Zurich's Digital Archives Are Drowning in Duplicate Images — The Numbers Tell a Costly Story

Zurich's public institutions collectively store tens of thousands of duplicate image files across their digital archives — a problem that costs money, slows retrieval systems, and has quietly become a line item in IT budget reviews across the city. According to general estimates published by European digital asset management bodies, duplicate and near-duplicate images typically account for between 20 and 40 percent of total image storage in large public-sector organisations. For a city the size of Zurich, with its layered bureaucracies, research universities, and pharmaceutical affiliates, that translates into a significant and measurable drag on infrastructure.

The issue has gained fresh urgency in 2026. Storage costs, while falling on a per-gigabyte basis, are rising in aggregate as institutions digitise historical records, expand surveillance infrastructure, and migrate legacy systems into cloud environments. For Zurich, which is simultaneously managing the fallout from the UBS-Credit Suisse merger's administrative consolidation and a tight municipal housing budget strained by the Wohnungsnot crisis, every unnecessary server rack matters.

Where the Redundancy Lives

At ETH Zurich, the federal polytechnic on Rämistrasse, research groups routinely accumulate image datasets that overlap across projects and departments. A single multi-year climate modelling project can generate satellite image archives in the low terabytes, with duplicates introduced every time a collaborator shares a dataset via a different transfer protocol or file-naming convention. ETH's IT services division has acknowledged in its publicly available annual infrastructure reports that data deduplication is an ongoing operational priority, though the institution has not published a precise figure for image-specific redundancy.

The Stadtarchiv Zürich, located at Neumarkt 4 in the Altstadt, faces a different version of the same problem. Its digitisation programme — which has been converting physical photographic collections into searchable digital formats since at least 2018 — has produced large image libraries where near-duplicate scans of the same photograph at different resolutions or under different archivists occupy separate storage allocations. City archivists have discussed the challenge at professional conferences of the Schweizerische Gesellschaft für Geschichte, noting that automated deduplication tools flag images as distinct when pixel-level differences exist, even if the content is functionally identical.

The Zurich Cantonal Hospital, the Universitätsspital on Rämistrasse, operates one of Switzerland's largest medical imaging databases. Radiology archives alone run into the millions of files. Hospital IT administrators in Switzerland broadly — per a 2024 report by eHealth Suisse, the federal coordination body — estimated that storage inefficiencies including duplication add between 8 and 15 percent to annual digital infrastructure costs across the Swiss hospital sector.

The Fix Is Technical, the Barrier Is Human

Automated deduplication software has existed for years. Tools using perceptual hashing — a technique that identifies visually identical or near-identical images regardless of file name or format — can process a one-terabyte image archive in under two hours on standard enterprise hardware. The commercial licensing cost for mid-tier solutions runs roughly between CHF 3,000 and CHF 12,000 annually depending on archive size, according to published price sheets from vendors active in the Swiss market.

The real obstacle is institutional. Deduplication requires someone to decide which version of a duplicate to keep, and in organisations governed by Swiss public records law — the Öffentlichkeitsgesetz at federal level and its cantonal equivalents — deletion of any file, even a redundant one, can require documented authorisation. At the Stadtarchiv, that means a formal review process that archivists must initiate manually.

Several Zurich institutions are now piloting integrated digital asset management platforms that build deduplication logic into the upload workflow itself, catching redundancies before they enter the archive rather than hunting for them after the fact. The Zentralbibliothek Zürich on Zähringerplatz began testing one such system in early 2026 as part of a broader digitisation grant funded through the Swiss National Science Foundation's infrastructure programme.

For city residents, the practical stakes are less visible but real. When the city's open-data portal, opendata.swiss, responds slowly to image requests, or when a planning document arrives from the Stadtentwicklung Zürich office with broken image links, duplicate and mismanaged files are often a contributing factor. Getting the numbers under control is less a technical triumph than a municipal housekeeping task — one that Zurich's data managers say is already overdue.