The ceiling wasn't demand. It was proof.

There is a specific moment that many growing MGAs will recognise, and it rarely appears in pitch decks. It is not the moment a new syndicate expresses interest. It is the moment right after — when someone on the leadership team has to answer, honestly, whether the organisation could actually run what it is about to be trusted with.

For one mid-market Lloyd's coverholder, that moment kept recurring. The business had built something real: several specialist commercial binders, distributed through a substantial network of agents, with genuine appetite from the market for more. Syndicates were interested in extending capacity. New classes of business were on the table. By most measures, this looked like a company on the way up.

Underneath it, the operation told a different story.

It is easy to say a business runs on spreadsheets and email and imagine that as mildly inefficient — annoying, but survivable. The reality is sharper.

Proposal forms moved between agents, customers and underwriters as documents, and the forms themselves changed shape depending on which binder they belonged to. There was no single underwriting record independent of whatever version of the form happened to survive in someone's inbox. When a decision needed to be explained later — to an auditor, a syndicate, or the coverholder's own leadership — the explanation had to be reconstructed rather than retrieved.

Visibility was the first casualty. Management could see that submissions were happening, but not clearly where they were stalling: which quotes sat incomplete, which decisions were pending referral, which approved risks had not yet reached issuance. Quoting could take weeks in a market that increasingly expects days. And in the worst version of the pattern, accepted quotations could wait too long for the operational work of issuance to catch up, creating avoidable uncertainty at one of the most important handoffs in the policy lifecycle — precisely the kind of control gap the operation needed to eliminate before it could scale confidently.

None of this reflected a lack of underwriting skill. It reflected a business whose operating infrastructure had not grown at the same rate as its ambition. For many growing coverholders, the limiting factor is not access to opportunities but the ability to demonstrate that additional authority can be operated safely. That was the ceiling here — not demand, but proof.

The redesign started from a simple reframing: a binder is not paperwork to be filed and occasionally consulted. It is a specification — of who can act, on what, within which limits, and what must be reported back. Once that specification is treated as executable rather than descriptive, most of the operational chaos has somewhere to go.

Agents received a single portal to initiate submissions rather than a form that assumed they already understood the underwriting rules. Based on the prospect and the risk information entered, the platform identified the relevant product and binder, surfaced the coverages that applied, and asked for the underwriting information that risk required. Required information is validated against the applicable workflow before the submission advances; missing information stopped being an untracked email and became a structured, assigned, traceable follow-up.

Binder permissions and underwriting authority now govern what each participant may submit, approve and bind — the authority a syndicate actually granted, applied by the system rather than remembered by a person. Every change to risk data and every underwriting decision is recorded, so the history from first submission to policy issuance can be reconstructed on request rather than hoped for.

The most visible change was faster, cleaner submissions. The more consequential one was structural: the coverholder now had an operating model that could support additional products without reproducing the manual complexity for each one. Underwriters make decisions inside a structured workflow instead of piecing together context from disconnected forms and email threads.

And — the part that matters commercially — the coverholder now had something concrete it could demonstrate to syndicates and capacity providers: not simply a promise that operations had improved, but a controlled model showing how additional authority would be exercised. That operating model strengthened the organisation's ability to support conversations about additional classes and deeper market relationships.

It would be easy to describe this as "automating underwriting." That is not quite what happened, and the difference deserves precision.

Facio executes the decisions that the approved product and binder rules permit it to execute — eligibility, validation, routing, the mechanics of moving a compliant risk forward. Where authority requires judgment or referral, the decision remains with an authorised person. The underwriter still underwrites. What changed is that both kinds of decision — automated and human — now happen inside the authority actually granted, and leave a record someone can examine later.

That is a more durable claim than "we digitised the forms," because it is the claim capacity providers actually care about when deciding whether to grant more authority to a coverholder asking for it.