David Anderson has been doing a lot of TOC thinking as relates to the software development process. Drumming in the Dark talks about what happens when the constraint isn't obvious:
I get asked often, "Where do you start?" Where do you start, if you want to make improvements and follow the ideas in the Theory of Constraints, when you don't know where the constraint is, or how to identify it?
David's focus is software development, but this question and the first-pass solution is the same other project management environments. It even works in manufacturing, which is where TOC (and drum-buffer-rope) got their start:
The answer is pretty simple. If you have no worthwhile measurements and no real insight to your ... productivity and performance, you simply drum at the rate of the output. As a batch of customer valued deliverables appear at the output, you allow a similar sized batch in at the input. This works regardless of your unit of value measure.
The answer is deceptively simple. People are always tempted to load up the system so that everyone is always busy all the time. But the only thing that results from that is piles of work-in-progress and no progress on the work that actually makes you money.
The theory of constraints is referred to as a process of ongoing improvement (POOGI), and this idea of setting a "simple" drum to the rate of output is a beautiful example. By following the process, one quickly discovers if there is an internal constraint. If there is, POOGI guides what to do once that constraint is unveiled: set the drum based on the constraint and/or strategically decide where the constraint should be and change the capacities elsewhere to be in line with that decision.
If you are curious about David's specific example, have a look at his presentation on the XIT SE Example, which he presented at the TOCICO conference in November (all presentations available online).