Skip to content
← All case studies

Construction

Making years of a Sydney builder's documents askable, then automating the estimates

A Sydney construction business had thousands of job files, site notes and tenders sitting as PDFs alongside Procore, and no way to ask them anything. We made every document machine readable on their own servers and gave them a private model to query it. Then we built estimating automations on top, including estimates drafted over WhatsApp from their own cost catalogues.

Job files, site notes and tenders made searchable
1000s

Job files, site notes and tenders made searchable

Documents sent outside their own servers
0

Documents sent outside their own servers

Where an estimate now starts
WhatsApp

Where an estimate now starts

Before

Thousands of PDFs nobody could search.

  • Procore

    Drawings, RFIs, site notes

  • Word and Dropbox

    Tenders, scopes, past quotes

  • Email

    Supplier threads and pricing

What we built

  1. 01On their servers

    Convert

    Every document is turned into machine readable text and indexed on the builder’s own servers. Procore stays the system of record.

  2. 02Private LLM

    Ask

    Staff ask questions in plain English through a private model. No document and no question leaves the business.

  3. 03Cited

    Answer, with the source

    Each answer cites the document it came from. A project manager checks the page rather than trusting a summary.

  4. 04WhatsApp

    Estimate

    The cost catalogue and past quotes feed a draft estimate, sent to the estimator over WhatsApp. The estimator reviews and owns the number.

Questions staff now ask

  • “What did we price this scope at two years ago?”
  • “Which supplier delayed the last three jobs?”
  • “Which jobs ran late last year, and why?”
Sydney construction business, Procore. Systems and steps as built. Client name withheld.

Where they started

A Sydney construction business running Procore for project management, with years of work behind it. Every job had left a trail: tender submissions, subcontractor quotes, site notes, variations, delay notices, supplier correspondence. Thousands of documents, almost all of them PDFs, spread across job folders and inboxes.

Procore held the live projects well. It did not hold the answers to the questions that kept coming up. Why did we go with that supplier over the other one on the last hospital job? What caused the six week delay on the school project? What did we price that scope at two years ago? The answers existed, but finding them meant someone who was there at the time, or a day of digging, and often both.

What we built first

The base layer is what we call a company brain. We took every document, current and historical, and converted it into a structured, machine readable form. We kept the full content rather than summaries, with the job, date, document type and parties attached.

All of it stayed on their own servers. The model that reads and answers questions runs privately inside their environment. No document, and no question about one, leaves the business. For a builder whose files contain subcontractor pricing and client commercial terms, that was the condition for doing it at all.

Every answer cites the documents it came from. A project manager can check the source rather than trust the summary. Procore stayed exactly where it was as the system of record, and nothing was migrated out of it.

What it made possible

Years of decisions became askable. Why a particular supplier was chosen over another on a past job, with the quotes and the email thread that decided it. That is the procurement question every project manager asks and nobody was able to answer quickly. What delayed a specific project, with the site notes and notices in order. How a similar scope was priced last time, and what it cost.

The people who had carried that knowledge in their heads stopped being the bottleneck for it. New project managers can find out how the business handled something before, instead of guessing or interrupting a director.

Then the automations

With the documents structured and the cost history reachable, workflow automation became straightforward, because the data the automations needed already existed in a usable form.

The one the team uses most is estimating over WhatsApp. An estimator or site lead sends a job description, and photos or a scope where there is one, to a WhatsApp number. The system drafts an estimate from the business’s own cost catalogues and past pricing, in their template, and sends it back for review. What used to be an afternoon at a desk is now a first draft waiting on a phone, with a person still owning the number.

Further automations followed on the same foundation. Each one was cheaper than the last because the business’s own information was already machine readable.

Why the order mattered

Most construction businesses ask for the automations first. We did the document layer first because every automation worth building for this client depended on the same thing: reliable access to their own history. Estimates from their cost catalogue, supplier decisions from their tender files, delay analysis from their site notes. Build the layer once and each workflow on top is a small job. Skip it and every automation has to solve the data problem again.

Next step

We map where your hours go and price the work that is worth automating.