Twelve years in AI and automation · Auckland, New Zealand 021 365 082 Taking new projects Book a free call
Home What I build Case studies FAQ Newsletter Contact Book a free call

Case study · Property development

Subcontractor agreements and supplier invoices that handle themselves

A New Zealand property developer runs several active builds with no administrative staff. Agreements now generate, send, chase and file themselves, and the accounts inbox sorts itself. One recurring task went from an hour to two minutes.

Time cut per agreement, an hour to two minutes97%
Saved a month on agreements alone15 to 20 hours
Admin hires, the stated goal0
DeliveryFixed price, free discovery

Background

The situation

The developer runs multiple active builds with no administrative staff. Every subcontractor line needed an agreement prepared from the job details, sent, chased, countersigned and filed. Every supplier invoice, statement, payment claim and reminder arriving in the accounts inbox had to be read, matched to a project and routed by a person.

The owner's brief was blunt: replace repetitive admin with AI rather than hire for it. Both automations run in the developer's own AWS account, under their own rules and templates, with the office team keeping the approve step.

Our approach

AP·01

Discovery on the real work

Walked the subcontractor process and the accounts inbox with the owner and the finance lead, line by line, and sized the build against logged hours rather than guesses.

AP·02

Their rules, their templates

The developer's own agreement templates, filing rules and numbering were kept exactly. The system fits the business, not the other way round.

AP·03

Shadow mode first

The invoice reader ran in shadow mode against live mail before it was allowed to file anything, and the agreement pipeline had a full dry run with test recipients before the first real line.

AP·04

Daily audit after go-live

Every working day an audit replays the flagged messages, checks nothing was misfiled, and ships fixes the same day. Rules that change are re-run against the backlog.

Challenges and solutions

What was slow, and what changed

Challenge 01Subcontractor agreements
Before

Each agreement took about an hour: find the line details, fill the template, send, chase the signature, countersign, file, tell the office.

  • About an hour per agreement, three to five a week
  • Chasing signatures by hand
  • Signed copies filed late or in the wrong place
What we built
  • Agreement built from the job line and the developer's own template
  • Sent for e-signature through DocuSign, chased automatically, countersigned and filed
  • Run sheet showing every line and its state, office owns the approve step
Result
  • Two minutes per agreement, in the owner's words
  • 15 to 20 hours a month back on this task alone
  • Live from 1 October 2026 after a full dry run
Challenge 02The accounts inbox
Before

Supplier invoices, statements, payment claims, reminders and receipts all landed in one mailbox and needed a person to read and route each one.

  • Up to 42 supplier messages a day
  • Each read, matched to a project and filed by hand
  • Easy to miss a claim buried in a statement thread
What we built
  • Every message classified: invoice, statement, claim, reminder, receipt
  • Filed under the developer's own rules and numbering, matched to supplier and project
  • Only the ones needing a decision reach the finance lead
Result
  • Zero misfiled across the October audits
  • Finance lead works a short queue instead of an inbox
  • Live from 26 August 2026

Measurable results

The numbers

Time reduction
97%
an hour per agreement down to two minutes
Hours saved
15-20
per month, at three to five agreements a week
Annual saving
$10K+
admin time on agreements alone, conservative estimate
Admin hires
0
the owner's stated goal, two months in
Before automation
1 hour per agreement
  • Templates filled by hand
  • Signatures chased by email
  • Inbox read daily by the finance lead
  • Filing depended on one person
After automation
2 minutes per agreement
  • Generated, sent, signed and filed automatically
  • Inbox sorted, only decisions surfaced
  • Run sheet and queue visible to the whole office
  • Daily audit, fixes shipped the same day

Implementation timeline

Aug 2026

Discovery and shadow mode on live mail

26 Aug

Invoice automation live

24 Sep

Agreement pipeline dry run passed

1 Oct

Agreements live, third project engaged

What changed day to day

Business benefits

Fixed price, their templates, their AWS account. If your office is drowning in agreements and invoices, this is the shape of the fix.

Book a free 30-minute call

BB·01

No admin hire

The goal was to replace repetitive admin rather than add a person. Two months in, the developer still runs with no administrative staff.

BB·02

No single point of failure

Agreements and invoices no longer wait for one person at a desk. The run sheet and the queue show the state of every item to whoever picks it up.

BB·03

Nothing lost in a thread

Every agreement is chased and filed, every invoice is matched, and the ones needing a human decision are flagged rather than forgotten.

BB·04

Audit trail by default

Signed agreements, invoices and statements land in the right folder with the right name every time, which makes month end and any later dispute simpler.

BB·05

Faster for subcontractors and suppliers

Agreements arrive the day a line is ready and replies are consistent, part of why the third project came within six weeks.

“We're a small company with no administrative staff, and our goal has been to replace repetitive admin tasks with AI. Skene has been the right partner for that. We've completed two projects with him so far, one of which cut a task from an hour down to two minutes, and we've just engaged him for a third.”

Callum, owner, Google review, October 2026

More case studies

Other industries

CS

Legal practice

Court documents, client intake and submissions automated for a New Zealand law firm.

CS

Retail

Inventory, supply chain and customer engagement automation for a NZ retail chain.

CS

FinTech

AI-driven IT operations and ticket triage for a New Zealand FinTech.

CS

Insurance

Claims processing, data entry and fraud detection for a NZ insurance broker.

CS

Air conditioning

A custom CRM, scheduling and billing system for an NZ-wide HVAC company.

CS

Hazard consultancy

Asbestos clearance reports read by vision and billed through simPRO.