Skills Demand-Supply Intensity Index™ (SDSI)
Measuring how tight a skill-specific labor market is relative to the overall economy.
1. Overview
The Intellectica Skills Demand-Supply Intensity Index™ (SDSI Index) measures how tight the labor market is for a specific skill category in a country, both in absolute terms and relative to the overall market. It provides business leaders and HR decision-makers with a clear, comparable metric that quantifies talent scarcity across skill domains.
The index answers a single question: how does demand-supply tension for a specific skill category compare to the overall labor market? It produces a percentage-based score anchored at 100%, where 100% represents market-average tension. Values above 100% indicate a tighter-than-average skill market; values below 100% indicate a looser-than-average one.
The SDSI Index is designed as a cross-sectional framework: 12 skill baskets are tracked simultaneously within a country, covering technology, business functions, design, and professional services. This allows direct comparison of relative talent tension across domains from the very first release. The time series builds forward with each monthly publication.
1.1 Why this index exists
Existing labor market indicators tend to be either too broad (economy-wide unemployment rates) or too narrow (individual company hiring metrics). Business leaders making talent strategy decisions need a middle layer: skill-specific demand-supply signals that can be compared across domains and tracked over time. The SDSI Index fills this gap by translating LinkedIn Talent Insights data into a standardized, interpretable metric.
1.2 Audience
The primary audience is business leaders and HR decision-makers — professionals asking questions such as: Should we be concerned about finding AI talent in Greece? Which skill domains are experiencing the greatest hiring pressure? Is the talent scarcity we feel in our sector real, or is the entire market tight? The SDSI Index provides the quantitative basis for these conversations.
2. Data Sources
The SDSI Index uses a single data source to maintain methodological simplicity and eliminate cross-source comparability issues.
2.1 LinkedIn Talent Insights
LinkedIn Talent Insights is a proprietary analytics platform providing real-time data on the professional workforce. It is the sole data source for the SDSI Index.
- Update frequency. Near real-time — professional profiles and job postings update continuously.
- Access type. Paid subscription.
- Historical depth. None. LinkedIn Talent Insights provides current-state snapshots only; there is no way to query historical values for past dates. The index begins live from launch.
- Known limitations. LinkedIn over-represents white-collar, urban, tech-adjacent professionals. Coverage is strong for knowledge-work domains but structurally incomplete for manual trades, the informal economy, and non-digital sectors.
3. Components & Dimensions
The SDSI Index is a measurement framework applied across 12 skill baskets spanning technology, business functions, design, and professional services. Each basket represents a distinct professional domain. The framework produces a SDSI score for each basket, enabling cross-sectional comparison.
Two core metrics are calculated per basket: the Tightness Ratio (absolute demand-supply balance) and the Skills Demand-Supply Intensity score (relative to the overall market).
3.1 Skill basket construction principles
Each skill basket is a curated list of LinkedIn skill tags representing a professional domain. The following principles govern basket construction:
- Domain essentiality. A skill belongs in a basket if it is essential to working in that domain, even if it is also used in other fields. The test: is this skill a core part of the practitioner's toolkit in this domain? For example, Python is not exclusively an AI skill, but it is essential to virtually all AI work. A skill that is merely incidentally associated with the domain does not qualify.
- Taxonomy-level consistency. Skills should be selected at mid-level specificity — not umbrella terms (e.g., not "Artificial Intelligence" as a catch-all) and not niche techniques (e.g., not "Hyperparameter Tuning"). Preferred level: Machine Learning, Deep Learning, Computer Vision, NLP, Generative AI.
- Dominance monitoring. If a single skill accounts for more than 40–50% of the basket's total professional count when queried independently, it is flagged as a dominant skill. Dominant skills are not automatically excluded if they pass the domain-essentiality test, but they are monitored at each release. The SDSI is tracked in parallel with and without the dominant skill; if the two values diverge significantly over time, the skill's inclusion is reconsidered.
- Basket size target: 10 ± 3 skills (7–13 per basket). Fewer than 7 creates fragility; more than 13 dilutes toward the general market. Not every domain will hit the same number. All baskets are documented with their exact skill lists.
