
Everybody is talking about integration and collaboration and participation these days.
Down with the silos of yesterday. Away with the walls that prevent ideas and information from flowing through. Let’s all work together. Collaborate. Be like one.
It’s good in theory.
The hard truth is that there will always be a division of labor in an organization . If there aren’t any silos (or departments or units), everybody will interact with everybody.
Now think about that possibility for a moment.
Do you really want to receive a meeting invitation every time ANY unit in your organization decides to discuss an issue?
Do really want to coordinate with EVERY other unit before you respond to a request from a customer?
Do you really want to involve EVERYBODY ELSE in your project?
I assume you answered “no” to the above questions. So, on closer inspection, we do not want to be connected to everybody.
It would be impossible, even if we wanted to. Ten people can be connected to each other in 90 different ways (n (n-1))=10×9). For one thousand people, it’s nearly one million (1000 x 999).
That’s a lot of Outlook invitations to respond to.
Yet some large firms design their organizations as if this mathematical axiom didn’t exist.
Which is one reason why our inboxes are overflowing and most of us spend 70% or more of our time in meetings.
Now before I go on, let me stress that I am not saying that the intent is always wrong.
The key strategic rationale is often to achieve some sort of synergy, or develop and sell “solutions” (combinations of products and services from multiple units). For some firms in some industries, this makes perfect sense.But some firms also experience that customers simply don’t want solutions, but rather specialized offerings.
We also seem to forget that collaboration is costly. It takes time to set up and participate in meetings and to negotiate with people from other units who pursue different goals.
I would suggest an alternative approach.
The first is to confirm that there in fact is a potential for achieving synergy of some sort.
But assume that we confirm this.
Then I think the idea of a starting with a well defined mandate and setting up a managed team is a better alternative (something Andrew Campbell proposed in this interview I did with him).
It’s basically about creating a reporting structure to support coordination around a specific initiative, instead of issuing general calls for more collaboration.
(I would add one thing: Such efforts should be resourced. Since it takes time and money to collaborate, the money has to come from somewhere.)
In some firms that suffer from fragmentation or a lack of resource sharing, the real problem is not that people are unwilling to “collaborate” but that the current systems or incentives make it nearly impossible to combine or share resources.
In such cases, the solution is to remove these barriers. Only senior executives have the authority to do that, so they need to take the lead.
Finally, one may need to consider a re-design of the formal boundaries between units. A “light” solution would be to add an integrator role or managed team to coordinate across two or more units, but with a strong and frequent need for coordination, one may need to formally integrate the units (or parts of them).
It’s not a realistic goal to remove silos. But, like in the image above, we should connect the silos to each other, and install some windows, so people can see what’s going on outside and communicate with each other.
Photo credit: padraic woods via photopin cc
I think the IT world has solved some of the challenges you highlight but we have not applied then to org structures (yet 😄). Maybe we could collaborate to investigate further 😉
Paddy, sounds interesting – if you want to discuss this issue, feel free to contact me via email after the Christmas break (due to risk of spam, I don’t publish my email address here, but you can use the contact form on the blog if you don’t have it). Regards, Nicolahy
Nicolay,
I’ll get back to you via e-mail – probably next week. I’m definitely interested in understanding better current org design best practices and seeing how some of the practices from my area – enterprise architecture – might apply.
Regards, Paddy