By the Numbers: Canberra's Duplicate Image Problem Is Costing Government Websites More Than Anyone Admits

Across ACT government websites, thousands of duplicate images are sitting in content management systems, consuming server storage, slowing page load times, and inflating digital infrastructure costs that ultimately land on the desk of territory taxpayers. An audit framework discussion circulated among Canberra-based digital communications teams earlier this year puts the scale of the problem in sharp relief: a single mid-size government agency can accumulate more than 40,000 image assets over a five-year content cycle, with duplication rates running as high as 35 percent of total file volume.

That figure matters now because the ACT government is mid-way through a broader digital transformation program, consolidating agency websites onto unified platforms ahead of a 2027 target date. When agencies migrate legacy content, duplicate images don't just travel with it — they compound. Every redundant JPEG or PNG file that crosses into a new content management system adds to storage costs, slows automated indexing tools, and creates accessibility compliance headaches when alt-text has been applied inconsistently across copies of the same image.

What the Numbers Actually Look Like on the Ground

At the Australian National University in Acton, the library's digital collections team has dealt with this problem at institutional scale. ANU's library holdings include digitised photograph archives running into the millions of files, and the university's cataloguing teams have spent years developing deduplication protocols that larger government agencies are only beginning to consider. The University of Canberra, on the Bruce campus in Belconnen, has similarly invested in digital asset management tooling as part of its research data infrastructure — tools that automatically flag identical or near-identical image files before they enter long-term storage.

For ACT government agencies operating out of premises along London Circuit and Civic's surrounding office precincts, the practical picture is messier. Communications staff at multiple directorates routinely upload new versions of the same ministerial headshot, event photograph, or infrastructure render each time a document is refreshed, without checking whether a copy already exists. Over a standard four-year election cycle, a medium-sized directorate website can accumulate between 8,000 and 15,000 image files. If the industry-cited duplication rate of roughly one-in-three files holds, that means anywhere from 2,600 to 5,000 redundant images sitting in live storage at any given time.

Storage costs for cloud-hosted government systems are not trivial. Commercially, object storage for government-grade environments typically runs between $0.023 and $0.045 per gigabyte per month depending on redundancy tier. A library of 15,000 unoptimised image files — many of them uncompressed TIFFs or high-resolution PNGs — can occupy 200 gigabytes or more. Multiply that across a dozen directorates and the annual bill for storing images that nobody needs, many of which are exact copies of something already in the system, climbs into tens of thousands of dollars before bandwidth and retrieval costs are added.

The Fix Is Technical, but the Barrier Is Behavioural

Digital asset management specialists point to a straightforward two-part solution: implement hash-based deduplication at the upload stage so the system automatically rejects an identical file, and run a one-time retrospective audit to clean existing libraries before migration. Several commercial platforms already include this functionality, and open-source tools such as those built around the IPTC metadata standards allow organisations to match images by content rather than just filename.

The ACT's Digital Strategy 2025–2028, which covers the consolidation work currently underway, flags data quality as a priority area, though its published documentation does not set specific targets for image asset rationalisation. That gap leaves individual directorates to solve the problem — or ignore it — on their own timelines.

For agencies preparing to migrate content to new platforms in the next 12 to 18 months, the practical advice from digital archivists is consistent: audit before you move, not after. Running a deduplication pass on a 15,000-file library before migration takes hours. Running it on a merged library from six agencies post-migration takes weeks, and the errors — broken links, missing alt-text, accessibility failures — tend to surface at the worst possible time, usually just after a new platform has gone live.