Corporate projects are complex, time consuming, fraught with risk, and expensive. That's why our profession exists. Our job is to minimize every one of those dimensions. ''Delivered, complete, on time, within budget'' is the embodiment of success toward those goals.
Over the past 40 years, I have managed hundreds of projects for hundreds of companies. I wasn’t always successful. But failure is a great teacher. When Thomas Edison was asked how he invented the light bulb, he said, “I tried everything that didn’t work.” I don’t know if I have tried everything that doesn’t work, but I’ve done my share. I eventually arrived at a set of principles which dramatically increase the probability of success for any project. And they all involve issues to be addressed at the start, when forming the project team:
- Size
- Quality
- Responsibilities
- Environment
- Operation
- Communication
Quality: It is false economy to use nothing but smart 22-year-olds. You need some gray hair. To keep the team as small as possible, the team members must be highly qualified at what they do. Each and every team member will have a broad focus, so they all must have extensive experience and a deep understanding of their particular domains of expertise. They must have extensive and expert-level qualifications in every aspect of the job they are to perform. More than anything else, the success of the project depends on the levels of performance these individuals can bring to the team.
Responsibilities: There are five positions of responsibility. Only one person is required in each position, except for the project domain experts, which can number as many as four.
- Project manager: A natural leader who can remain focused on the goals of the project, communicate effectively with other team members, and keep things moving at a rapid pace with no delays. Expert-level qualifications in the domain of the project are also required.
- Business representative: Someone who understands the goals and objectives of senior management and has the authority to speak for them.
- Documenter: A journalist who can keep a log of all team decisions and create all required external team and project communications in language that anyone can understand.
- Quality control: Someone who is focused on the quality of the project or product to be created.
- Project domain experts: As many as four people responsible for producing the project content or result, such as programmers for software or engineers for product development.
Operation: The entire team will participate in short, frequent meetings. Every member will be free to express opinions and ideas regarding any aspect of the project. Most often, everyone will be in agreement. Sometimes, however, agreement will not be forthcoming. At that point the team will leave all final decisions on business-related aspects of the project to the business representative. He or she is in the best position to have an intuitive feel for the right business decision for the company. Final decisions on all project domain aspects of the project will be left to the project manager.
Communication: If the team is small enough and close-knit enough, then almost all communication can be verbal. Verbal communication is by far the most efficient and, when intended for the team only, is sufficient. The only items that must be committed to writing are project team decisions and communications intended for external distribution.
Limiting the total team size to eight enables efficient communication. This is a crucial element in the success of any corporate project. Larger teams are a great resource if you are digging a tunnel. In that case the more shovels the better. But in a complex corporate project where communication is critical, too many lines of communication dramatically slow the project.
As the number of people on a project increases linearly, the required lines of communication increase exponentially. The number of lines of communication is equal to n(n-1)/2, where n equals the number of people in the group. A group of eight, therefore, has 28 lines of communication, while a group of 120 has an astounding 7,140 lines of communication, making it 255 times more complex! No wonder the probability of success falls as team size grows.
Limiting the number of domain experts to four maximizes business value creation. This principle holds true no matter how large the intended output of the project. These are all qualified experts in their fields, all with the same general responsibility on the team. It may not seem logical, but including more than four will not appreciably improve the throughput or even the delivery date of the project.
In the late 1800s Maximilian Ringelmann, a French agricultural engineer, discovered that the greater the number of people pulling on a rope, the less effort each individual contributed. When there is a group of people, all performing the same task, there can be diminishing returns beyond just a few members.
In 1972 psychologist Ivan Steiner researched the effects of incrementally increasing the size of a team. Starting with one member and adding one at a time, he determined that each additional member increased the group’s potential productivity but also introduced inefficiencies: group members became less motivated, and group coordination became much more difficult. Steiner demonstrated that team productivity peaked at about five team members. After that point adding more members actually began to hurt the team’s performance.
For some current examples in today’s even more complex world, since 1978 Quantitative Software Management (QSM) has led the IT industry in software estimation, project tracking and control, and software metrics analysis. In a study published in 2005, they used metrics from more than 7,000 completed projects to analyze productivity and error rates. The results were astounding but predictable if we are to trust my principles:
- A two-person team developed 40,000 lines of code in 40 person months.
- A 29-person team only improved the time frame by 12 calendar days.
- The larger team produced six times as many defects as the small one.
- Teams with 34 people on average completed a 100,000-line project in 5.6 months at a cost of $2.1 million.
- Teams of four people on average took about two weeks longer but cost just $294,000.
- Shaving two weeks off the schedule cost some companies as much as $1.84 million.
As of 2001 the International Software Benchmarking Standards Group valued one function point at about $1,000. This places an approximate value of the solution we developed at $60 million. It was created over an 18-month time frame. That equates to a business value creation rate of $500,000 per person year, an incredible level of productivity.
Books, articles, papers, and studies have been published for decades, all coming to the same conclusion. More resources in the way of people applied to corporate projects bring a greater decline in incremental productivity and quality. It does seem logical that for large corporate projects a large team is required, but that is absolutely not the case.
The exact optimum number is debatable — even the experts cannot agree. But they do agree it is not big. According to Fortune magazine in 2006, it was 4.6. According to Jeff Bezos, CEO of Amazon, it is the number of people who can be fed with just two pizzas. But for me eight is enough.
About the Author
With over four decades of experience, Art Pennington is president of the Profit Research Institute, the founder of four successful software companies, an author, a keynote speaker, a consultant, a holder of multiple patents, and the creator of the “profit method” of business success. Art’s free book, PROFIT, is available at www.profitmethod.com.