- Longevity over trendiness. Prefer skills that will remain meaningful for 5+ years. Emerging skills may be included but are flagged as candidates for the 6-month review cycle. Brand-name skills (e.g., AWS, Salesforce, Figma) are included only when they function as de facto industry categories with 10+ years of stability.
- Selection-time checklist. Each skill is validated at selection time using five questions: Does it pass domain essentiality? Is it at mid-level specificity? Does it show a non-trivial professional count in the target country? Does it trigger dominance monitoring? Is it likely relevant in 5+ years? All five must be assessed for inclusion.
3.2 Basket overview
The SDSI Index tracks 12 skill baskets covering 121 skills across the knowledge-work economy:
| Basket | Skills | Domain Layers |
|---|---|---|
| AI & Data Science | 13 | Core AI/ML · Domain · Generative AI · ML Frameworks · ML Infrastructure · Essential Tooling · Essential Analytical |
| Cybersecurity | 10 | Core Defense · Offensive Security · Operations · Cloud Layer · Access Control · Threat Analysis · Security Tooling |
| Cloud Computing & DevOps | 10 | Core · Cloud Platforms · Methodology · Container Orchestration · Containerization · IaC · Deployment Pipelines · Automation |
| Software Engineering | 11 | Core · Essential Language · Frontend Framework · Backend Runtime · Design · Integration · Version Control · Essential Tooling |
| Digital Marketing | 11 | Core · Search · Social · Content · Analytics · Automation · Channels · Paid Media · Strategy |
| Finance & Accounting | 10 | Core · Analysis · Reporting · Assurance · Planning · Risk · Strategy · Operations · Regulatory |
| Human Resources & Talent Management | 10 | Core · Development · People Operations · Total Rewards · Strategy · Culture · Essential Tooling |
| Sales & Business Development | 10 | Core · Client Relations · Essential Tooling · Pipeline · Sales Model · Deal Skills · Strategy · Operations |
| Project Management | 9 | Core · Methodology · Portfolio · Communication · Planning · Transformation · Operations · Execution |
| Supply Chain & Logistics | 10 | Core · Sourcing · Warehouse · Planning · Storage · Analytics · Supplier Relations · Distribution · Purchasing |
| UX/UI & Product Design | 9 | Core · Research · Design · Prototyping · Testing · Structure · Essential Tooling |
| Legal & Compliance | 9 | Core · Specialization · Compliance · Privacy Regulation · Transactions · Practice · Operations |
3.3 AI & Data Science (13 skills)
| Skill Tag | Domain Layer |
|---|---|
| Machine Learning | Core AI/ML |
| Deep Learning | Core AI/ML |
| Neural Networks | Core AI/ML |
| Natural Language Processing (NLP) | Domain — Language |
| Computer Vision | Domain — Vision |
| Generative AI | Generative AI |
| Large Language Models (LLM) | Generative AI |
| Retrieval-Augmented Generation (RAG) | Generative AI — flagged for review |
| TensorFlow | ML Frameworks |
| PyTorch | ML Frameworks |
| MLOps | ML Infrastructure |
| Python (Programming Language) | Essential Tooling — dominant skill |
| Data Science | Essential Analytical |
Python (Programming Language) accounts for approximately 78% of the basket population when queried independently. Retained under the domain-essentiality principle; monitored with-and-without at each release. RAG flagged for 6-month longevity review. Excluded: Artificial Intelligence (AI) umbrella tag, brand-specific skills (ChatGPT, GPT-4, DALL-E, LangChain), and insufficiently specific skills (Image Processing, OpenCV, Prompt Engineering).
3.4 Cybersecurity (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Network Security | Core Defense |
| Information Security | Core Defense |
| Cybersecurity | Core Defense |
| Penetration Testing | Offensive Security |
| Vulnerability Assessment | Offensive Security |
| Incident Response | Operations |
| Cloud Security | Cloud Layer |
| Identity & Access Management (IAM) | Access Control |
| Threat Intelligence | Threat Analysis |
| SIEM | Security Tooling |
Minimal cross-basket overlap expected. Cloud Security may capture some professionals also in the Cloud Computing & DevOps basket.
