A standard product is never finished.
That might sound contradictory. Isn't standardising supposed to lock things in place?
But the best CPQ projects I see don't treat their standard as an end point. They treat it as something that keeps moving.
In practice, I see this happen mostly with models that have already been live for a while. Not right after a configurator is built, but once a model has been in use for some time and the market has shifted in the meantime. New opportunities emerge, and that's the moment the standard needs to expand if it's going to stay relevant.
A customer asks for something that isn't in the standard yet. Rather than rejecting it outright, or agreeing to it blindly without thinking it through, it gets deliberately sold as a special solution. From that sale, engineering gets the job of actually working it out.
Nothing new so far — that's simply customisation.
But here's where it gets interesting. If the same request comes in again, and again, that's a signal. And that signal doesn't always come from a customer explicitly asking for something. More often, it's the market shifting — and if you don't respond to it, your competitor will. That's when the solution, which started out as an exception, gets added to the standard.
"That's how your standard grows in step with what the market actually asks for. Not with what you assumed was needed beforehand."
That does require discipline. You need to actively track this pattern: which exceptions keep coming back, through which channels, and how often. Without insight into the data your configurator processes every day, it stays a matter of chance which exceptions ever make it into the standard, and which simply remain exceptions while the market has long since moved on.
Do you have a way of structurally translating market shifts into new standard options, or does this mostly happen incidentally?