evidoria

← Back to browse

Good practice Imported

Croatia's First National Web Accessibility Audit — No Public Site Cleared 50% Full Compliance

Croatia · Zagreb · See the Croatia profile · See the Zagreb profile

Evidence: Observational / pre–post Top 30% 88/100 · Ask Evidence Copilot about this practice

Croatia's Information Commissioner tested 158 public-sector homepages and 8 apps against 87–102 EN 301 549 requirements using disability-led testers: average scores ranged from 86% in health to 31% in transport, and no site fully met every requirement.

158
Public-sector homepages tested (2020-2021)
18
Sites tested in-depth (2021)
8
Mobile apps tested (2021)
86 %
Average compliance — health sector (highest) (2020-2021)
69 %
Average compliance — education/science/sport sector (lowest) (2020-2021)
31 %
Lowest individual-site score (transport) (2020-2021)
0
Websites fully meeting more than 50% of all tracked requirements (2020-2021)
95 %
Best-scoring mobile app (MerlinMobile) (2021)
58 %
Lowest in-depth score among tested bodies (Croatian Employment Service) (2021)

Details

Maturity
Established
Promoter
Office of the Information Commissioner of the Republic of Croatia (Ured povjerenika za informiranje)
Period
2020–2021
Region (NUTS)
HR05
Keywords
digital accessibility, government transparency, disability rights

Context

Croatia's 2019 Accessibility Act transposed EU Directive 2016/2102, giving the independent Office of the Information Commissioner legal responsibility for monitoring, inspecting and reporting on public-sector website and app accessibility, working alongside the government's central digital-transformation office (SDURDD).

Objectives

To produce Croatia's first statutory, independently-tested picture of public-sector digital accessibility using disability-led manual evaluation rather than automated tools alone.

Activities

For the first statutory monitoring cycle (websites: 1 January 2020 – 22 December 2021; apps: tested September–December 2021), external evaluators from national disability umbrella organisations — including testers with motor and vision impairments, the latter using the NVDA screen reader — manually checked 158 public-body homepages against 87 EN 301 549 requirements, plus 18 sites and 8 apps in further in-depth detail against 87 and 102 requirements respectively.

Results

Average accessibility varied sharply by sector: highest in health (86%) and social welfare (82%), lowest in education/science/sport (69%), with individual-site lows of 31% in transport and in environment/recreation/culture against individual highs of 90–100%. Regionally, Adriatic Croatia and Northern Croatia scored highest (81% and 80%), while the City of Zagreb and Pannonian Croatia lagged (71–74%). No monitored website satisfied more than 50% of all tracked requirements in full — the Commissioner attributes this partly to roughly half of all requirements being inapplicable to any given site, meaning genuine full compliance remains rare even where average scores look strong. Among apps, the University Computing Centre's MerlinMobile (95%) and the Health Ministry's CovidGO (87%) scored best, while ZOSI (62%), the City of Zagreb (60%) and the Croatian Employment Service (58%) scored lowest among in-depth-tested bodies.

Conclusions

The Commissioner has since run follow-up compliance checks (2020 and 2021) and paired the monitoring with public-sector staff training, showing the audit was designed to drive an ongoing improvement cycle rather than stand as a one-off report.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Long (> 3 years)
Staffing & skills
Office of the Information Commissioner (Ured povjerenika za informiranje), central digital-transformation office SDURDD, external evaluators from national disability umbrella organisations, including testers with motor and vision impairments

Conditions for success

  • A statutory mandate (2019 Accessibility Act) assigning independent inspection powers to a body separate from the audited agencies
  • Disability-led manual testing (not automated-only scanning) to check every one of 87-102 requirements per site
  • In-depth sub-sampling (18 sites, 8 apps) to go beyond homepage-level checks
  • Pairing monitoring with public-sector staff training and repeat compliance checks to drive improvement, not just measurement

Common failure modes

  • No monitored website satisfied more than 50% of all tracked requirements in full, even in the highest-scoring sectors
  • Sharp sectoral and regional gaps (31% to 100% at individual-site level) show uneven implementation
  • Some disability-facing bodies themselves (e.g. the disability employment agency ZOSI) scored among the lowest in in-depth testing

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful