A project’s success or failure is judged by its adherence to the guidelines established when it began.
This is both typical and reasonable, but only if those guidelines are well constructed and attainable.
Often, an executive declares a project a failure based on the failure to operate within the allotted budget, failure to complete on time, or failure to produce a product of high enough quality. But such a judgment can be complicated. A project within its budget but failing to meet all the stakeholder’s ever-shifting needs may be considered a failure, while an ambitious project may go way over budget but still be considered a success because at least it was properly completed.
These subjective definitions of failure reveal that the measurement of a project’s success is not always about delivery, but about perception.
If the project guidelines can be better defined before it starts, it has a greater likelihood of achieving success.
Data can improve these expectations. Data about past projects can help you set appropriate, attainable goals, and inform you as to how much of the unexpected should actually be expected.
1. Planned Versus Actual Start/End Dates for Similar Projects
This data point requires that you’ve performed projects of this kind before. Simply collecting the actual start dates and actual end dates of previous projects (not the “projected” dates) will give you an average time-to-completion estimate, as well as high and low possibilities.
For further predictive accuracy, find the number of work days between those two dates, accounting for holidays, corporate training days, employee PTO vacation days, and unexpected sick days. Then find the number of problems faced, how long it took to address them, the number of adjustments made to the original plan, and time lost due to plan adjustments. Much of this data may not be included in the official report, so it may require digging through emails, records in your project management tool, meeting minutes, or questioning those involved.
Use this data to compare with the holidays, corporate training, PTO days, etc. that are planned to occur within the timespan of your upcoming project. Throw in the average unexpected setbacks that occurred in each previous project. This estimate, being based on previous experiences, has a higher likelihood of accuracy than a gut instinct, or even average project time. It may also be more convincing when attempting to explain why a project will simply take longer to complete than the executives would like.
2. Project Team Productivity Data
This data point does not require prior projects of its kind per se, but it does require that you have some sort of data on individual employee productivity for similar tasks to those required by the current project.
This requires a certain degree of willingness from your staff to either allow the monitoring - or participate in the monitoring - of how long it takes them to accomplish certain tasks. Ideally, this would be more than just a generalized, “It takes me two hours to do this.” Rather, it needs to be consistent, non-intrusive observations of the rate of task completion (such as noting each time an item is completed). Whether its measured in words per minute, lines of code per hour, or pots painted per day, this data can be used to create average productivity estimates, as well as minimum and maximum productivity levels.
Once you’ve identified a realistic rate of production, you can either use that information to calculate team progress manually or with the help of project management software. Many PM applications allow you to assign tasks to a team of workers. With productivity data for each team member already entered, the software will give you an estimated completion time for the entire project. This allows you to estimate time-to-completion of a team’s assignments based on of the cumulative productivity of each employee, with data to back it up.
One thing to be wary of is that many employees are uncomfortable with particularly detailed monitoring of their task progress. Assure them that the monitoring is purely for the sake of setting reasonable goals on future projects, which in turn will lead to higher success rates, and fewer project cancellations.
3. Resources and Materials Consumption Data
Material consumption is some of the easiest data to find accurate values for, and can save you serious money. Analyzing the relationship between the amount of materials anticipated, requested, used, and leftover from previous projects should help you better calculate the amount required for your upcoming project.
Similarly, quality assurance data on the frequency of defective products will help you anticipate the cost and delay of making up for imperfections.
Aside from the materials alone, entire budgets of prior projects can be broken down and analyzed to find which areas frequently overspend, and therefore which areas need to be better planned for. Whether that be a particular resource, labor, or tool, it’s likely to go over budget again, unless real changes were made.
All of this data collection is purely for the purpose of improving your project’s expectations beforehand. It’s the difference between simply picturing a finished bridge or piece of software in three months, and actually calculating out the time it will take for each team to perform their tasks based on numbers from their previous performances. Which method would you use to determine a project’s eventual success or failure?