AI & automation services

How we work: from a free call to your foundation build, one step at a time

Every GCIT AI and automation engagement follows the same path. A free discovery call of up to 30 minutes, then the AI Discovery Workshop when there is a problem worth exploring, then your foundation build, then quarterly build cycles or gcOS Care. You decide at every step whether to go further.
The discovery call is free. The AI Discovery Workshop is A$695 ex GST, credited in full if you go ahead.
Two GCIT team members smiling as they look at a screen together at a desk in the Burleigh Heads office
This is how gcOS by GCIT is delivered. gcOS is your data in one governed place, automations doing the repetitive work, and your own portal for your team, run and secured by GCIT. The steps below are how you get there, one proven step at a time.
If you are comparing AI implementation partners, this page is the whole process in one place. It is written for owners of Australian small and medium businesses who want to know what they are buying, what they get at each step, and where they can stop. Not planning to build anything? The Claude for Business Workshop gets your team doing real work with Claude as it ships today. Back to AI & Automation Services.

What are the steps of an AI engagement with GCIT?

Seven stages, from a short call to software that keeps running. Each one ends with something you can use, and you decide at two points whether to go further. Each stage shows what happens, what we need from you and what it costs.
  1. Discovery callFree, up to 30 minutes
  2. AI Discovery WorkshopA$695 ex GST
  3. Access and feasibilityIncluded
  4. AI outcomes presentationFindings and proposalYou decide

    Or stop here, and keep the findings

  5. Foundation buildFixed price, with your workshop fee credited
  6. Keep buildingFixed price or quarterlyYou decide
  7. gcOS CareStarts after sign-off

    Start again whenever you are ready

Stage 1 of 7, find the problem

A short call to see if it is worth exploring

We introduce how we work and find out whether your problem is worth exploring together in a workshop. It might take five minutes or thirty.

From you
The person who owns the problem.
You leave with
A clear answer on whether a workshop is worth your time, or an honest answer that it is not time yet.
What it costs
Free, with no commitment.

More about the discovery call

Stage 2 of 7, find the problem

Pick the one problem most worth solving

We sit down with the people closest to the job, pick the one problem most worth solving first, and agree how success would be measured. Nothing is built in the workshop.

From you
The owner of the problem, someone who does the work every day, and the person who can authorise access.
What to bring
The last real example of the job, how often it happens and roughly how long it takes.
You leave with
One agreed problem, how we would measure success, and a plan for secure access.
What it costs
A$695 ex GST, covering the workshop and the feasibility findings. Credited in full against your foundation build if you go ahead.

More about the AI Discovery Workshop

Stage 3 of 7, prove it on your data

Tested on your real data

Your system owner authorises read-only access, never in a document or an email. We test the problem on your real data, measure how the job runs today and build a demo of your own portal in your logo and colours.

From you
Access approved by your system owner, and quick answers about the data.
How long
It depends on how quickly access can be set up. We agree the timing after the workshop.
Your data
Access stays read-only. A person approves any later change before it happens.
What it costs
Included in the workshop fee.

Stage 4 of 7, prove it on your data

A clear verdict and a fixed price

In person or remote, with the person who can approve a build. We show what the test found, click through your demo portal with you and give you a clear verdict: proceed, proceed with conditions, or stop.

You receive
Feasibility findings, how it would work, how success is measured, your demo portal, and a fixed-price foundation build proposal including running costs.
From you
The person who can approve a build. We also agree who will own the build and who will give day-to-day feedback.
If the answer is stop
We tell you, and you keep the findings.
You decide
Go ahead with the foundation build, or stop here. If you go ahead, your A$695 is credited in full.

More about the AI outcomes presentation

Stage 5 of 7, build it

One process working in your own portal

Your foundation build fixes the one problem, on your real data, and is closed against tests agreed before we start. Your team uses it on real work before it is signed off.

How it runs
  1. Most of it built and tested before build day 1
  2. Build day 1 onsite, your team uses it on real cases
  3. One to three feedback weeks
  4. Build day 2 onsite, checked against the agreed tests
  5. Written sign-off, then gcOS Care starts
From you
A project owner who owns the result, and a feedback lead who uses it every day and tells us what is wrong.
What it costs
A fixed price, set in your proposal. One invoice before the build starts, with your A$695 credited.

More about the foundation build

Stage 6 of 7, keep it running and growing

Choose how to keep building

At build day 2 we show you the next outcome worth going after, written as the change you would see and how it would be measured.

