Tracking Projects

by Carl Angotti, Managing Director

This paper discusses the need for methods to keep track of Project Progress by measuring the project plan against what is actually happening as the plan unfolds. This is one of the most tricky, and often underused part of the Project Manager's tools.

The State of NPD Project Visibility

If ever there was a “Emperor's New Clothes” phenomenon in New Product Development (NPD) Project Management it is the lack of the use of viable measurements to show project status in meaningful ways to both the team and upper management. We have seen this effect many times, as have many of our associates. Project status often turns out to be obscured in the presentation of actual scope, cost and schedule compared to the original plan during the often long implementation phase. This often leads to poor management and team decisions plus anger and frustration all around.

The Cat-and-Mouse Game

The upper management and the project team seem to play a “cat and mouse” game with each other when the major project variables are set and reviewed. At the beginning of the project, management sometimes doesn't even allow the team to provide any input into the schedule or cost for a project. Or, if the team creates a plan for schedule and cost, then management sometimes ignores these estimates when setting project goals. Upper managers frequently have the tendency to require shorter schedules and lower costs than the team believes are possible, while the project team tends to overestimate these variables.

In addition, the management often has a hidden “fudge” factor for cost and schedule. They then require the team to meet a short (compared to the “fudge adjusted,” “very tight,” schedule, that the team knows full well is impractical. They both know the schedule will slip, but the desire for management is to get the most work out in the shortest time, and, at what they believe to be the lowest cost. This causes the team to react by “hiding” the actual progress from critical review. “… to expect a project to be accurately predicted at the beginning almost borders a belief in “magic.”After all, who wants to constantly report “bad news.” Management then tries to get the “real” status, but it frequently seems elusive. Meanwhile, the Cat and Mouse game is underway. This approach could actually result in higher costs and longer schedules than those originally estimated by the team.

A Natural Tension

There is a natural tension between the goals of both of management and the team. It would be ineffective for management to blindly accept the estimates of the team, and it is just as ineffective for the management to set the schedule and associated costs by fiat.

New Product Projects Have Lots of Risks

The real problem is that NPD projects have lots of risks associated with them, and to expect a project to be accurately predicted at the beginning almost borders a belief in “magic.” On the other hand, we believe that much more can be done to set project measurements, both direct and indirect to improve the management of projects.

What Can Be Done?

Project Managers and management need to work together to stop the cat-and-mouse cycle. This requires the acceptance of the risks involved, and respect for each others' position in the overall development function. This means creating an environment of working together with management as part of the team, not adversarial to it. We believe the best information should be presented to all of the parties, regardless of the fact that rigid goals are not being met. In our opinion, the best decisions are made when the best information is available.

In addition, effective, high quality, measurements need to be presented to both the management and the team, so they can focus on getting the project completed in scope, and with minimal budget and time used.

Beyond MS Project Basics

Many project managers consider it sufficient to just “update” the schedule during implementation. “Gantt charts… are helpful, but often not adequate for tracking real NPD projects effectively.” This is used to show status via a new schedule, but the original, or baseline, schedule is not presented alongside. If this were done, it would show an almost constant schedule slip.

Measurements must move well beyond just using basic Microsoft Project Gantt charts and a spreadsheet of project milestones, due dates, actual dates and percentage complete. These are helpful, but often not adequate for tracking real NPD projects effectively.

Use Earned Value to Track Project Progress

One straightforward method to track projects is to use the MS Project built in “Earned Value” feature. This can produce a data table that can be exported to an Excel spreadsheet. The data in the spreadsheet can then used to produce a plot of estimated versus actual accumulated project progress. This feature is rarely used, but it can be very effective, since it is very visual. One could even use some fancy math in Excel to improve the ability to do a much more valid estimate to complete.

What it does is to allow the Project Manager to show the accumulated resource cost “earned” by the completion of tasks, as the project progresses. This is plotted along side what was originally estimated. As each task completes, it “earns” its value as the percentage completion increases to one hundred percent. Hence, both expended resources and completion are shown together. Such an approach produces the famous “S” curve of completion for a project, but in this case, the curve is weighted by the value of the resources used. This method shows progress in a way that no milestone chart or Gantt chart could ever do. It provides a “dead reckoning” look at where the project is, compared to where it was estimated to be.

Other Helpful Indirect Graphs

To make projects still more visibly effective, a few other indirect charts are needed. If this isn't done, it might be possible to “sub-optimize” the project by over focusing on the primary variables. Such graphs create a number of measures that report underlying problems that might not otherwise be detected. These could include measures such as:

1. A plot of estimated versus actual accumulated project cost over time.
2. A plot of Open, and Closed Action Items or “Bugs” versus time. This plot would have the Action Items weighted by priority, to give more emphasis to more critical items.
3. The Risk Management List of Risks and Mitigations, with the number of Risks and Number of Mitigated Risks being plotted together over time. Again, the risks would be weighted by their impact as they are plotted. This would emphasize the status of the more impactful risks.
4. Other Plots tuned to the special needs of individual projects.

A Dramatic Improvement in Visibility

The approaches described here require somewhat more work than just producing a Gantt chart and milestone list, but the results can be astonishing. The combination of the graphical plots shows intuitively how the overall project is doing. It is then possible to see trends and estimate future milestones much more accurately.