3.5 Cloud Computing & DevOps (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Cloud Computing | Core |
| Amazon Web Services (AWS) | Cloud Platforms |
| Microsoft Azure | Cloud Platforms |
| Google Cloud Platform (GCP) | Cloud Platforms |
| DevOps | Methodology |
| Kubernetes | Container Orchestration |
| Docker | Containerization |
| Terraform | Infrastructure as Code |
| CI/CD | Deployment Pipelines |
| Infrastructure as Code | Automation |
AWS, Azure, and GCP are brand names but function as stable skill categories on LinkedIn (10+ years established). They collectively define the cloud market and pass the longevity test.
3.6 Software Engineering (11 skills)
| Skill Tag | Domain Layer |
|---|---|
| Software Development | Core |
| JavaScript | Essential Language |
| Java | Essential Language |
| TypeScript | Essential Language |
| React.js | Frontend Framework |
| Node.js | Backend Runtime |
| Web Development | Core |
| Software Architecture | Design |
| RESTful APIs | Integration |
| Git | Version Control |
| SQL | Essential Tooling |
SQL also appears in data contexts but is essential to virtually all software engineering work (domain-essentiality principle). JavaScript likely triggers dominance monitoring given its prevalence across the web development population.
3.7 Digital Marketing (11 skills)
| Skill Tag | Domain Layer |
|---|---|
| Digital Marketing | Core |
| Search Engine Optimization (SEO) | Search |
| Search Engine Marketing (SEM) | Search |
| Social Media Marketing | Social |
| Content Marketing | Content |
| Content Strategy | Content |
| Google Analytics | Analytics |
| Marketing Automation | Automation |
| Email Marketing | Channels |
| Pay-Per-Click (PPC) | Paid Media |
| Marketing Strategy | Strategy |
Google Analytics is a brand name but has been the industry standard for 15+ years and functions as a skill category on LinkedIn. Digital Marketing as a tag may trigger dominance monitoring.
3.8 Finance & Accounting (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Financial Analysis | Core |
| Accounting | Core |
| Financial Modeling | Analysis |
| Financial Reporting | Reporting |
| Auditing | Assurance |
| Budgeting | Planning |
| Risk Management | Risk |
| Corporate Finance | Strategy |
| Management Accounting | Operations |
| Taxation | Regulatory |
Risk Management also appears in other professional contexts (e.g., project management, compliance). Cross-basket overlap is expected and documented.
3.9 Human Resources & Talent Management (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Human Resource Management | Core |
| Learning & Development | Development |
| Employee Relations | People Operations |
| Organizational Development | Development |
| Performance Management | Development |
| Compensation & Benefits | Total Rewards |
| Talent Management | Strategy |
| Workforce Planning | Strategy |
| Employee Engagement | Culture |
| HRIS | Essential Tooling |
Recruiting and Talent Acquisition were excluded — they function as demand-side umbrella tags in HR job postings, inflating job-post counts across the entire HR domain. Empirical testing confirmed this: their inclusion produced a severely distorted SDSI driven almost entirely by recruiter demand rather than HR market tension. The umbrella tag Human Resources (HR) was also excluded on the same grounds.
3.10 Sales & Business Development (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Sales Management | Core |
| Business Development | Core |
| Account Management | Client Relations |
| CRM | Essential Tooling |
| Salesforce.com | Essential Tooling |
| Lead Generation | Pipeline |
| B2B Sales | Sales Model |
| Negotiation | Deal Skills |
| Sales Strategy | Strategy |
| Pipeline Management | Operations |
Salesforce.com is a brand name but has been the dominant CRM platform for 20+ years, functioning as a skill category. Negotiation may overlap with the Legal & Compliance basket.
