---
type: Static
title: AI Automation Assessment
description: A fixed-fee engagement that finds the first workflow worth automating, agrees a definition of good enough, and produces a quote for building it.
resource: https://nationaldigital.com.au/services/ai-automation-assessment/
tags: []
timestamp: '2026-09-08T00:00:00.000Z'
language: en-AU
publisher: National Digital
author: Cameron Young
authorRole: CEO
datePublished: '2026-09-08'
dateModified: '2026-09-08'
pillar: static
canonical: https://nationaldigital.com.au/services/ai-automation-assessment/
---

_Home — AI Automation Assessment_

# AI Automation Assessment

*Fixed-fee starting point*

**Find out which workflow AI would pay back on before you build anything. One process examined as it is actually done, a definition of good enough agreed, and a quote for the first automation.**

- [Book the 30-minute call](https://calendly.com/cameron-at-national-digital/ai-automation-assessment)
- [See what you get](#what-you-get)

- **Fee** — $2,950 fixed, ex GST
- **Timeframe** — About a week from the workshop to playback
- **You keep** — The process as mapped, an agreed definition of good enough, a ranked shortlist and a fixed quote for the first build
- **Credit** — Credited in full against the first automation build within 90 days

## Trusted by Australian organisations

- [NRMA Parks and Resorts](/case-studies/nrma-parks-and-resorts/)
- [Luxico](/case-studies/luxico/)
- [MyStrengths](/case-studies/mystrengths/)
- [Micro-X](/case-studies/micro-x/)
- [Gratofy](/case-studies/gratofy/)

*Is this the right starting point?*

## Who it is for

The AI Automation Assessment fits when a team is spending hours each week on a task that looks automatable and you want a payback figure before anyone builds a demonstration.

- A team spends hours each week on triage, extraction, drafting or routing that follows a pattern.
- The board has asked what the business is doing about AI and the honest answer so far is not much.
- An earlier pilot produced a convincing demonstration that never reached production.
- You want to know whether the data behind a process can be read reliably before spending anything on a model.

If you already know the workflow and want it built, the [AI automation](/ai-automation/) pillar carries the published band for the build itself. If the question is broader than one workflow, the [Technology Roadmap](/services/technology-roadmap/) sequences the whole estate first.

*What you get*

## What you get

A written pack about a week after the workshop, and an hour of playback to go through it.

- The process as it is actually performed, mapped with the people who do it.
- Findings on the systems the task depends on: whether their data can be read reliably, whether an answer can be written back, and what rate limits or approval steps sit in the way.
- An agreed definition of good enough, which the build then scores against real historical cases from your business.
- A shortlist of candidate workflows ranked by payback period.
- A fixed quote for the first automation, and what it would take to run it.

The definition is the part most assessments skip. Without one there is no way to tell a system that works from one that reads well, and scoring it against your real cases is the first step of the build. The pack is yours, and it is written so that another team could build from it.

## How it runs

1. **A 30-minute call.** We confirm which process the assessment examines and who needs to be in the workshop. Nothing is invoiced before this call.
2. **The workshop.** Two hours with the people who do the task, on the work as they actually do it.
3. **Data and systems check.** Whether the inputs the task depends on are reachable and clean enough to automate against, and where an answer has to land.
4. **The definition of good enough.** Agreed with you in plain terms, so the build can be scored against the outcome a person produced on the same input.
5. **The pack and playback, about a week in.** The ranked shortlist, the fixed quote and an hour on the reasoning.

> National Digital provided extensive support to our development team throughout all stages of the project. Their expertise in modern web technologies was invaluable.

— Jeff Dermann, Product Owner at NRMA Parks

- Verified review

*After the pack*

## An AI position for the board

Boards increasingly ask for a written position on how a business uses AI: which tasks are automated, what a person still approves, where data goes and how a wrong answer is caught. That document can be added to the assessment and is quoted separately. It draws on the same findings, so it describes what the business actually does rather than restating a generic policy.

## What happens after

Where the shortlist has a workflow worth building, the fixed quote is for that build and the assessment fee is credited in full against it if it proceeds within 90 days. Where the answer is that a workflow should be left alone, usually because the data behind it cannot be read reliably or the volume does not justify the build, that is written into the pack. It is far cheaper to learn here than during a build.

## Questions buyers ask

### How is this different from a strategy deck?

It examines one workflow rather than surveying a business, and it ends in a quote for building something. The deliverable is a definition of good enough, a shortlist and a price, which are the three things the decision needs.

### Will you recommend building something regardless?

Sometimes the answer is that a workflow should be left alone, usually because the data behind it cannot be read reliably or the volume does not justify the build. That answer is written into the pack, and it is far cheaper to learn here than during a build.

### What do you need from us?

Two hours with the people who do the task and read access to the systems it touches. If the assessment leads to a build, a sample of real historical cases comes next, because scoring against them is what turns good enough from an opinion into a number.

### How long does it take?

About a week from the workshop to the playback session, depending on how quickly system access arrives.

### Do you build the automation as well?

Yes, and the fee is credited in full against that build if it proceeds within 90 days. The pack is also written to be handed to another team, which is the point of paying for it separately.

### Which models and tools do you use?

Whichever fits the measure and the constraints, chosen after the evaluation rather than before it. Model choice matters less to whether an automation works than the retrieval around it and the fallback path for cases it should refuse to handle.

## What an AI Automation Assessment costs

A fixed-fee engagement that finds the first workflow worth automating: the task as performed, whether its data is reachable, an agreed definition of good enough, a shortlist ranked by payback and a fixed quote for the first build. Scoring that definition against real cases is the first step of the build.