Fixed-price build
One defined outcome, scoped and priced before we start.
Quarterly engagement
Embedded or Strategic. You choose the outcomes, each with a baseline and a target date. A set range of onsite visits each month, billed monthly a quarter at a time, with gcOS Care included.
You decide
Keep building with the next outcome, or rest on gcOS Care. You can start again whenever you are ready.

More about quarterly build cycles

Stage 7 of 7, keep it running and growing

What we build, we keep running

Nothing we build is left without support. gcOS Care keeps your portal, automations and connections running, secure and up to date.

What is included
Hosting, monitoring, patching, break-fix and minor adjustments for what we delivered.
When it starts
Only after you sign off what was delivered.
Coming back to build
Your list of next outcomes is kept, so you can start again whenever you are ready.
What it costs
Priced in your proposal, and included while a quarterly engagement runs.

More about gcOS Care

Certified Services Partner, Claude Partner Network

GCIT is Certified in the Claude Partner Network

Elliot Munro, Phoenix Asange-Harper and Shay Brokensha hold Anthropic’s Claude Certified Architect - Professional credential. A baseline and a success metric are agreed before we build.

Read the announcement

What a first build looks like

One process, for one team, measured against how the job ran before. Client results are anonymised.

A local metal fabrication business

About 99% of quote lines matched the estimators’ own quotes on replayed orders, with zero automation errors.

A commercial real estate agency

Property enquiries are answered in about a minute, around the clock, with the correct Information Memorandum attached every time (100%). Uncertain matches go to a person.

Read the story →

A precast concrete manufacturer

The forecasting portal was running on day one. It found it was reading a small fraction of the workbook and now refuses short reads.

A local accounting practice

A scheduling portal went live four days after the scope was agreed. It replaced a legacy macro workbook, and the SharePoint sync is read-only with no stored secret.

How much does AI implementation cost with GCIT?

The first step costs nothing. The second is one fixed fee, credited in full if you go ahead. Everything after that is priced in your proposal, with running costs, before you commit to it.
What each step costs, what you leave with, and what you commit to
StepPriceWhat you leave withCommitment
Discovery callFreeA clear answer on whether a workshop is worth your time, or an honest answer that it is not time yetNone
AI Discovery WorkshopA$695 ex GSTA feasibility report (verdict, data map, baseline and the acceptance tests a build must pass), a branded demo of your portal, and a fixed-price proposal for your foundation build, including running costsOne fee, credited in full if you go ahead
Your foundation buildPriced in your proposalOne process working in your own portal, closed against the agreed tests with a named sign-offOne invoice, paid before the build starts
Quarterly build cycles (Embedded or Strategic)Priced in your proposalEngagements run as quarterly sprints, billed monthly, with a set range of onsite visits each monthA quarter at a time
gcOS CarePriced in your proposalHosting, monitoring, patching, break-fix and minor adjustments for what we deliveredCharges start only after delivery
All prices ex GST. The AI Discovery Workshop is the only step with a published price, because every build after it is different. Its price, scope and running costs are set out in writing before you decide.
Not planning to build anything?The Claude for Business Workshop gets your team doing real work with Claude as it ships today. We set Claude up properly for your business and work one business function through end to end with your team. A$695 ex GST, with nothing to build.

How do you find out if it is worth building?

We build in order: your data first, then your portal, then automations and agents.

The discovery call: free, up to 30 minutes

A conversation with the person who owns the problem and someone who does the work every day. We introduce how we work and find out whether your problem is worth exploring together in a workshop. It might take five minutes or thirty. Nothing is built for this call.
We will tell you honestly when the better first step is a data clean-up, an integration or a plain automation rather than AI.

The AI Discovery Workshop: tested on your real data

When there is a problem worth solving, your system owner authorises secure access, never in a document. We check the idea against your real data, measure today’s baseline and build a branded demo of your own portal.
Then the AI outcomes presentation. You see the feasibility report with its verdict (proceed, proceed with conditions, or stop), a map of your data, the baseline, the acceptance tests a build would have to pass, your demo portal, and a fixed-price proposal for your foundation build, including running costs. There is no separate proposal meeting.

What happens in your foundation build?