3.11 Project Management (9 skills)
| Skill Tag | Domain Layer |
|---|---|
| Project Management | Core |
| Agile Methodologies | Methodology |
| Scrum | Methodology |
| Program Management | Portfolio |
| Stakeholder Management | Communication |
| Project Planning | Planning |
| Change Management | Transformation |
| Business Process Improvement | Operations |
| Cross-functional Team Leadership | Execution |
Project Management as a tag is very widely listed and likely triggers dominance monitoring. Agile Methodologies and Scrum may overlap with Software Engineering professionals.
3.12 Supply Chain & Logistics (10 skills)
| Skill Tag | Domain Layer |
|---|---|
| Supply Chain Management | Core |
| Logistics Management | Core |
| Procurement | Sourcing |
| Inventory Management | Warehouse |
| Demand Planning | Planning |
| Warehouse Management | Storage |
| Supply Chain Optimization | Analytics |
| Vendor Management | Supplier Relations |
| Transportation Management | Distribution |
| Sourcing | Purchasing |
Especially relevant for Greece given its shipping and trade economy. LinkedIn coverage is reasonable for mid-to-senior supply-chain professionals but may underrepresent operational warehouse staff.
3.13 UX/UI & Product Design (9 skills)
| Skill Tag | Domain Layer |
|---|---|
| User Experience (UX) | Core |
| User Interface Design | Core |
| UX Research | Research |
| Product Design | Design |
| Interaction Design | Design |
| Wireframing | Prototyping |
| Usability Testing | Testing |
| Information Architecture | Structure |
| Figma | Essential Tooling — flagged for review |
Figma is a brand but has become the de facto industry tool. Flagged for 6-month review given potential displacement risk. At 9 skills, this is the smallest basket in the framework.
3.14 Legal & Compliance (9 skills)
| Skill Tag | Domain Layer |
|---|---|
| Legal Research | Core |
| Corporate Law | Specialization |
| Contract Law | Specialization |
| Regulatory Compliance | Compliance |
| GDPR | Privacy Regulation |
| Due Diligence | Transactions |
| Legal Writing | Practice |
| Intellectual Property | Specialization |
| Contract Management | Operations |
GDPR is regulation-specific but is the defining EU privacy framework with 10+ year expected stability. This basket carries the highest thin-signal risk in smaller markets; per the methodology, any basket with fewer than 50 job posts in a country is reported with an explicit low-confidence flag.
4. Calculation Formula
4.1 Step 1 — Tightness Ratio (TR)
The Tightness Ratio measures the absolute demand-supply balance for a specific skill basket in a specific country. It represents the number of active job postings per professional in that skill domain.
TR(s) is the Tightness Ratio for skill basket s; JP(s) is the total active job posts matching at least one skill in basket s; TP(s) is the total professionals with at least one skill from basket s.JP(AI&DS) = 1,105 job posts; TP(AI&DS) = 64,865 professionals
Interpretation: there are approximately 1.7 job postings for every 100 AI & Data Science professionals in Greece — roughly one job post per 59 professionals.
4.2 Step 2 — Market Tightness Ratio (TR-Market)
The same calculation applied to the entire labor market in the country, with no skill filter. This serves as the universal benchmark.
JP(m) = 17,830 job posts; TP(m) = 1,773,314 professionals
4.3 Step 3 — Skills Demand-Supply Intensity (SDSI)
The SDSI expresses how tight a skill market is relative to the overall market, indexed to 100%. The overall market always equals 100%.
Interpretation. The AI & Data Science skill market in Greece is 1.70× tighter than the overall labor market. For every job post per professional in the general market, there are 1.70 job posts per professional in AI & Data Science.
4.4 Rounding convention
All published SDSI values are rounded to whole percentages (e.g., 170%, not 169.5%). Tightness Ratios are reported as percentages rounded to one decimal place (e.g., 1.7%). Decimal places beyond these thresholds imply false precision given that LinkedIn counts fluctuate daily.
