Canberra's public sector is sitting on a storage crisis hiding in plain sight. Across federal departments headquartered in the city's parliamentary triangle and ACT government offices from Civic to Gungahlin, IT managers have spent the past two years grappling with a specific, unglamorous problem: duplicate images clogging shared drives, content management systems and cloud repositories at a scale that is now costing real money to maintain.
The trigger for renewed urgency is the federal government's Whole of Government Data and Digital Strategy, which set a July 2026 deadline for agencies to complete digital asset audits as part of broader data hygiene obligations. That deadline, combined with rising cloud storage costs, has pushed the duplicate image problem from a background nuisance to an active line item in departmental budgets.
How the Problem Accumulated
The roots go back to the early 2010s, when agencies like the Australian Public Service Commission on King Edward Terrace and the Department of Finance on One Canberra Avenue began migrating physical records and early digital archives into shared network drives. Staff saved the same campaign image, the same ministerial headshot, the same infrastructure photograph in multiple folders — sometimes across multiple systems — with no centralised tagging or asset management protocol in place.
When cloud migration accelerated after 2018, those duplicates moved with everything else. A single high-resolution photograph of, say, the National Arboretum Canberra or the Gungahlin town centre might exist in dozens of slightly different file names across SharePoint, an agency's content management system, and a legacy departmental intranet. Multiply that across hundreds of communication teams and you have an archive problem that grows faster than any single IT team can manually address.
The Australian National University's digital preservation unit, which manages its own substantial image archive on Acton Peninsula, identified the same structural issue in an internal review completed in late 2024. University of Canberra's library services on Kirinari Street have similarly flagged duplicate digital assets as a priority in their ongoing collections management work, though neither institution has publicly detailed the full scale of the problem.
The Mechanics of Duplicate Image Replacement
Duplicate image replacement — the process of identifying redundant copies, designating a single canonical version, and systematically replacing or deleting the rest — sounds straightforward. In practice, agencies are finding it labour-intensive without dedicated software. The challenge is that not all duplicates are exact pixel-for-pixel matches. A photograph cropped for a web banner and the same photograph saved at full resolution for print are technically different files. Perceptual hashing tools, which compare images by visual similarity rather than file size or name, are now the standard approach, but deploying them across an agency's full archive requires time, testing and staff training that smaller teams simply haven't had.
The Australian Signals Directorate's Australian Cyber Security Centre, located in Brindabella Business Park near Canberra Airport, has noted in its Essential Eight guidance framework that unmanaged file repositories — including image stores — represent a legitimate surface area for data governance failures, even if they are not a primary cyber threat vector. That framing has helped digital teams make the internal case for funding the cleanup.
Cost estimates vary widely depending on agency size, but cloud storage pricing from major providers means even modest duplicate reduction can translate to thousands of dollars in annual savings for a mid-sized department maintaining a five-to-ten terabyte archive. For larger agencies, the figure is proportionally higher.
Agencies working through their July 2026 digital asset audit obligations now face a practical set of choices: procure dedicated digital asset management software, contract the work to one of the Canberra-based IT firms that have built specialist practices around public sector data, or run the cleanup manually using in-house staff. The last option is increasingly being rejected as unscalable. Whatever path they choose, the underlying lesson is simple — storing digital assets without governance from day one creates compounding problems that eventually become impossible to ignore.