Use “Quasi-earned Value” to Track Project Progress

As mentioned above, one straightforward method to track projects is to use the MS Project to create Reports suitable for this “Earned Value” feature. Unfortunately, this information is only available in the Reports feature of MS Project. “ACWP [and] BCWP [plots] give a powerful look into the project progress”The data must be manually transferred from the Reports to a spreadsheet. Generating this information is not straightforward and easy. But the work is worth the effort, if one wishes to make project status more visible. The report data is entered into a spreadsheet that can then used to produce a plot of estimated versus actual accumulated project progress and cost.

Throw Away Preconceived Notions of Earned Value

To make this method work, one has to throw out any preconceived notion of “Earned Value” and any relationship it bears to the actual financials on a project. We are looking here for a method of obtaining a special plot in the easiest manner possible.

What we want is to allow the Project Manager to show the accumulated resource cost “earned” by the completion of tasks, as the project progresses. This is plotted along side what was originally estimated. As each task completes, it “earns” its value as the percentage completion increases towards one hundred percent. Hence, both “expended” resources towards completion and the original plan are shown together plotted over time.

Such an approach normally produces the famous “S” curve of completion for a project, but in this case, the curve is weighted by the value of the resources used. An earlier approach that was used had the completion of tasks marked off and then accumulated. This weighted all tasks equally instead by the size of the resources used. The “Quasi-Earned Value” method shows progress in a way that no milestone chart or Gantt chart could ever do. It provides a “dead reckoning” look at where the project is, compared to where it was originally estimated to be.

What does the Graph Look Like?

The result looks like this,

Note that the graph shows three plots marked as follows,

• BCWS—Budgeted Cost of Work Scheduled (in blue)
• BCWP—Budgeted Cost of Work Performed (in red)
• ACWP—Actual Cost of Work Performed (in green)

These are Earned Value terms.

Each of these graphs shows an accumulated number, plotted as a function of time. The BCWS is the accumulated value of the work that was originally scheduled. Hence, it can be plotted at the beginning of the project. It is derived from a crosstab custom report of cumulative cost of each task versus each 1/3 month period for the project.

This is the baseline that is placed into a spreadsheet at the beginning of the project.

The BCWP plot is the accumulated value of the “work performed” as a function of time. It is placed in the BCWS spreadsheet, and is created and updated each week during the project. It is derived from a custom report that shows the BCWP by week. This is also taken from a custom crosstab report that can be created each week during the project. It is then accumulated in the spreadsheet to create the data for the graph. The data changes as the percentage completed for each project is updated each week. This is the “Earned Value” for the project. Tasks that are not complete definitely show up in the graph as a schedule slip.

The ACWP plot is the accumulated value of the ACWP for the project and is derived in a similar manner to the BCWP, but for ACWP. This is a plot of the actual cost of the work performed, accumulated by week. This plot shows how the project actual (scheduled) cost is doing compared to what was originally estimated. Both of these give a powerful look into the project progress.

What does this Graph Indicate?

The graph illustrated does not have the usual “S” curve, because the project is a “repeatable” one with no need for much “front end” work. This is unlike most Product Development projects. It shows that the project, near the end of September, is about two weeks behind schedule for all tasks being tracked. It is also spending a bit more than planned, about $5,000 more out of about $30K. This is true without regard to what the Gantt chart Critical Path indicates. These are not good trends, and may require corrective action.

Using “Quasi-critical Chaing” to Track Project Progress

Critical Chain Style project management has some interesting concepts that are useful in managing projects. Although we are not “true believers” in the Critical Chain approach, our group and others think it has some merits worth looking at.

One important concept of Critical Chain is the idea of having a schedule buffer built into a project at several points, but especially at the end. These buffers allow for a possibility of having a high probability of creating a project that finishes “on time.”

More about the Critical Chain Concept

We have two papers on our website, by two PM associates that cover the idea of Critical Chain in greater depth than is presented here.

The first is from a lecture by Dr. Dohn Kissinger, of Profit Solutions, called “Faster, Better, Cheaper.” In addition to the lecture slides, you can read my notes from the lecture. Dr. Kissinger is a certified Goldratt Institute trainer.

The second paper is “Critical Chain as an Extension to CPM” by Vincent McGevna, PMP. This paper describes some practical ways to use Critical Chain together with Microsoft Project, and without any special “add on” programs.

In both instances, for visibility, the focus is on the buffer size to be used to create a practical contingency.

Using Critical Chain Method for Visibility

To create visibility in a project, the Project Manager can show other managers and the project sponsor a plot of the Critical Chain buffer size plotted as a percentage of the buffer size used over time, starting from the project beginning and continuing throughout the project. This plot shows the actual amount, and the rate of consumption of the buffer on the project.

From a project management point of view, the standard way to use this approach is to consider the point at which the buffer size reaches 50% of the available buffer to be very significant. This can be assumed to be a major trigger point for a complete project plan update. If the methods outlined in the Kissinger and/or McGevna presentations are used, the buffer should be about equal to 1.0 to about 2.0 times of the Root Sum Square (RSS) of the difference between the nominal and most pessimistic estimates to complete for the tasks along the CRITICAL PATH. This becomes the buffer that is then added to the end of the project plan and then tracked via the plot.

Give these a Try

The approaches described here are not normally used by project managers. And, except for the up front work, they take very little effort to help track a project and provide very valuable information to help with management decisions.

Questions

The preceding description of tracking tools is a very short presentation of how these methods might work on a real project. If you want to have a more in depth understanding of the approaches presented, please contact us at cangotti@360strategicpartners.com.

This paper and the material it links to are published with permission of the authors.  All rights to these materials remain with the authors.