The common version
A consultant runs workshops, writes down what people say they need, marks each line as fit, gap or partial, and hands over a spreadsheet with several hundred rows. It looks thorough. It is nearly useless, because it tells you nothing about what any of it costs or which items actually matter.
Worse, it captures the process as described, which is rarely the process as run. The spreadsheet documents an idealised version of the business that nobody actually follows.
What good looks like
A fit-gap that earns its fee produces four things, and you should ask for them by name.
- The current state as it is actually run, workarounds included. If the workarounds are not in the document, the analysis is incomplete, because the workarounds are the requirements.
- A gap register where every gap carries an estimate and a recommended treatment: configure, extend, change the process, or accept it.
- A phased scope, with a stated reason for the sequence and an explicit list of what is not in phase one.
- The decisions that need a sponsor, surfaced early, with the cost of each option.
The workaround test
There is a quick way to tell whether an analysis is real. Find someone in finance or operations who has been there five years and ask them what they do outside the system. Then check whether that appears in the document.
It usually will not. Those workarounds exist because the system does not do something the business needs, and they are the highest-value requirements you have. An analysis that misses them will produce a configuration that gets worked around in exactly the same way.
Who should own the output
You should, unconditionally, whether or not the firm that produced it goes on to deliver. A fit-gap held hostage to the implementation contract is a sales instrument.
It should also be readable by a finance director, not only by a consultant. If it cannot be handed to your sponsor and understood without translation, it will not be used to make the decision it was commissioned for.