### The task and its data

The work as it is actually done, and whether a machine can reach what it depends on.

- **Process workshop**: Two hours with the people who do the task, because the documented process is rarely the real one.
- **Data and systems reachability check**: Whether the inputs the task depends on are reachable, clean enough to automate against, and writable where an answer has to land.

### The shortlist and the quote

An agreed definition of good enough, and a ranked view of what to automate first with a price on the first one.

- **Definition of good enough**: What a correct result looks like for this task, agreed in plain terms before anything is built.
- **Ranked shortlist and fixed quote**: Candidate workflows ranked by payback period, with a fixed quote for the first one and what it would take to run it.
- **Playback session**: An hour walking through the shortlist and the definition, so the reasoning is understood rather than just received.

**Total Fixed Fee:** $2,950

**Payment terms:** Payable on booking. Credited in full against the first automation build if it proceeds within 90 days.

**Key Assumptions:**
- Priced as a single fixed-fee package: the lines above are what the engagement is made of, not a menu of separately priced components.
- One process family. Assessing several unrelated workflows is quoted separately.
- Scoring the agreed definition of good enough against real historical cases is the first step of the build and is priced in the build, not here.
- A written AI governance position for the board is available as an addition and quoted separately.
- The fee is credited in full against the first automation build if it proceeds within 90 days.
- The pack is the client's to keep and take elsewhere.

## Find the first workflow worth automating.

Book a 30-minute call with an engineer. We confirm which process the assessment examines and who needs to be in the workshop. The fee is invoiced after the call and credited in full against the first automation build if it proceeds within 90 days.

- [Book the call](https://calendly.com/cameron-at-national-digital/ai-automation-assessment)
- Email: enquiries@nationaldigital.com.au
- Phone: 1300 800 855

*A different situation?*

## The other ways to start.

Each engagement is priced on its own and credited against the work that follows it, so the right one is the one that matches your situation.

### Technology Roadmap

**Fixed $4,950 · credited against the first initiative**

Several systems need rationalising and the order matters.

[See the Technology Roadmap](/services/technology-roadmap/)

### Platform Health Check

**Fixed $3,950 · credited against a rebuild**

You run one platform and need to know what condition it is in.

[See the Platform Health Check](/services/platform-health-check/)

### Product Development Plan

**Fixed $3,950 · credited against the build**

You know roughly what you need built.

[See the Product Development Plan](/services/product-development-plan/)

### Product Support

**From $1,495 a month ex GST**

You want the platform monitored, maintained and moving.

[See Product Support](/services/product-support/)

Still not sure which one fits? [Compare every engagement](/services/) on the services page, or book the call and we will tell you on it.

## Related

**Related:**
- [National Digital - Custom Software, AI & Platforms](/okf/home.md)
- [Talk to an engineer.](/okf/contact.md)
- [Careers at National Digital](/okf/careers.md)
- [Graduate Full-Stack Web Developer](/okf/australia-remote-graduate-full-stack-developer.md)
- [Senior Software Engineer (Frontend)](/okf/australia-remote-senior-software-engineer-frontend.md)
- [Senior Software Engineer (Backend)](/okf/australia-remote-senior-software-engineer-backend.md)
- [Privacy Policy](/okf/privacy.md)
- [Website Terms of Service](/okf/terms.md)
- [Accessibility Statement](/okf/accessibility.md)
- [Small team. Serious platforms.](/okf/about.md)
- [Real platforms, in production.](/okf/case-studies.md)
- [Civic software we build and run](/okf/projects.md)
- [Senior engineering, end to end.](/okf/services.md)
- [Knowledge library.](/okf/expertise.md)
- [QA & Test Automation Engineer](/okf/australia-remote-qa-automation-engineer.md)
- [Booking systems built around your operation.](/okf/booking-systems.md)
- [Custom software, built by senior engineers.](/okf/custom-software.md)
- [AI that survives contact with production.](/okf/ai-automation-services.md)
- [NewBook Integration & Custom Development](/okf/newbook-integration.md)
- [cruisePAL (Resco) Integration & Custom Development](/okf/cruisepal-integration.md)
- [Customer portal development, built around your access rules.](/okf/customer-portals.md)
- [An Australian software development company you can talk to directly.](/okf/software-development-company.md)
- [Custom appointment scheduling, built for how your service actually books.](/okf/appointment-scheduling-systems.md)
- [Venue booking systems built around your capacity, not a vendor's template.](/okf/venue-booking-systems.md)
- [Web application development for products that have to hold up.](/okf/web-application-development.md)
- [AI chatbot development that answers, escalates and knows its limits.](/okf/ai-chatbot-development.md)
- [Document and workflow automation, wired into the systems you already run.](/okf/document-workflow-automation.md)
- [AI consulting from the engineers who would have to build it.](/okf/ai-consulting.md)
- [Governance and Credentials](/okf/governance.md)
- [Information Security and Privacy Statement](/okf/governance/information-security-and-privacy.md)
- [Quality Policy](/okf/governance/quality-policy.md)
- [Work Health and Safety Policy](/okf/governance/work-health-and-safety-policy.md)
- [Environmental Policy](/okf/governance/environmental-policy.md)
- [What will it cost to build?](/okf/estimate.md)
- [Product Development Plan](/okf/services/product-development-plan.md)
- [Replacing a legacy booking system — without losing a booking.](/okf/legacy-booking-system-replacement.md)
- [Product Support](/okf/services/product-support.md)
- [Platform Health Check](/okf/services/platform-health-check.md)
- [Technology Roadmap](/okf/services/technology-roadmap.md)
