Fictional client
By then a contractor was taking the crop off at three times the cost, and Kerrindale’s client had learned something about Kerrindale. Thirty-one deals missed their season last year and fourteen never came back — $18.6m of lending and $112,000 of commission, lost not to a competitor’s rate but to a week. Only two of the thirty-one were the client leaving it too late. Agribusiness finance is a business where the deadline is set by the weather, and the software either works backwards from it or it does not.
Kerrindale Rural Finance does not exist. We cannot show live client systems, so this study is built around a fictional brokerage — invented firm, invented brokers, invented farming families, invented lenders and invented figures. What is not invented is the work: the group structures, the seasonal windows, the security classes and the failure modes are all real, and the screens are genuine working pages rather than pictures of pages. Every mockup on this page opens and responds to clicks.
On scope: this is written for agribusiness finance broking in Australia, with a Riverina firm as the example. Commodity mixes, seasonal calendars, water markets and lender appetite differ enormously by region and by year — in a real build those are configuration rather than code. All figures are Australian dollars.
Kerrindale places agribusiness debt across the Riverina from offices in Wagga Wagga and Griffith. Six brokers, 180 client groups, $840m under management and relationships measured in decades rather than deals. Its problem is not finding clients or lenders. It is that every deal begins by rebuilding, from scratch, a picture of a family that has not fundamentally changed since the last one.
Agribusiness broking has two things no other lending niche has: a client that is nine entities rather than one, and a deadline that nobody can negotiate with.
Two ideas, and they are the two things this business already knows and cannot act on. First, the client is the group, not the borrower. Nine entities, four people, who guarantees what, which title sits in the super fund, who holds the water — assembled once and reused on every deal, every review and every conversation, instead of rebuilt over nineteen days each time.
Second, a seasonal deal cannot be opened without a season date, and every date on it derives backwards from that — machine on farm, settlement, approval, submission, group position. Not a reminder. A plan that reads late when it is late, on the day it becomes late, rather than in October.
Seven screens follow. Each one opens as a real page — click through them.
Every live deal against its window, ranked by float — the number of days that can still be lost before the season date is missed. Underneath, the honest analysis of the thirty-one deals that missed last year, and the finding that only two of them were the client leaving it too late.
Then the table nobody at Kerrindale has ever seen on one page: which accounting firms actually produce financials, and how long they take. Two firms account for fifteen of the twenty delayed deals.
The screen this study is built around. Harvest starts on 12 October and nobody can move it, so every date on the deal is derived backwards from it — machine on farm five days before, dealer needs a week, lender needs a fortnight from approval, credit needs a fortnight from submission.
Read forwards, this deal looks comfortable: it came in on 11 August for an October harvest. Read backwards from the only date that matters, it is already three days late. That difference is the whole screen.
The idea the whole system rests on. A farming client is not a borrower — it is a family group. Nine entities, four people, five facilities across four lenders, one property leased between related parties and a separate company holding the water.
It also records the thing that governs everything and is currently written down nowhere: Cameron guarantees two of the five facilities and Sophie guarantees none. That was a succession decision made in 2021, and it constrains what can be asked of which lender.
An agricultural balance sheet is four different kinds of thing. Land sits still. Water trades separately from the dirt. Livestock breeds, dies and gets sold between one count and the next. Crop is an estimate standing in a paddock.
The screen answers the first question every lender and every client asks — what is actually free — which currently takes an analyst half a day and is out of date the moment it is finished.
Fourteen lenders whose appetite changes by commodity, region and security — and changes in a phone call between one BDM and one broker, which never reaches the other five. Five deals last year went to a lender whose appetite had already moved. Six weeks each, on deals with a season date.
It also catches the second-order consequence: one lender now wants a water valuation under twelve months old, and the group’s largest unencumbered asset has quietly stopped being usable with their main lender.
$840m under management, $96m of it maturing inside twelve months, and thirty-eight of those facilities in nobody’s diary. A balloon that arrives unannounced is a client ringing their bank directly.
And the number that should worry a principal most: trail has stopped on facilities worth $38,000 a year and nobody knows why. Either clients have refinanced elsewhere or a lender stopped paying. Both matter.
The family group sees its own position — every entity, every facility, when each one comes up for review, and what is holding up the deal in front of them.
It also does something a broker rarely does in writing: it tells the client that the hold-up is their own accountant, explains exactly why the dates are what they are, and says plainly that a call from them will move it faster than another email from the broker.
Four phases of two weeks each — eight weeks end to end. The group record goes first: the security position, the deal plan, the maturity diary and the client portal are all the same information seen from different angles, and none of them can be built until the group exists. Each phase ends with something the team actually uses, not a demo.
Weeks 1–2.
Weeks 3–4.
Weeks 5–6.
Weeks 7–8.
Eight weeks is possible because most of it already exists. The entity and relationship records, document handling, obligation and expiry tracking, escalation rules, the messaging engine and the portal are running in production across six sectors already — including finance broking, where the same guarantor-and-entity structure is the spine of the system. What actually gets built for an agribusiness brokerage is the sector core: the seasonal backwards plan, the four security classes and the group position. That is the part this study is about, and it is the part we would spend the eight weeks getting right.
This took a fictional brokerage and made its process visible. The same exercise works better with a real one. Tell us how your business actually runs and we will build a working mockup of it — your districts, your lenders, your brand — before anybody commits to a build.
Start a Conversation See other sector blueprints