The Numbers Behind Perth's Duplicate Image Problem: What the Data Reveals

Perth businesses are sitting on bloated digital libraries stuffed with duplicate images, and the cost is measurable. Across the property, resources and construction sectors — three pillars of the WA economy — unmanaged duplicate image files are inflating cloud storage bills, slowing internal workflows and, in some cases, creating legal exposure around image licensing. The problem is not new, but the numbers have caught up with it.

The timing matters. Western Australia's housing construction pipeline remains under pressure, with the State Government's Metronet expansion driving precinct development from Yanchep in the north to Thornlie-Cockburn Link in the southeast. Property marketing teams at agencies along St Georges Terrace and in the Leederville commercial strip are processing thousands of listing images weekly. Duplication rates in high-volume digital environments like these are rarely trivial.

What the Data Actually Shows

Industry research published by technology consultancy Gartner in 2024 estimated that between 30 and 40 per cent of files stored in unmanaged enterprise digital asset systems are exact or near-exact duplicates. For a mid-sized Perth real estate group running 20 active agents and shooting 60 to 80 property listings per month, that translates directly into wasted spend. Commercial cloud storage at the enterprise tier — through providers such as Microsoft Azure or AWS, both of which operate Australian data centres — currently runs at roughly $25 to $35 per terabyte per month at list pricing. A library carrying 40 per cent redundancy on a 10-terabyte archive is burning around $100 to $140 a month on files nobody needs.

The issue compounds when licensing enters the picture. Stock image platforms including Getty Images and Adobe Stock embed metadata in their files. When duplicate images are distributed internally without tracking, organisations risk deploying the same licensed asset beyond the terms of a single-seat purchase. For a resources company operating out of West Perth with marketing functions spread across multiple project sites — think the Pilbara corridor from Port Hedland down — that compliance gap is a genuine legal exposure, not a theoretical one.

Locally, two organisations have moved to address this systematically. The University of Western Australia's digital collections team, based on Crawley's Stirling Highway campus, began an internal audit of its research image repositories in early 2025, prioritising near-duplicate detection across its institutional photography archives. Separately, the WA Museum — whose Boola Bardip venue at Yagan Square on the Perth Cultural Centre precinct holds one of the largest digitised collections in the southern hemisphere — has publicly committed to a multi-year digital asset management overhaul, with duplicate rationalisation a stated component of that work.

Why Perth's Economy Makes This Worse

Perth's dual exposure to a resource boom and a residential construction surge means the volume problem is structural, not cyclical. Iron ore royalties flowing into the State Budget — WA recorded a surplus of $3.7 billion in 2024-25 — have funded expanded government communications and marketing functions across multiple agencies. Each agency generates imagery. Each project generates imagery. Without centralised governance, duplication multiplies across departmental shared drives and Microsoft SharePoint tenancies.

The AUKUS defence precinct developing around HMAS Stirling on Garden Island is generating its own documentation and communications imagery load as contractors from Osborne Road in Fremantle through to the Henderson marine industrial area come on board. Defence procurement environments carry strict data governance obligations, which means duplicate image management is not discretionary in that context — it is a contractual requirement.

For Perth businesses that have not yet run a duplicate audit, the practical starting point is a file hash analysis across existing storage — most enterprise IT teams can run this with existing tools in under a week on libraries up to 5 terabytes. Third-party platforms such as ImageDeduplicator and Canto offer automated near-duplicate detection that catches near-matches, not just byte-identical files. Prices start around $200 per month for small-to-medium teams. The return on a single month's audit, for any organisation holding more than a terabyte of imagery, will almost always outrun the cost of not doing it.