Back to Experiments

Swedish Analyst Jobs Analysis

AnalyticsSep 4, 2025
Priority:
(Highest)
SQLTableauData Analytics

Result Summary

15,000+ job postings from Arbetsförmedlingen is analyzed to extract insights


Better Reading Experience Here.

 

🕵️‍♀️
My First Deep Dive into Data (And What I Learned About My Future Career)

 

Table of contents

Unsupported block type: table_of_contents

📝 TL;DR

How It Started

As I am getting into data analytics, I wanted to figure out: which analyst path should I go for?

Data Source

15,000+ analyst job postings from Arbetsförmedlingen (2016–2024).

Skills Demonstrated

SQL, Excel, Tableau, AI assisted Python, end-to-end project design.

Key Findings

  • Digital Analysts make up the largest share of postings (38.5%), followed by Business Analysts (29.9%) and Data Analysts (15%).
  • The job market exploded during COVID recovery (a 100% surge in 2021-2022), but it cooled off in 2023–2024 as the market normalized and the economy weakened.
  • Half of all opportunities are in Stockholm, but Västra Götaland offers a stronger concentration of Business Analyst roles.
  • SQL is the most in-demand technical skill, and Looker and Python show the fastest growth.

  • 1. ❓ Introduction: Why This Project

    The Motivation

    When I first got into data analytics, I realized there are tons of different analyst positions and I wasn't sure what makes them different.

    This project serves three purposes:

  • Figure out which analyst path makes the most sense for someone with my background and interests.
  • Get hands-on practice with a complete analytics workflow - from extracting and cleaning data to analyzing and visualizing it.
  • Test out AI-assisted coding, even though I only know basic SQL and have zero Python experience.

  • 2. 🛠️ Data Source & Methods

    Data Source

  • Arbetsförmedlingen (Swedish Public Employment) API https://arbetsformedlingen.se/om-webbplatsen/apier-och-oppna-data. /em
  • Coverage: 2016-2024, ~15,000 analyst job postings, 6 analyst roles (Digital, Business, Data, BI, Operations, Commercial) chosen based on personal career interests.
  • Methods

  • AI-assisted coding: I used AI to help write the Python scripts, then adapted and validated the code myself. Since I'm still early in my learning journey, I focused on understanding the overall data pipeline rather than spending excessive time on syntax. This gave me a solid big-picture view of what data professionals actually do day-to-day.
  • Role classification: Job titles were mapped to analyst categories using a Python keyword mapping script (covering both Swedish and English).
  • Skill extraction: Python text-mining rules were applied, with synonym mapping (e.g., “SQL” vs. “Structured Query Language”).
  • Handling duplicates: For time-series analysis, I kept duplicate postings with unique IDs since they show real demand patterns; But for skill analysis, I consolidated duplicates by content to avoid skewing the results.
  • Data removal: Headhunter companies were removed from employer analysis since they don't represent the actual hiring companies.
  • Validation: I manually reviewed 100 job postings to check data accuracy.
  • Process

  • Data Extraction: Used AI-assisted Python to script job postings from the API.
  • Data Cleaning: SQL queries to pull relevant field and Python text mining to extract skills from job descriptions.
  • Validation A: Manual review of 100 postings.
  • Analysis: Excel.
  • Visualization: Tableau dashboards.
  • Validation B: Light comparison with industry reports.
  • Limitations

  • Single data source: Arbetsförmedlingen tends to over-represent large companies and public sector jobs, while under-representing startups and international companies.
  • Time lag: Data covers only 2016–2024, so may miss the most recent market changes.
  • Limited Validation.

  • 3. 🧐 Analysis: Job Market Reality Check

    Market Demand Hierarchy:

  • Digital Analyst: 38.5% (highest demand, driven by companies' focus on marketing analytics).
  • Business Analyst: 29.9% (second largest, but sensitive to economic conditions. When the economy slowed in 2023-2024, Business Analyst postings dropped significantly).
  • Data Analyst: 15.0% (smaller share, but growing fast at +27% CAGR).
  • BI Analyst: 11.4%.
  • Operations Analyst: 2.9%.
  • Commercial Analyst: 2.4%.
  • Economic Impact Patterns:

  • COVID Boom (2020-2021): Job postings doubled due to company expansions and digitalization.
  • Reality Check (2023-2024): Job market cooled off significantly as the market normalized and the economy weakened.
  • Seasonality: 40% fewer postings in July and August (Swedish summer effect).
  • Dashboard A - Market Overview


    4. 📍 Analysis: Geographic Intelligence

    Analyst opportunities are highly concentrated: nearly 90% of all postings are clustered in just three regions - Stockholm, Västra Götaland, and Skåne.