4.5 Score interpretation
The SDSI score is unbounded — it can range from near 0% to well above 250%. The following interpretation bands provide guidance for readers:
| SDSI Range | Interpretation |
|---|---|
| Below 50% | Significantly looser than market — oversupplied skill domain |
| 50–80% | Moderately looser than market |
| 80–120% | Near market average |
| 120–180% | Moderately tighter than market |
| 180–250% | Significantly tighter — talent scarcity signal |
| Above 250% | Acute tension — severe demand-supply imbalance |
These bands are initial estimates based on the first release. They will be reviewed and potentially adjusted after 6 months of cross-sectional data across multiple skill baskets.
5. Anchor / Baseline Protocol
5.1 Self-anchoring design
Unlike indices built on relative data sources (e.g., Google Trends), the SDSI Index is self-anchoring. The market-wide Tightness Ratio serves as the baseline and is recalculated fresh with every monthly release. A SDSI of 170% means 1.70× the market tightness right now, not relative to a historical reference period.
This eliminates several common methodological challenges: there is no anchor drift, no re-anchoring events, and no scale-factor instability. The index is always expressed in current-market terms.
5.2 Implications
Because the baseline is recalculated each period, month-over-month SDSI changes reflect shifts in relative position, not absolute levels. A skill basket's SDSI could decrease even if its absolute Tightness Ratio increased, provided the overall market tightened faster. The auxiliary Tightness Ratio (reported separately) provides the absolute context needed to distinguish these scenarios.
6. Auxiliary Metrics
6.1 Tightness Ratio (TR)
The Tightness Ratio is reported alongside each skill basket's SDSI score as an auxiliary metric. While the SDSI captures relative market position, the TR provides the absolute demand-supply signal.
The combination of SDSI and TR enables a two-dimensional interpretation:
| High SDSI (> 120%) | Low SDSI (< 80%) | |
|---|---|---|
| High TR | Genuine talent crunch Tight market, tighter than average — the strongest scarcity signal. | Broad-market tightness Overall market is tight; the skill is not specifically scarce. |
| Low TR | Relative scarcity in a soft market Skill stands out despite loose overall conditions. | Unremarkable Oversupplied skill in a loose market. |
6.2 Market Tightness Ratio (TR-Market)
The overall market Tightness Ratio is reported as a standalone metric to provide macroeconomic context. It represents the economy-wide job-to-professional ratio and serves as the denominator for all SDSI calculations.
7. Data Collection
Each monthly release is built from a single LinkedIn Talent Insights snapshot per country, taken on a fixed collection date within the publication month. The same date applies to every basket in that country-month, so all values within a release are mutually consistent.
Two categories of measurement are recorded for each snapshot:
- Market baseline. Total Professionals (
TP-market) and Total Job Posts (JP-market) for the country with no skill filter applied. Collected once per country per month; serves as the denominator for every SDSI calculation in the release. - Per-basket measurement. For each of the 12 skill baskets, Total Professionals (
TP(s)) and Total Job Posts (JP(s)) matching at least one skill in the basket. The OR-logic filter convention is described in Section 9.
From these inputs, the formulas in Section 4 yield TR(s), TR(m), and SDSI(s). All raw counts and derived values are archived alongside their collection date to build the forward-looking time series; recalculation of past values is never performed retroactively, since LinkedIn Talent Insights does not support historical queries.
8. Publication Format
8.1 Release components
- Headline ranking table: all 12 skill baskets sorted by SDSI, with the SDSI value and a plain-language label from the interpretation bands
- Auxiliary Tightness Ratios alongside each basket
- Market Tightness Ratio as macroeconomic context
- Trend chart showing SDSI over time per basket (from month 3 onward)
- Intellectica commentary: 2–3 sentences highlighting the most notable movements or comparisons
8.2 Publication channels
The SDSI Index is published via an interactive dashboard on the Intellectica website, with supporting distribution via LinkedIn posts and an email digest.
8.3 Cadence
Monthly. Each release is published within the first week of the calendar month and reports the previous month's snapshot.
9. Limitations & Considerations
- LinkedIn coverage bias. LinkedIn over-represents white-collar, urban, tech-adjacent professionals. SDSI values for knowledge-work skill baskets are reliable. For manual trades, hospitality, construction, or informal-economy skill domains, the underlying data is structurally incomplete. This defines the scope boundary of the index.