One process, working in your own portal on your real data, and closed against tests you agreed before we started.
  1. Before build day 1
    Most of the build happens here
    We build most of it on the demo portal you saw at the workshop meeting, test it on your data and rehearse the workflow, so build day 1 starts with something working.
  2. Build day 1, onsite
    Your team tries it on real work
    The people who do the job use it on real cases. We agree a short feedback list together, and new ideas are parked for later.
  3. One to three feedback weeks
    We make the agreed changes
    Your team keeps using it on real work while we make and retest the agreed changes. There is no extra visit.
  4. Build day 2, onsite
    Closed against the agreed tests
    A named person on your side signs off against the tests agreed before the build. The build isn’t closed until those tests pass. You get a user guide and a written record of what was delivered, how it was tested and who supports it.
How it is paidOne invoice, paid before the build starts. Your A$695 workshop fee is credited in full against it.
Your two peopleA project owner who owns the result: the person who runs the process today, there on both build days. A feedback lead who uses it every day and tells us what is wrong. We agree both at the AI outcomes presentation.

Is your problem ready for a foundation build?

  • Worth it. The time or cost the job takes today should outweigh the build, using your own numbers.
  • Rule-shaped where possible. Clear business rules come first, with AI added at decision points and a person in the loop.
  • Owned by someone in the room. The person who runs the process today is there on both build days.
  • Access secured first. Working access to the data, arranged and tested before build day 1.

Every foundation build is

  • One process, end to end. Data mapping is part of the work, not unpaid extra.
  • Always a portal. Your team sees and steers the work in its own web portal.
  • One department. A specific team, never a company-wide rollout.
  • Up to three connections to your existing systems, and ideally two.
  • Dashboards and automations, or new connections. Not both in the same build.

A foundation build never

  • Replaces your core platform, such as your CRM, ERP or practice system
  • Touches anything running in production. It can be switched off with nothing changed.
  • Faces your customers. Agents and anything customer-facing come later, in quarterly build cycles.

How do you prove an AI build is worth it before you start?

We only start a build when we can show the return: how long the process takes you now and what we think we can get it to.
A baseline and a success metric are agreed before we build. The metric is in business terms, such as hours a week, turnaround time or error rate, and the baseline uses your own numbers with the agreement of the person who runs the process. A rough baseline is fine at the start. Precision can improve later, but agreement cannot be added after the fact.
We’ll only do this if we can demonstrate that it’ll have an ROI benefit… how much is this process taking you now
Elliot Munro, CISO, GCIT

Agreed before anything is built

  • The metric. One measure your business already cares about.
  • The baseline. What the process costs you today, in your numbers.
  • The acceptance tests. Written before the build, so finished means the same thing to both of us.
  • The owner. The named person in your team who signs off on build day 2.
Good to knowIf the numbers do not stack up at the workshop, we say so and stop there. You still keep the feasibility report.

What do you need to have ready before we come on site?

Working access to your systems, tested in advance. We always secure API access before we go on site, and we test against your full history before anything ships. A build day spent chasing logins is a day wasted, so if access is not ready we reschedule rather than burn the day.
Keep the room small. A build day with an audience turns into a presentation.
A GCIT team member in glasses smiling at his desk in the office

Our pre-visit checklist

  • Accounts ready for the systems involved, set up to the agreed scope
  • A named admin contact on your side who can unblock access on the day
  • A real record retrieved end to end before the day, so we know the data is reachable
  • The person who does the job today in the room, plus the sponsor for the start and the wrap-up
After build day 2

What happens after your foundation build?

You choose: a quarterly build cycle, gcOS Care on its own, or stop.

Quarterly build cycles

“Rather than hand you a report, we put an engineer in your business who maps your processes, builds the automation, proves the time it saves, and stays to keep improving it.”
Engagements run as quarterly sprints, billed monthly, with a set range of onsite visits each month.
There are two rhythms, Embedded and Strategic. The difference is how fast you want to change, not the size of your business. After a completed quarter you can rest on gcOS Care and come back when there is more to build.
Priced in your proposal.

gcOS Care

Ongoing care covers hosting of your portal and connectors, monitoring, patching, break-fix for delivered features, minor adjustments and continuous security. New features are scoped separately. Charging starts only after delivery.
gcOS Care is included while quarterly build cycles run. We run, secure and maintain what we build, the way we have run IT for our clients for years.
On its own, priced in your proposal.

What do you receive along the way?

Three documents, each one yours to keep whether or not you go further.

Feasibility report

From the AI Discovery Workshop. The verdict, a map of your data, the baseline and the acceptance tests a build would have to pass.

Proposal for your foundation build

Shown at the AI outcomes presentation. What is in and out, the tests, the two build days, the fixed price and the running costs.

