Finally, there may be many reasons why construction companies innovate in one way or
another. The most obvious explanation seems to be the characteristics of the interaction both in
the internal network and the external network. The relatively weak internal couplings between
the project and the company level imply that exploratory behavior at either level is rarely
transferred to and exploited on the other level. Thus the feed-forward and feedback processes
(Crossan et al., 1999) are challenging. This is a well-known problem in project-based
organizations (Brady & Davies, 2004), and the companies in the study report that they put lots
of efforts and work more systematically on this issue. The study illustrates a greater recognition
of the role of external actors, particularly clients in developing and implementing new solutions.
However, what the construction companies must acknowledge is their own role as customers,
driving supplier development. Standardization and industrialization can imply either more
interaction or less interaction within the supply chain. On the one hand it can be about codeveloping solutions with producers and users, and on the other it can be standardized solutions
developed mainly internally which then need to be accepted by the rest of the network. Such
solutions, if they are accepted, do not require any deeper interaction. While the latter way of
working seems to have been a characterizing feature of the industry in the past we do see some
examples of a more integrated and interactive way of developing new standardized solutions,
even though it cannot yet be stated to be a representative feature of the industry. Another factor
influencing how the supply chain interacts is the strong focus on price in all parts of the chain.
The way that suppliers are usually selected does not encourage long-term relationships but
rather fortifies the uniqueness of the constellation of actors and resources in each project. The
character of the construction network, with intense interaction during separate projects but little
long-term interaction over several projects, enables explorative learning during the projects but
impedes exploration at other organizational levels, as well as exploitation of explorative
solutions at project level. In sum, one explanation for why construction companies innovate in
the way they do – the core of their innovation logic, may be found in the characteristics of the
interaction and connections in the internal and external network.
Referring back to the model presented in Figure 1, it is clear that this is an idealized image of
how the feed-forward and feedback processes should appear. We have shown that for these
processes to take place interaction between different actors, resources and activities, including
both the internal and external network of the single company, is needed. Thus, for innovation
to be achieved through such processes, interactions stretching across different network levels
and involving explorative and exploitative learning need to take place