list of projects, we assume that the data was collected and/or analyzed on or about the same time. We don’t expect that there will be 3 projects which were updated yesterday and the rest updated 6 months ago.

We also assume that the method of aggregating the data makes sense. When we look at summaries of the data, we don’t expect a 3 day project using 6 man-days to carry the same weight as a 6 month project using 1,000 man days.

Project data is, by its very nature a batch paradigm. Since every activity in the project can affect every other activity in this project and perhaps even in other projects, we are used to collecting data on a regular basis, perhaps weekly or monthly and then, after analysis and consideration, posting the results. Yet may executives want a “real-time-dashboard” as the first thing we deploy.

This happened to me recently while talking to a CIO at a large retail organization in the US. We were about to go in and start the deployment of an enterprise project management tool and on the eve our design meeting, I was told by the project manager that the CIO had asked that we make sure he got a dashboard as part of our first wave of deployment.

It fell to me to give him the bad news. Yes, he could have his dashboard but, no, it would not be useful until we could ensure that we had not only a process to gather the data in a consistent manner but also a tool where virtually all project data would be stored. He could have his dashboard in 6-9 months, I told him.

If management is asking you to make sure they get a dashboard so they are enabled to make decisions, then here are a few things that have to come first:

First, “all data in one place”. It’s like a mantra for our consultants. They know that if we can get all the project data into one source, using a single tool, then exciting tools like dashboards become possible. If the data can’t be brought together, then this is all wishful thinking.

Next, organize a central process to account for the data in question. There may be multiple sources of data to deal with. Aside from pure project data, financial data is often desired. You may need to account for operational workload such as technical support hours or emergency maintenance by people who would otherwise be available for project work. This has to be accounted for if you’re doing resource capacity reports.

When you get to the technical side of things, you’ll find a number of tools. Start by searching the Internet for Scorecard or “Balanced Scorecard” products. We’ve taken a liking around here to Microsoft’s Business Scorecard Manager 2005 but, whichever tool you choose, make sure it’s flexible enough to reach into multiple data sources to get the information you need for presentation.
Finally, manage management’s expectations. They may really need this kind of thing right now but if what you’re going to give them will lead them to the wrong decisions, it’s the most costly thing you could give them. Take your time and do it right.