Learning from the sewers

The Fowl Meadow today

The basic stages of building infrastructure--plan, design, build, and maintain--apply equally well to information systems and physical infrastructure like pipes, roads, and power systems. Over the years, the science of project management has developed many tools to help such projects stay on schedule and on budget, yet many projects fail to meet their timelines and their financial milestones. Why a project fails is often viewed as a sui generis combination of internal management and technical glitches and external forces. But there is often one common factor: A failure of the organization's leadership to adopt a supervisory approach that helps keep the project on track. Oddly enough, this can occur even when the leader of the organization is strongly committed to its success.

If you read through the Washington Post's summary of the path taken to construct the federal health care insurance exchange system, you can find lots of explanations for the project's failure.  Some are internal to the agencies involved, while others may have derive from intense political opposition or fear of that opposition.  Nonetheless, it seems incredible that such could have occurred when the President's commitment to the program was unquestioned, and when the stakes were so high for him personally and for the country.

I'd like to suggest that one reason was that the President simply did not understand how to supervise this kind of project in a manner that would identify implementation problems in real time and would help ensure that they were dealt with expeditiously and effectively.  I am not saying this critically.  There is no point in piling on.  I am saying it because the experience offers lessons for future leaders, whether at the federal or state and local level.

The solution is remarkably easy and is illustrated by a town official and a sewer project. When I was executive director of the MA Water Resources Authority, we had literally dozens of water and sewer construction projects in our, er, pipeline.  One project was designed to repair structural and hydraulic deficiencies in an old sewer line that were causing sewage overflows into a wildlife area called the Fowl Meadow in the towns of Canton, Norwood and Milton, which also threatened the water supplies for Canton, Dedham, and Westwood.  The project was complicated, involving tons of permitting issues because we were building through a wetland, because of the overlapping municipal jurisdictions, and because of numerous technical issues.  The project would ultimately comprise 17,000 feet of 48-inch diameter force main sewer line, a 50 million gallon per day pumping station, and 10,000 feet of 48-inch diameter gravity sewer.  Physical challenges included five pipe jacking crossings at major highways and railroads, a siphon crossing of the Neponset River, a pumping station foundation, controlled blasting removal of bedrock, dewatering of silts and fine sands, and sewer bedding and backfill. In addition, it was necessary to protect the unreinforced 48-inch pipe immediately adjacent to the old sewer line throughout the construction period.

At that time, the town manager from Norwood, John Carroll, served as one of the MWRA board members.  John had been commissioner of public works for Massachusetts and had years of experience in managing construction projects there and in his own town. Of all the dozens of projects facing the MWRA, this one was John's pet. His goal was to stop the disgusting overflows of sewerage which were wreaking environmental damage in his town. He understood the long-lived nature of such projects and wanted to implement a supervisory approach that would keep the project on track. He asked me if I would please arrange a status meeting between him and the project team every six weeks. He said, "I just want them to tell me how it's going, what problems they are facing, and how they are planning to solve them."

John knew that sometime before the team met with him--maybe around week four--they would start preparing for their next session.  They never wanted to come to the meeting unprepared. If they had encountered a problem, they wanted to show up with an idea as to how they would solve it.  It's not that John was trying to impose his judgment on the team. He never did that. He would just ask simple questions. He was always supportive of the team, encouraging them to tell him everything.  Knowing that he would ask those questions was not only enough to keep the team on track, but also made them feel comfortable telling John about their problems. They knew he was an advocate for them and could use his involvement more broadly. In particular, John's well-known interest in the project gave the team the status it would need to acquire support from other parts of the agency. After all, no project team has all the answers. There is always a need to get help from other parts of the organization, but there has to be a way to encourage those other divisions to cooperate, to take time away from other projects they might be working on.  The result: The multi-year, multi-million dollar project stayed on track, helping to enhance the entire community. You see the result in the pictures here.

Now, the WaPo article makes clear that President Obama had regular status meetings with his subordinates:

For months beginning last spring, the president emphasized the exchange’s central importance during regular staff meetings to monitor progress. No matter which aspects of the sprawling law had been that day’s focus, the official said, Obama invariably ended the meeting the same way: “All of that is well and good, but if the Web site doesn’t work, nothing else matters.”

I wasn't there, but I'd like to hypothesize that the President's approach was not the same as John's. For one, he probably had the wrong people in the room. I bet he didn't have the actual project team. For another, I bet he did not portray an atmosphere of support. More likely, he demanded results--“We’ve got to do it right"--and he comes across to me as a person who is not interested in excuses. I'm guessing that people would not have wanted to give him bad news.  Hence the October surprise:

Only during the weekend after HealthCare.gov’s Oct. 1 opening did the president’s aides begin to grasp the gravity of the problems, the White House official said. Obama soon began getting nightly updates on the performance of the Web site, which has still been unavailable to Americans for hours at a stretch over the past week.

Too late, as it turned out.

I don't mean to suggest that a sewer project is the same as a massive information system, but there are similarities and lessons to be learned.  I've seen John Carroll's approach used in many organizations that faced high priority complicated physical and information system infrastructure projects, including those loaded with technical and political challenges (and bigger than HealthCare.gov.) The simplicity of the approach is deceptive: This is a powerful management technique. The leader's expectation of high performance by the project team is fulfilled not by appearing to hold them accountable or by making the technical decisions for the team. Instead, the leader appears as an interested and engaged project advocate, curious about obstacles and successes, empathetic to the concerns of the team. Information is shared in real time throughout the pendency of the long-lived project. The team thereby holds itself accountable to a high standard of performance. Success emerges.