User guide and handover record

At build day 2. How to use it, what was delivered and tested, and who supports it under gcOS Care.
Later, separatelyWe come back after handover to review how your team is using it. That review is separate from the build and never triggers a charge.

Why choose GCIT as your AI implementation partner?

GCIT is a managed IT and security provider first. What we build is handed over at build day 2 with a written handover record, then run and secured under gcOS Care, so it is not left with you to keep running. We build with Microsoft 365 and Claude by Anthropic.
We follow the approach in Anthropic’s guide to building effective agents: start with the simplest thing that works and add complexity only when it earns its place. For governance, the international standard for AI management systems, ISO/IEC 42001, is a useful reference point for any business putting AI into daily use.
For your IT lead: how the builds are put together
Every project inherits a shared portal kit: Entra ID sign-in, an MCP tool package, audit logging and data mapping. Improvements flow to every project.
Identity
Staff sign in to the portal with their own Microsoft Entra ID accounts in your Microsoft 365 tenant.
Connections
Model Context Protocol (MCP) servers and connectors to your line-of-business systems. Three connectors at most in a foundation build, and it can be switched off with nothing changed.
Testing
Our shadow runs score the automation against what staff actually did. Going live means adding the part that acts on the results, not flipping a switch. If there is no data to score, the result shows as no data, never as 100%.
History
We test against your full history before anything ships.

Is there another way to start?

If you already know the area you want to look at, two focused workshops sit beside this path.

Conversational Intelligence Workshop

A structured session on how AI can capture, score and act on the conversations your business already has.

Microsoft Copilot Readiness Workshop

For Microsoft 365 businesses planning Microsoft 365 Copilot: the use cases worth paying for, licensing scenarios and a rollout plan.

Questions buyers ask us about how we work

The discovery call is free. If there is a problem worth solving, the AI Discovery Workshop is A$695 ex GST, credited in full if you go ahead. Your foundation build, quarterly build cycles and gcOS Care are priced in your proposal, with running costs, before you commit.

You can stop after any step. Stop after the workshop and you keep the feasibility report. Stop after your foundation build and you keep a working build, closed against the agreed tests. Quarterly build cycles run a quarter at a time, so you can stop at the end of any quarter. gcOS Care only starts after delivery, so there is no ongoing fee for work you did not go ahead with.

Because it is real work on your real data, not a sales meeting. Your system owner authorises access, we test the idea on your data, build a branded demo of your portal and write up the verdict and a fixed-price proposal. The A$695 is credited in full if you go ahead.

One process that is worth it on your own numbers, shaped by clear rules where possible, owned by someone in the room, with access to the data secured first. Repetitive work done by a named person every week is usually the best place to start.

No. Your foundation build is one internal process in your own portal, for one department, with nothing customer-facing. Agents and anything your customers see come later, in quarterly build cycles, once your data and portal are in place.

A baseline and a success metric are agreed before we build, and the acceptance tests are written first. On build day 2 a named person on your side signs off against those tests. Later we measure against the same baseline, using the same method, with the person who runs the process.

You choose: a quarterly build cycle (Embedded or Strategic), gcOS Care on its own, or stop. gcOS Care is included while quarterly build cycles run, and on its own it starts only after delivery. We come back later to review how your team is using it. That review is separate from the build and never triggers a charge.

Time and capacity back. We measure every build in the hours or turnaround the people doing the work get back, using the baseline you agreed. On our own service desk, technician productivity rose by over 30%, and most tickets are still worked by technicians using AI tooling.

No. We build in order: your data first, then your portal, then automations and agents. If Microsoft 365 Copilot is already part of your plans, the Copilot Readiness Workshop is the Microsoft-focused starting point.

Start with a free discovery call

Up to thirty minutes to see whether your problem is worth exploring. If there is one, the AI Discovery Workshop tests it on your real data before you spend anything on a build.
Your dataTo test your idea we only need to look at your data, never change it. The person who manages your systems sets up our access, so you never send us a password in an email or a document. Nothing changes in your systems unless someone on your side approves it first, and GCIT is ISO 27001 certified for information security.
AI Discovery Workshop: A$695 ex GST, credited in full if you go ahead.
Last reviewed
Office
1/16 Dover Drive, Burleigh Heads QLD 4220
ISO 27001  |  ISO 9001  |  ISO 14001  |  Essential Eight assessment certified (TAFE Queensland)  |  Microsoft Solutions Partner
  • What happens next?

    Our team will read your enquiry, get in touch to understand what you need and agree the next step with you.

Ready to secure and simplify your IT? Talk to a GCIT expert today.