Please use one from below list to access this site
| IE 11 | Edge | Chrome | Firefox | Safari |
The NDIS Demand Map is a data analytics tool that provides information to help providers to grow in the NDIS.
Use the NDIS Demand Map to help understand the services and locations NDIS participants may choose by 2025.
The NDIS Demand Map provides an up to date forecast of NDIS demand by postcode across Australia. The forecast estimates the expected demand when participants have minimal barriers to accessing services.
Use the map to find out:
The forecasts use de-identified NDIS and the Australian Bureau of Statistics' 2021 census data to predict what the NDIS will look like by 2025.
The forecast will be updated regularly based on new NDIS data as it becomes available.
It is important to note that the forecast is based on participant spending patterns to date, in particular the spending patterns of experienced participants in developed markets. This pattern may change in the future if participant's preferences change.
The Demand Map provides a forecast range for future participant numbers, spending and workforce. This range reflects the inevitable uncertainty in making forecasts, particularly in areas where the NDIS has not been operating for long or areas with low populations. We estimate that the actual value will fall somewhere within the forecasted range 90% of the time.
The forecast and current demand for neighbouring postcodes can be combined using the 'Select any bordering postcodes you are interested in' feature. This is particularly useful when the ranges are large due to low populations in the desired postcode. However, users should not attempt to independently sum results across postcodes without using this feature. The ranges for individual postcodes will not sum to the actual estimated range as the forecasts are not independent of one another.
We expect 90% of postcodes to fall within the reported ranges; but not necessarily the midpoint of the range.
Website users should not sum categories or sum totals across postcodes, as this will produce inaccurate results.
Users should make their own assessments when using this data.
We welcome any feedback you have on how we can improve the Demand Map. Please send your views through our feedback form.
In interpreting forecast ranges, Demand Map users should note:
The Demand Map uses a 'Deep Autoregressive Neural Network' (DeepAR) machine-learning technique to forecast NDIS participation, and a 'Multivariate Regression' (MR) to forecast the expected annual spend and required workforce to deliver these services. The model draws on census data and NDIS administrative data from over 2,500 postcodes in Australia to project the number of participants in each postcode. The expected value of spend per person for different categories is estimated for these participants using the MR model. Only summary statistics at a postcode level are used in modelling, such that individual data observations are not linked across sources. Job estimates are inferred from a projected volume of services to be delivered to these participants.
The DeepAR model is a global forecasting method. It leverages the historical change in participant numbers as well as census data to predict the future number of participants in a postcode. The ranges produced for the model are directly calculated as part of the modelling procedure, which directly estimates the models variance.
The MR spend model estimates the relationship between postcode participant characteristics, other demographic characteristics and spend per person, for each spend category. The model is then used to predict the average predicted spend necessary to meet the demand of a mature participant in a mature market. The MR model allows for correlation in errors for different categories of spend to ensure more robust modelling. A similar procedure is used to estimate the expected workforce.
To ensure robustness, modelling is evaluated out of sample through step ahead cross-validation. This procedure trains the model on past values, and evaluates performance against the remaining subsample. The average performance of the model is compared to benchmark forecasts and the range of the forecast is evaluated to test whether the 90% of values are falling within the predicted range.
Current demand is defined as the demand for NDIS services over the 12 months period from 1 March 2022 to 28 February 2023, including: the total spend on NDIS services provided during the specified period by postcode, and the number of participants with active plans by postcode at the end of the period (28 February 2023).
There is a toggle made available on the Demand Map to allow users to switch between current and forecast demand figures.
To protect privacy, current demand figures are not available for postcodes with very low resident populations (less than 100 people as per the 2021 census) or when less than 10 NDIS participants reside in the postcode.
These forecasts have been produced by the Department of Social Services. Every effort has been made to provide the most current, correct and clearly expressed information possible on this site. Nonetheless, inadvertent errors can occur and applicable laws, rules and regulations may change. The information contained on this site is general and is not intended to serve as professional advice. No warranty is given in relation to the accuracy or reliability of any information. Users should not act or fail to act on the basis of information contained herein. Users should not rely on the information for any business, commercial or other purpose, and are strongly encouraged to seek professional advice concerning the information provided on this site before making any decision. Users should not rely on sum categories or sum totals across postcodes, as this will produce inaccurate results. Users should not use this tool for the purpose of re-identification. All contributors to this site disclaim all and any liability to any person or organisation in respect of anything, or in consequence of anything, done or omitted to be done by any person, organisation or other user in reliance, whether in whole or in part, upon any information contained herein.
No part of the text or graphics on this site may be reproduced or transmitted in any form or by any means, electronic or mechanical, including by photocopying, facsimile transmission, recording, re-keying or using any information, storage and retrieval system. Certain links on the site lead to resources located on servers maintained by third parties. As such, no representations are made as to the accuracy, currency or any other aspect of the information contained on such servers or the timely, accurate or complete transmittal of such information.
