evidoria

← Back to browse

Good practice Imported

OSKA — Estonia's AI-Assisted National Skills-Forecasting System for Curriculum Planning

Estonia · Tallinn · See the Estonia profile · See the Tallinn profile

Evidence: Observational / pre–post Top 24% 60/100 · Ask Evidence Copilot about this practice

Estonia's Qualifications Authority runs OSKA, forecasting national skills since 2015 to guide VET/HE curriculum planning. AI partner Headai used NLP to mine 10,876 job ads against six mechatronics curricula, mapping real skills gaps.

10,876 job ads
Job advertisements analysed in the Headai mechatronics NLP pilot
6 curricula
Vocational/university curricula compared against job-ad demand
about one third
Estonian jobs requiring vocational education
about one quarter
VET graduates' share of the labour force
OSKA — Estonia's AI-Assisted National Skills-Forecasting System for Curriculum Planning

Details

Maturity
Established
Promoter
Estonian Qualifications Authority (Kutsekoda), with Headai
Period
2015-present
Region (NUTS)
EE00
Keywords
vocational education, curriculum planning, labour market forecasting, skills intelligence, NLP/AI analytics

Context

OSKA is Estonia's national labour-market monitoring and future-skills forecasting system, run by the Estonian Qualifications Authority (Kutsekoda) since 2015 under an OSKA Coordination Council that includes multiple government ministries and social partners.

Objectives

Forecast skills needs across all major economic sectors on a rolling five-to-ten-year horizon to guide VET and higher-education curriculum planning, sharpened by AI-based analysis of job-market data.

Activities

The Estonian Qualifications Authority worked with AI company Headai, which used NLP and 'Cognitive AI' techniques to machine-read 10,876 Estonian job advertisements alongside six vocational and university mechatronics curricula, producing visual skills maps comparing employer demand with existing course content.

Results

An external evaluation cited by Cedefop found OSKA 'successful in creating practical and usable labour market intelligence for its target groups'; five years in, OSKA's own research found that about a third of Estonian jobs require vocational education while VET graduates make up only about a quarter of the labour force.

Conclusions

Neither source quantifies how many curricula have actually been revised as a direct result, so evidence stops at forecasting quality and system uptake rather than proven downstream outcomes.

Implementation
Indicative cost
Medium (€50k–€500k) — No published budget; an ongoing national government-run monitoring function with an external AI vendor engagement (Headai) documented for one sectoral pilot.
Time to results
Long (> 3 years) — Operating continuously since 2015 on a rolling five-to-ten-year forecasting horizon; still active as of the most recent evaluation.
Staffing & skills
Estonian Qualifications Authority (Kutsekoda), OSKA Coordination Council (multiple government ministries and social partners), AI partner Headai (NLP/Cognitive AI)

Conditions for success

  • A multi-ministry coordination council anchoring the system across government and social partners
  • An AI partner supplying NLP-based labour-market text mining at national scale

Common failure modes

  • No quantification of how many curricula are actually revised as a result, so demonstrated uptake beyond intelligence-production is unclear

Commonly funded by

National / regional programmes ESF+ — European Social Fund Plus

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.

Attachments

Similar practices you may find useful