The hard part of repricing a fixed-fee client isn't the math
Since my first post about Reprice, I've been digging deeper into how bookkeeping firms actually handle fixed-fee clients.
One thing has become more interesting than I expected.
The problem isn't necessarily detecting that a client's workload has changed.
Some firms already track things like:
- transaction volume
- number of accounts
- new entities
- reconciliation workload
- account complexity
- changes in scope
That makes sense.
But then there's a harder question:
When has the change become significant enough to actually review the client's pricing?
For example, suppose a client's transaction volume increases.
Is one unusually busy month enough?
Probably not.
What if the higher volume continues for three months?
What if the client also adds two accounts?
What if the workload increase comes from additional services rather than transactions?
At some point, the situation has changed from:
"The client had a busy month."
to:
"The original pricing assumptions may no longer fit."
But where exactly do firms draw that line?
That's the part I'm trying to understand.
I'm particularly interested in how experienced bookkeepers handle this in practice.
Do you have:
- a numerical threshold?
- pricing bands?
- a baseline established when the client starts?
- a rule based on several months of data?
- scope-change triggers?
- or is it mostly professional judgment?
And once you've decided a review is justified, what happens next?
Do you normally:
- increase the fee,
- change the scope,
- move the client to another package,
- renegotiate the engagement,
- or sometimes decide the client is no longer worth keeping?
I'm still building Reprice around this problem, but I'm deliberately keeping it small rather than turning it into another SaaS platform.
At this stage, I'm more interested in understanding the real workflow than adding features.
For those who manage bookkeeping clients: what actually triggers a pricing review in your firm, and what do you do after that decision?
I'd especially like to hear about your actual process rather than what you think firms should do.
The trigger is the interesting part. In firms you’ve studied, is repricing governed by a repeatable rule, or does it still depend mostly on judgment?
From the conversations I've had so far, it looks like there are repeatable signals, but not necessarily a universal threshold.
Transaction volume, account count, added entities, reconciliation workload and scope changes come up repeatedly. The harder part seems to be deciding when those changes are significant enough to justify action.
So my current hypothesis is that the inputs can be fairly systematic, while the final trigger still involves some professional judgment.
I'm trying to get more data before treating that as a general rule.
You mentioned you're now seeing repeatable signals but are still trying to determine whether they support a general rule. That's a different point from where we left off 8 days ago. If that uncertainty is now becoming the harder part than collecting more observations, I think the Evidence Sprint may be relevant.
Yes, I think that's a fair distinction.
At this point, I'm less interested in collecting more examples just for the sake of having more examples, and more interested in seeing whether the same sequence keeps appearing across independent firms.
I'm curious about the Evidence Sprint you mentioned. What would the process look like specifically for this kind of question — distinguishing a repeatable pricing-review trigger from signals that are still mostly judgment-based?
Just replied by email with the details and how I’d apply the Sprint to the Reprice question you described.
This feels like the right problem to stay narrow on. The trigger is not really “transactions went up,” it is “we now have enough evidence that the original pricing assumption is no longer true.”
I reckon I’d separate temporary noise from structural change. One busy month should probably create a note. Two or three months plus a scope signal, like extra accounts, new entities, more reconciliations or more client questions, should create a review task.
The product may not need to decide the new price at first. It may just need to give the bookkeeper a clean case file: original baseline, what changed, how long it persisted, and which conversation to have next. That feels much easier to trust than an automatic “raise fee by X” recommendation.
I think that temporary vs. structural distinction is particularly useful.
A single busy month doesn't necessarily invalidate the original pricing assumption. Persistence plus a scope signal gives a much stronger reason to review it.
I also like the idea of separating the review decision from the new-price calculation. A clean case file with the original baseline, what changed, how long it persisted, and the relevant scope signals could be more useful than pretending there is always a correct automatic price increase.
I'm treating that as a hypothesis for now, though. I want to see whether other bookkeepers describe their process in a similar way before changing the product around it.
That sounds like the right level of caution.
I’d probably validate it by asking bookkeepers about the last client they actually repriced, not their ideal process. What was the first warning sign, how long did they wait, what finally made it feel justified, and what did they show the client?
That should tell you whether the “case file” idea matches reality. If they already mentally build a case before having the pricing conversation, Reprice can make that work visible and easier to defend.
The trust point matters too. A review prompt feels like help. An automatic price increase can feel like the software is overstepping before it has earned the right.
That's a very useful way to test it.
I agree that asking about the last client they actually repriced is probably more revealing than asking about their general process. The sequence you suggested — first warning sign, time waited, what made the review feel justified, and what they showed the client — should help separate the real trigger from the process they think they follow.
The “case file” question is especially interesting to me. If several bookkeepers already gather or reconstruct that evidence before a pricing conversation, then the opportunity may be less about calculating a new price and more about making that reasoning structured and visible.
I'll test that before changing the product. The trust distinction between “here's why this deserves a review” and “increase the fee by X” also seems worth validating directly with practitioners.