- Snapshot, not flow. LinkedIn Talent Insights shows current stock of professionals and current open postings. It does not capture hiring velocity, time-to-fill, offer acceptance rates, or salary pressure. The SDSI measures structural tension, not hiring outcomes.
- No historical baseline. LinkedIn Talent Insights provides current-state snapshots only. There is no way to query historical values for past dates. The index begins live from its launch date. No backtested values are produced or implied. The time series builds forward.
- OR-logic skill filters. LinkedIn's skill filter uses OR logic — a professional with any one skill from the basket is counted. Broadly-skilled professionals therefore appear in multiple baskets simultaneously. Cross-basket SDSI values are not fully independent; they share talent-pool overlap.
- Domain-essential skills inflate overlap. Skills included under the domain-essentiality principle (e.g., Python, SQL, Negotiation) are shared across multiple domains. This amplifies cross-basket talent-pool overlap. The effect is documented and monitored but does not invalidate within-basket SDSI calculations.
- Daily fluctuation in counts. LinkedIn's professional and job-posting counts shift daily as profiles and postings are added or removed. Monthly snapshots mitigate this, but a single collection date introduces minor timing sensitivity.
- Sensitivity to overall market conditions. Because the SDSI is benchmarked to the entire market, it is sensitive to macroeconomic shifts. If LinkedIn job-posting volume surges across all sectors, all skill-specific SDSI values will shift even if nothing changed in those specific skill markets. The auxiliary TR provides the absolute context needed to identify this effect.
- Interpretation bands are initial estimates. The score-interpretation table is based on the first release and limited cross-sectional data. Bands will be reviewed and potentially recalibrated after 6 months of operation across all 12 skill baskets.
- Thin-signal risk for certain baskets. Legal & Compliance and UX/UI & Product Design may have limited professional and job-posting counts in smaller markets. If a basket produces fewer than 50 job posts in a country, the SDSI is reported with an explicit low-confidence flag.
10. Versioning & Updates
10.1 Versioning convention
v1.x (minor versions). Adding or removing individual skills within an existing basket, adding a new skill basket, documentation improvements, cosmetic refinements. No structural impact on the SDSI formula or interpretation framework.
v2.0+ (major versions). Changes to the SDSI formula, modification of interpretation bands, changes to the market baseline definition, changes in publication frequency, fundamental redefinition of basket construction principles.
10.2 Component updates
Skill baskets are reviewed every 6 months. Triggers for component changes:
- A skill shows decaying professional count or negligible job-posting activity over multiple consecutive periods
- A new skill emerges with strong, continuous signal that fits an existing basket's domain
- Technological or market shifts make a skill obsolete or fundamentally change its domain association
- A dominant skill's with-and-without SDSI values diverge significantly, indicating distortive effect
10.3 Adding new skill baskets
Adding a new skill basket is a minor version event (v1.x) since it does not affect existing baskets or the SDSI formula. The new basket is simply added to the cross-sectional ranking from its first month of publication. No phased inclusion is required because each basket is independently calculated.
10.4 Major-version communication template
"Starting this [month], the Skills Demand-Supply Intensity Index methodology has been updated. Specifically: [list of changes]. Previous periods have been recalculated under the new methodology where applicable. Full methodology change notes are available at [link]."
11. Roadmap
- Cross-country expansion. Apply the framework to additional countries with their own market baselines, enabling comparison of skill-specific tension across geographies.
- Interpretation-band recalibration. After 6 months of cross-sectional data, reassess the SDSI interpretation thresholds based on observed distributions across all 12 baskets.
- Sub-basket analysis. Within large skill domains, explore sub-basket breakdowns (e.g., within AI & Data Science: Core ML vs. Generative AI vs. ML Infrastructure).
- Additional baskets. Evaluate Green Energy & Sustainability, Data Engineering, and other domains as LinkedIn coverage and signal warrant.
- Integration with Intellectica AI Pulse Index. Explore complementarity between the SDSI Index (talent-market tension) and AI Pulse Index (broader AI adoption and discourse signals).