I’m often surprised by what topics will spark interest in people, and cycles vs. innovation/disruption would not be one on my list of topics expected to interest people. But a couple asked me if I could elaborate my example a little more clearly, and so I’m going to go for a specific example currently facing my team.

We have a large branch, some 500+ people. Before the last round of cuts and reorganization, that number was closer to 700. Ten directorates dropped to 7, we moved a lot of financial processing people (back-office types) to a service delivery branch, etc. But the part I want to talk about is the regular financing files for non-salary costs.

These costs are not extensive, maybe 10-15% of salary costs, and include things like travel, hospitality, equipment, newspapers, water machines, software licenses, training, etc. A lot of small costs that require a bit of transaction time. During the reorg a few years ago, it was felt that there were economies of scale and increased consistency to be had by centralizing the macro entry of planning figures for finances by our finance branch. They left basic processing in the branch, and each directorate has an admin officer that handles that. They also coordinate the planning inputs from various divisions, roll them up, send them off to the finance branch, etc.

Now, however, the finance branch is looking at it as something that can be done directly by the same admin officers, particularly now that we have a new system with smaller margins for error. Still complicated, still painful systems to work, but less manual stuff in a way. So, with the fear of the finance branch that they are spending too much time on processing/planning, and not enough on strategic budget management, they have delegated all of that planning and coordination function back to branches. For every branch except mine. We’re the last hold out.

We will eventually have it too, but that leaves us with a question. Most of the other branches have full branch management services in their corporate areas — in other words, finance is centralized in their branch, with 4-6 finance officers handling most of the support function. For our branch, the finance branch would recommend the same — create a centralized system in our branch. Except I’m resisting that being the “obvious conclusion”.

The reason they are pushing it is that they are saying “the main problem with centralization in finance branch is that it is too time-consuming for their group, crowding out the strategic role they’re supposed to be playing” and the way to solve that is to move that centralized function to another centralized function, in each branch. Except those are not the only problems. They are also doing double-entry with some of the functions now — the “centralized group” is keeping two sets of books, and if truth be told, so are the individual admin officers. So almost three or four copies/sets of entries. If I simply accept the move from point A to point B, that function won’t change. And any “strategic” role that my team will play will get overwhelmed by the process, just as the finance branch people were.

Another problem is that the planning function gets lost in the transactional world, and things like salary forecasts in the system are decidedly out of date as soon as they are entered, and increasingly so until the next “period” update. What if I could find a way to get the admin officers to enter the non-salary data directly, to take responsibility for more regular updates on salary, and at the very least, reduce “triple or quadruple” entry down to “double” or even “single” window entry?

Equally, can I leave some of the coordination/strategic/support function that comes from the finance branch in my team? To decentralize as much as possible, or as much as proper, to make sure the forecasting and operations are as close to the actual delivery people as possible, but still retaining a bit of the challenge / strategic / coordination / support function in my team?

There are some problems that need to be addressed — multiple sets of data entry, transactional processes crowding out strategic management, lags in updates to some types of expenses, etc. But there are some good things too — decentralized for processing, centralized for strategic challenge function, centralized for support to make sure consistency in approach, etc.

Solving just the one problem (transactions > strategic) could force me into this long cycle of centralization / decentralization / centralization / decentralization. But what if I unpack the parts that work from the parts that don’t work, and see if I can keep some of what works at the centralized function and decentralize the rest?

Because what started this mess in the first place was the perception that decentralization was too expensive and inefficient and wasn’t working. Now, we’re now finding that centralization is inefficient and *it* isn’t working.

Sooooo, I’m trying to unpack to see if there is a hybrid solution that will stop the “yo-yo” from bouncing between the two extremes and allow me to keep the best of both worlds while minimizing the worst of either world. Yet without that conscious desire to stop the cycle, how do people inside the cycle see that it even is a cycle? Particularly when so many of the actors have changed and all they see is that the “current” isn’t working so we go to the opposite model instead of moving to the middle.

Or at least that is what is on my mind. Knowing full well that there is no single “right” answer.