    Regional Specialization

    Stockholm (≈50% of all jobs):

  • Strong focus on Digital Analysts (+8% vs Västra Götaland)
  • Under-represented in Business Analysts (-11% vs Västra Götaland)
  • Reasons: tech and startup ecosystem
  • Västra Götaland (≈25%):

  • Leads in Business Analyst opportunities
  • Reasons: concentration of manufacturing and traditional industries
  • Skåne (≈15%):

  • BI Analyst specialization (+6% vs Stockholm and Västra Götaland)
  • A balanced and diversified analyst job market
  • Dashboard B - Regional Analysis


    5. 🏢 Analysis: Employer Landscape

    Top Hiring Companies (100+ postings)

    Traditional Enterprises: Swedbank, IKEA, Volvo, Electrolux, Scania Digital Agencies: Noor Digital, Leadstar Media, Mild Media Consulting Firms: Columbus, Sogeti, Sopra Steria

    Role-Specific Employer Patterns

    Demand varies significantly by employer type.

  • BI Analysts: Concentrated in consulting firms (Columbus, Sogeti), suggesting project based demand for specialized expertise.
  • Business Analysts: Dominated by large enterprises (Volvo, IKEA, SAAB), where process optimization, stakeholder alignment, and operational excellence are business priorities.
  • Digital Analysts: Split among specialized agencies, e-commerce companies and startups, where digital performance and customer acquisition are fundamental.
  • Data Analysts: Retail and automotive leaders (H&M, ICA, Volvo), where customer insights and operational data drive competitive advantage.
  • Dashboard C - Employer Landscape


    6. 🎓 Analysis: Skills & Requirement

    Skills Analysis

    Universal Requirements:

  • Languages: Swedish (39% of all postings), English (38%)
  • Soft Skills: Collaboration (27%), Communication (25%)
  • Tools: SQL (15%), Excel (15%)
  • Growth Skills (2022-2024 trend):

  • Python: +8% growth
  • Looker: +70% (challenging Tableau/PowerBI)
  • Role Differentiation

    Unsupported block type: column_list

    Dashboard D - Skill Matrix

    The size of each square represents the frequency of a skill within a role, the larger the square, the more frequently the skill is required.

    The color of each square shows how over- or underrepresented a skill is across roles, red indicates overrepresentation, blue indicates underrepresentation.

    Dashboard E - Key Skills Evolution


    7. 💬 Takeaways & Reflections

    What I discovered through this project:

  • Business Analyst roles appear highly communication-driven and more vulnerable to economic fluctuations. Given my non-Swedish background, these roles might be more challenging in the Swedish work environment.
  • Data Analyst roles are more technical and show stronger long-term growth. Given my background in business, I can differentiate myself in Data Analyst positions where technical and business skills intersect.
  • Although I had hesitations about relocating to Stockholm, the data clearly shows that it remains the central hub for analyst careers in Sweden. This insight is shaping how I think about location choices.
  • I discovered Looker for the first time - a tool I now want to explore.
  • Regardless of role, I need to strengthen SQL, Python, and soft skills.
  • This project confirmed that I enjoy the investigative aspect of data work.

  • 8. 📈 What Could Be Improved

    This project was designed as a first exploration, not a final product. Several areas could be improved to strengthen both reliability and depth:

    Data Quality and Validation

  • Expand validation beyond a 100-posting sample and cross check more industry reports.
  • Test the impact of duplicate handling.
  • Compare Arbetsförmedlingen postings with other sources (e.g. LinkedIn) to assess representativeness.
  • Analytical Depth

  • Break down results by industry to understand sector-specific demand patterns.
  • Analyze skill co-occurrence (e.g., SQL + Python vs. SQL + Excel) to identify common skill bundles.
  • Benchmark Sweden’s results against the EU and global market to find Swedish specific patterns.

  • 9. 🗂️ Documentations & Codes