The theme of this week is “the balancing act”.
Several papers you read this week are about seemingly opposing things. The article “From firefighter to fire lighter” by Barber and Warn puts reactive against proactive behavior. The Chapter of Winch, managing construction projects consummately tries to bring all kinds of opposing things together. Winch puts construction as artifacts against construction as assets; he sets project integrity to process integrity. The paper by Koppenjan et al. distinguishes different management approaches: predict & control versus prepare & commit. Interestingly he brings the two also together by stating that the approaches are “seemingly opposing”.
Winch states that the central tenet of his book is that construction project management is a problem of management of uncertainty through time. The solution to deal with this problem is to use a variety of seemingly opposing “tools”. To be successful as a manager and to be able to steer a project to a conclusion a large variety of skills, methods, approaches, strategies, attitudes, and performance are at your disposal and will be needed…… and they change through time as the project moves forward.
For me, managing is taking a position on the scale between proactive and reactive, between being strict and flexible, between predict & control and prepare & commit, between contact and the people, between anger and joy, between directive and cooperative, between wearing a suit or wearing jeans and sneakers.
I would like to give an example.
Some years ago, I worked as a manager for the refit of the monumental Vondelpark Pavilion; I referred to the project in my post of week 4. The building used to be the Dutch Film Museum for years and needed a thorough refit when the museum left to move to the EYE building in the North of Amsterdam. The owner of the building is the Amsterdam municipality and they found new tenants, a broadcasting corporation; AVRO, and a grand café-restaurant. My team needed to take care of all the user requirements. That is comprehensive for broadcasting studios, internet studios, and a large restaurant with all the equipment and technical stuff needed. In the meantime, we needed to take the monumental status of the building into account.
When I started the project, I thought that it was just a straightforward restoration and refit job. The reason I started to compose a design team and value the work in a kind of technical and transactional way. One of the things we made was a schedule. The client, the Alderman of the city district, did not want to rush things. He wanted a solid project structure and wanted us to take care of the quality of the monument and also diminish risks, especially related to budget and time. A realistic timeline and carefulness were important. The client stated that if the schedule was realistic the chances of delay are lower than with an overly optimistic schedule. The project would take about two and half years to design, construct and complete. For the owner, the project was partly an asset to rent out. The Alderman also wanted a smooth process because he had bad experiences with another project with public and political indignation about budget overruns and delays.
That was an error of judgment related to the tenant. For the future tenant, the building was a secondary thing, an artifact, and speed was everything. The prime concern for the AVRO director was to have an excellent broadcasting studio for television, radio, and the internet as soon as possible. It became clear to me that the AVRO, existing for almost 100 years, wanted to celebrate its anniversary a year later in the new building. That was, given the time schedule impossible.
A week after my official assignment, I met the director of the AVRO and members of the board and we ended the meeting with a disagreement. The schedule destroyed their dream of a great anniversary in the Vondelparkpavilion. The director was so angry that she phoned the Alderman to complain about my inflexible stand and she demanded to speed up.
Immediately I understood that the project, or at least part of the project, was not technical at all, and trying to convince the director by explaining the schedule and the complexity of the project would only make things worse. The project was a big personal thing and of high symbolic value for the user. I needed to change my approach completely. Instead of organizing a “technical thing” I needed to design a plan of collaboration, of mutual sensemaking, and approach the tenant in a transformational way. I needed to gain trust, and I needed to find solutions to solve this incompatible situation of more than a year. I was in an uncomfortable position and I needed to solve that.
My solution was twofold and the process was a balancing act. I assigned and contracted a design team to speed up the process of design and to move forward with the preparations for the refit. In the meantime, I dedicated most of my energy to the dialogue with the new tenant. The tenant hired a lawyer and a management consultant firm with the aim of forcing me to speed up delivery. In months thereafter it became clear to their consultants that it was impossible to deliver the project in one year. Especially because we had to deal with a state monument in the city center and it turned out the tenant had extra requirements to incorporate in the design. They understood that my schedule was realistic and we informed the AVRO director about this. She trusted her consultants and changed the anniversary plans.
Now I needed to deliver according to my plans. The focus of the design team was to follow the schedule as well as possible, and at the same time to solve designery constraints mentioned above. I needed this to gain a little trust and to show I was in control, honoring my promises and serving the needs of the AVRO. I needed to steer this process proactively.
Building a good relationship with the tenant (AVRO) was more about “prepare and commit” and I needed to be open and clear. The approach for this part of the project was also proactive and on time because I needed conformity before I could contract a builder.
In the end, the project was delivered after almost two years, faster than promised. We had multiple constraints along the way, complaining and protesting neighbors, challenges with the condition of the building, several discussions with the historic department, and many difficulties with the integration of the broadcasting equipment in the monument.
After delivery, it took the AVRO several months to test and try-outs before the first broadcasts.
Luckily, I had a good contractor.
Barber and Warn (2005). Leadership in project management: from firefighter to firelighter.
Koppenjan, J., Veeneman, W., van der Voort, H., ten Heuvelhof, E., & Leijten, M. (2011). Competing management approaches in large engineering projects: The Dutch RandstadRail project. International journal of project management, 29(6), 740-750.
Walker, D. H. (2015). Playing the Project Manager. International Journal of Managing Projects in Business, 8(2), 393-400.
Winch, G. M. (2010). Conclusions: Managing Construction Projects Consummately. In: Managing Construction Projects; An Information Processing Approach. West Sussex, UK: John Wiley & Sons