Search for financial modelling services and you will find a great deal about methodology and almost nothing about price. Every provider asks you to book a call. The reason is straightforward: pricing is easier to defend in a conversation than on a page, and nobody wants to be the cheapest or the most expensive line in a comparison.
We think that is the wrong trade. You cannot budget for something you cannot price, and making a buyer sit through a discovery call to learn whether a service is a $3,000 decision or a $30,000 one wastes their time and ours.
So here are our numbers, what drives them, and where we think you should spend less.
What we charge
Financial model build: from $5,000, typically $5,000 to $20,000. Two to four weeks.
Model review and remediation: $2,000 to $5,000. Three to six business days.
Fixed fee in both cases, quoted after a scoping call, agreed before any work starts. We do not bill hourly for model builds. Hourly billing rewards the provider for taking longer, which is not an incentive you want pointed at your deadline.
What moves a build within that range
Five things, roughly in order of impact.
Number of entities and currencies. A single-entity model in one currency is straightforward. A group with four subsidiaries across three currencies needs consolidation logic, intercompany elimination and translation mechanics, and each of those is somewhere an error can hide.
Whether historical data needs reconstructing. If you have clean, reconciled management accounts for three years, we build from them. If the historicals live across several systems, disagree with each other, or were never properly closed, reconstructing them is real work and it happens before modelling begins. This is the single commonest reason a quote comes in higher than a client expected.
Debt structure. A term loan is simple. Multiple tranches with different tenors, margins, amortisation profiles and ranking, with covenant testing and a cash waterfall, is a different exercise. Project finance and acquisition models sit at the upper end of our range for this reason.
Number of scenarios and the depth of sensitivity work. Three switchable cases with a handful of sensitivities is standard. Twenty scenarios with probability weighting is not.
What the model has to survive. A model for internal planning and a model that will be pulled apart by a lender's credit committee are built to different tolerances. The second takes longer because more of the time goes into making it defensible rather than making it work.
What you get at each end of the range
At the lower end, around $5,000, expect a single-entity three-statement model, driver-based, monthly then annual, with a clean input sheet, a scenario switch, basic sensitivities, and documentation. That is a genuinely useful model for a company raising a seed or Series A round, or planning internally.
At the upper end, around $20,000, expect multi-entity consolidation, multi-currency, a full debt structure with covenant testing, a cash waterfall, extensive scenario and stress testing, a valuation bridge, and output formatted for an external audience who will challenge it.
Above that range we would normally tell you the scope has changed into something else, usually a transaction advisory engagement rather than a modelling one, and price it separately.
When you should spend less than we charge
This is the part most providers leave out.
If you need a forecast, not a model, and the business is simple, a competent finance person with a spreadsheet and a week will serve you better than an external engagement. Not every planning exercise needs a built model.
If you are pre-revenue and pre-team, and the model exists mainly because an accelerator asked for one, a good template will do. You will rebuild it within a year anyway, because the drivers you assume today will not be the drivers you actually have.
If you already have a model and the question is whether to trust it, do not commission a rebuild. Get it reviewed first. A review costs a fraction of a build and frequently concludes that the model is sound and needs extending rather than replacing. We would rather sell you the cheaper thing and be right about it.
If the only purpose is to satisfy a checklist, and nobody will interrogate the output, buy the cheapest thing that clears the checklist. Spending advisory fees on a document nobody reads is waste, however good the document.
What drives cost at the large firms
We are not going to publish anyone else's fees. What we can say is directional and uncontroversial: engaging a Big 4 firm for equivalent modelling work will cost you several multiples of the numbers above, and the difference is mostly structural rather than a quality premium. Larger firms carry a cost base that mid-market fees cannot support, so the economics only work at a certain ticket size.
There are cases where you should pay it anyway. If you need a recognised name on the cover for a regulator, a court, or a public-market transaction, the name is the product and you should buy it. For everything else, the question is whether you are paying for capability or for the logo.
Red flags in how a model is priced
An hourly rate with no cap. Your interests and the provider's are opposed from the first hour.
A quote given before anyone has seen your data. Nobody can price this accurately without knowing whether your historicals are clean. A quote without that question having been asked is a number that will change later.
A price far below the range above. Financial modelling is time-intensive skilled work. A $500 model is a template with your logo on it, and it will be identified as one by the first person who opens it properly.
Unclear ownership. If you do not get the live file, with unlocked cells and no ongoing licence, you have rented a model rather than bought one. Ask explicitly.
No revision period. Your assumptions will change during a raise or a credit process. A provider who treats every change as a new engagement has misunderstood what the model is for.
How to get an accurate quote quickly
Four things, and any provider who cannot price from them is guessing:
- ·What decision the model supports, and who will read it. A lender, an investment committee and an internal board want different things.
- ·The structure. How many entities, which currencies, what debt exists or is contemplated.
- ·The state of your historicals. Be honest here. "Three years of reconciled management accounts" and "we have the bank statements and a lot of invoices" produce very different quotes, and the second is not a problem as long as it is known upfront.
- ·The deadline, and whether it is real. A term sheet date is a constraint. An internal preference is not.
With those four, we can usually quote a fixed fee within a working day.
The underlying point
The cost of a financial model is almost never the expensive part of the decision it supports. The expensive part is raising at the wrong valuation because the model overstated cash, or committing capital on a projection that did not survive contact with reality, or losing credibility in diligence over an error that a review would have caught in an afternoon.
Price the model against that, not against the invoice.
Working on a raise, a refinancing or an acquisition? We quote a fixed fee after a scoping call, at no charge. If your requirement is simpler than the range above, we will tell you that too.