Zurich's Duplicate Image Crisis: What Happens Next and the Key Decisions Ahead

Zurich's municipal digitisation programme has a growing problem buried inside its servers. Thousands of duplicate image files — accumulated over years of parallel scanning drives, departmental uploads and legacy migrations — are clogging the city's central document management infrastructure, and administrators now face a concrete set of decisions about how to clean it up, who pays, and what legal obligations apply under Swiss data-retention law.

The issue crystallised earlier this year when the city's IT directorate, Departement der Informatik, completed an audit of the central archive repository used by offices across Stadthausquai and the civic administration cluster near Rathausbrücke. The audit found that a substantial share of stored image assets — including scanned permits, planning documents and heritage photographs — existed in two or more identical or near-identical copies, creating storage redundancies and, more critically, compliance ambiguities under the cantonal archiving ordinance.

Why the Timing Matters

Switzerland's federal archiving law sets strict rules on what must be kept, in what format, and for how long. Cantons and municipalities layer additional requirements on top. For Zurich, the pressure is not simply administrative tidiness: the city is mid-way through a broader e-government modernisation push, with several departments migrating to a unified platform by the end of 2026. Any duplicate image data that crosses into the new system will cost more to store, harder to audit, and potentially in conflict with data-minimisation principles under the revised Federal Act on Data Protection, which came into full effect in September 2023.

ETH Zurich's Digital Preservation Lab, based on Rämistrasse, has studied this class of problem across European municipal archives. Researchers there have documented that duplicate-rate inflation in unmanaged digital repositories typically ranges between 18 and 34 percent of total stored assets within a decade of initial digitisation — figures consistent with what Zurich's own audit is understood to have flagged internally. The cost per terabyte of government-grade, compliant cloud storage in Switzerland sits around CHF 80 to CHF 120 annually depending on redundancy tier, meaning even modest reductions in duplicate volume translate into meaningful budget savings over a five-year horizon.

Stadtarchiv Zürich, located on Neumarkt, has been involved in parallel conversations about how heritage and historical photograph collections — many digitised through the Zürich liest programme and related civic initiatives — intersect with the deduplication question. The risk is not trivial: aggressive automated deduplication tools can misidentify slightly different versions of a historical image as duplicates when they are, in fact, distinct documents with separate archival value.

The Decisions That Cannot Wait

Three choices now sit on the table for municipal IT leadership. First, whether to run a fully automated deduplication pass — fast and cheap but risky for nuanced heritage collections — or a hybrid workflow that flags suspected duplicates for human review. Second, which department absorbs the cost: central IT or the originating units that generated the redundancies. Third, how the city communicates its data-handling decisions to the public, given Zurich's tradition of transparency under the cantonal information law.

A hybrid review model would likely require contracting external digital archivists for an estimated three-to-six-month engagement. That timeline is tight against the end-of-2026 platform migration deadline. Postponing the migration is an option, but it would delay efficiency gains that departmental budgets are already anticipating.

Civic technology advocates in the city, including groups that have engaged with past open-data debates at the Zurich main library on Zähringerplatz, argue that whatever methodology is chosen, the criteria and results should be published openly — consistent with the city's existing open-data portal. That demand adds another layer to an already crowded decision tree.

The coming weeks will likely see the IT directorate bring a formal recommendation to the Stadtrat. Once that lands, the political negotiation over budget allocation begins in earnest. For departments already watching their 2026 appropriations closely, the deduplication question is no longer a technical footnote — it is a line item waiting to be written.