Start writing without building a company first.
The hardest part of a new underwriting agency is not the underwriting. It is everything you have to buy, build and hire before the first policy earns a cent — and every dollar of it is spent at the moment you know least about whether the book will work.
Cuttleflow exists to move that cost off your starting line. The systems a new coverholder needs on day one are already built, and already reporting in the shape your capacity provider expects.
The problem
Breakeven arrives long after the spending does.
A new agency traditionally funds a rating tool, a policy administration system, document generation, a broker portal, taxes and levies logic, bordereau tooling and a compliance framework — before it has a book to spread any of it across. Then it hires one or two people to run the parts the software does not. Both are fixed costs, and fixed costs are the reason a small agency has to write a surprisingly large amount of premium before it makes money.
That is a structural problem, and no amount of budgeting fixes it. It decides which agencies get to exist: not the ones with the best underwriter, but the ones who can afford the year before the underwriting starts paying.
What changes
No system to build
Intake, rating, issuance, endorsements, taxes and levies, the audit ledger and the monthly report are already built. Your product is configuration on top of them — weeks, not a build.
Cost follows the book
A subscription scaled to what you carry, plus usage that follows the work. Your largest systems cost arrives as the premium does, rather than a year before it. Pricing →
Fewer people to run it
The month-end report builds itself from the records the policies were written on, so the back-office headcount most agencies hire for their first bordereau is not needed on day one.
The part capacity actually asks about
You will be asked how you control the pen — this is the answer, in writing.
A binder is granted on trust and renewed on evidence. For a new agency with no track record, the systems are the track record — and the diligence questions are always the same four. What your capacity provider will ask — and what you can show them →
“How do you stop someone writing outside authority?”
Authority limits are enforced at the moment of bind, not audited afterwards. Push a risk or a discount past an underwriter’s limit and the system stops offering to bind and starts offering to refer. You can demonstrate that in a meeting with the rating demonstrations.
“Can you reproduce a premium from two years ago?”
Rule packs are versioned and immutable once live. Any historic quote replays to the same number, and shows which rule version and which wording version applied on the day.
“What is your audit trail?”
A hash-chained ledger written before each state change takes effect. Nothing is overwritten; corrections reference what they correct. It is evidence rather than a report you assemble when asked.
“Will our bordereaux arrive clean and on time?”
Generated from the same records the policies were bound against, to the delegated-authority standard, with the Australian premium components carried as named fields from the day of binding. This is the question that decides whether a binder grows or is not renewed.
A five-person agency running this can answer those four questions in writing, the same way a much larger carrier would have to. That is not a claim about us; it is what happens when compliance is a by-product of the work instead of a department.
Small agencies weren’t a strategy — the cost of infrastructure just made them the only option.
What you still have to bring
We are not selling you a licence or a binder.
To be plain about the boundary: you bring the licence or authorised representative arrangement, the capacity relationship, the appetite and the underwriting judgement. We bring the machinery underneath, and the evidence that it behaved. Cuttleflow is not licensed to bind on its own account and does not place your capacity for you.
What we can do is make the systems question the easy part of your launch instead of the expensive one — and, separately, the Marketplace (proposed) is being designed so a new agency could be found by brokers on day one rather than spending its first year on distribution.