Healthcare · NDIS allied health · Operations automation

DisabilityAssessments: 90% of operations automated for an NDIS allied-health provider

DisabilityAssessments is an NDIS allied health provider. ComplxAI built the administration around that service as one system: custom onboarding flows, a central knowledgebase, automated matching and alerting, end-to-end invoice processing for providers and plan managers, and automated reconciliations, all on secure storage for patient data. The client reports 90% of its operations automated.

90%

of operations automated (client-reported)

10%

human resource the operation runs on (client-reported)

$1–3M

per year, the return (client-reported range)

The problem

Six admin steps that had to run as a single system

The administration around an NDIS allied health service is a chain: onboarding, the knowledge the team draws on, matching, alerting, invoice processing for providers and plan managers, and reconciliations. Each step feeds the next. DisabilityAssessments needed that chain run as one system rather than by hand.

Two things shaped the build: the system holds patient data, so secure storage had to underpin everything; and invoice processing covers two parties, providers and plan managers, so it had to run end to end for both.

Constraints

Patient data throughout, so secure storage underpins the whole system
Invoice processing for two parties: providers and plan managers
One connected system from onboarding to reconciliation

At a glance

Client: DisabilityAssessments (disabilityassessments.com.au)
Industry: Healthcare · NDIS allied health
Scope: Onboarding, knowledgebase, matching and alerting, invoice processing, reconciliations, patient-data storage
Capabilities: Custom software development, AI automation, AI integration, AI for healthcare
Ownership: Client owns the IP; documentation and handover included

The solution

Six parts, connected from intake to reconciliation

The build follows the operation in order. Custom onboarding flows take intake in at the front, and a central knowledgebase holds what the team draws on, so both live inside the same system that runs everything after them. Matching and alerting run automatically from there. At the back, invoice processing runs end to end for both parties it covers, providers and plan managers, and reconciliations run automatically behind it. Secure patient-data storage sits underneath all six parts. Because the parts are connected, what enters at onboarding carries through to reconciliation, and the client reports 90% of operations automated, the operation running on 10% human resource, and a return of $1–3M a year. The four layers below show how the parts sit together.

Architecture

Four layers, front to back

Described at the level of its capabilities; the stack behind them is the client's to share.

Frontintake and knowledge
Custom onboarding flowsCentral knowledgebase
Middleautomation
Automated matchingAutomated alerting
Backfinance automation
End-to-end invoice processingProvider invoicingPlan-manager invoicingAutomated reconciliations
Datastorage
Secure patient-data storage

For how we engineer automation across operations and connect it to the systems already in place, see our AI integration services.

How we work

What an engagement like this looks like

Five stages. Each is priced in writing before it starts, you can stop at any boundary, and the IP is yours from the first day of any paid stage. The model in full is on our how we work page.

Intro callThirty minutes, free, on which parts of the administration are worth automating first.
Pre-discoveryAn hour, also free, to test whether the case holds before anything is scoped or priced.
DiscoveryTwo to four weeks from a fixed fee. The output is a use-case map, prioritised and scored on impact, effort and risk.
StrategyTwo to four weeks, priced upfront. The output is a board-ready implementation plan with a build, buy or defer call on each item.
ImplementationBuilt in 12-week blocks, with working software every couple of weeks. Documentation, handover and a 14-day support window after launch are included, then paid support if you want it.

Capabilities used

Outcome

An operation that runs on 10% human resource

DisabilityAssessments reports 90% of its operations automated, the operation running on 10% human resource, and a return of $1–3M a year. All figures are client-reported; the dollar figure is a range.

The 10% is the part that still needs a person. We design automation that way: the system carries the repeatable work, and a person stays responsible for the decisions that need one.

Client-reported

90% automated

operations run on 10% human resource

$1–3M / yearclient-reported range

Related capabilities