Methodology · Last updated Q2 2026
How we calculate every score on this platform.
We refuse to be a black box. Every metric you see — Survival Score, AI Resilience, Germany Opportunity, Salary Trajectory — is built from documented frameworks and public data. This page exists so you can interrogate our logic before you trust it.
Principles
Four rules we score by.
Task-decomposition over job titles
We never score "is your job at risk." We score the tasks inside your job. Following Acemoglu (MIT) and Brynjolfsson (Stanford), each role is decomposed into task types. Repetitive, template-driven, single-modal-output tasks are AI-substitutable. Strategic, judgment-driven, relational, and physically-embedded tasks resist substitution.
Composite over single-factor
No single number captures career safety. Survival is a weighted composite of four pillars: AI Resilience (35%), Germany Opportunity (30%), Transition Capacity (20%), Career Upside (15%). Weights reflect the relative durability of each factor over a 5-year horizon.
Directional, not predictive
A score of 72 does not mean a 72% probability of any specific outcome. It is a position estimate on a calibrated 0–100 scale relative to a peer cohort. Use it as a strategic compass, not a forecast.
Conservative on uncertainty
Where the labor-market data is thin or contested, we err toward the more pessimistic estimate. Better to flag a risk that does not materialize than miss one that does.
Composite formula
The Survival Score, fully decomposed.
// Composite Survival Score (0–100)
survival = (
0.35 × AI_Resilience // task-substitutability inverted
+ 0.30 × Germany_Opportunity // sector demand × language × visa
+ 0.20 × Transition_Capacity // adaptability × runway × velocity
+ 0.15 × Career_Upside // leverage × seniority × judgment
)
AI Resilience
35%
Inverted task-substitutability. Highest weight because AI exposure is the largest 5-yr structural force.
Germany Opportunity
30%
Sector demand × salary strength × English-friendliness × visa fit × credential recognition.
Transition Capacity
20%
Learning velocity × financial runway × language growth × adaptability index.
Career Upside
15%
Seniority × judgment density × specialization depth × AI-augmentation capacity.
Data sources
What we built this on.
Acemoglu (2024) — "The simple macroeconomics of AI"
Task-substitution model + productivity growth ceilings.
Brynjolfsson, Mitchell, Rock — task-level AI exposure scores
Per-occupation AI-exposure rankings (anchoring our resilience axis).
OECD Future of Work reports (2023–2025)
Cross-country labor displacement baselines.
Bundesagentur für Arbeit (BA) labour-market vacancy data
Germany sector demand index + Mangelberufe weighting.
Statistisches Bundesamt (Destatis) wage statistics
Salary baselines + inflation-adjusted real wage trajectories.
Eurostat structural labour indicators
Cross-EU benchmarking and sector-growth signals.
Goethe-Institut + BAMF integration outcome data
Language-fit weighting in the Germany Opportunity pillar.
Disclaimers
What this is — and what it isn't.
This is not financial, legal, or immigration advice. Visa eligibility, recognition of credentials, and tax treatment depend on individual circumstances and applicable law. Always consult a qualified professional before acting on a major career or relocation decision.
Forecasts are scenario projections, not predictions. The labor market is shaped by policy shifts, technology jumps, and macroeconomic shocks we cannot fully anticipate. Treat 5-year salary projections and AI-substitution timelines as scenario models, not certainties.
Cohort benchmarks are calibrated, not census-grade. Peer percentile estimates are calibrated against our test panel and published labor-market distributions. They become more accurate as our user base grows. We will publish methodology updates each quarter.
No score replaces judgment. The Survival Score is a starting point for strategic conversations, not an oracle. The most useful number on this platform is the one that sparks the right next question.
Found an error or want to challenge a weighting?
We publish quarterly methodology revisions and welcome rigorous pushback. Email the lab directly.
research@workingermany.io →