Applied economics, from firms to labor markets
I study how organizations make workforce decisions — and how those decisions accumulate into labor-market outcomes. My work combines econometrics, people analytics, data engineering, and business intelligence.
From employment spells and hiring pipelines to wages, job flows, and business dynamics, I build measurement systems that make economic questions testable.
Keegan Johnson
Applied Economist | Workforce Analytics & Data Science
About
I am an applied analytics practitioner with an MS in Economics and a professional specialization in workforce data. My work connects microeconomic questions inside organizations — hiring, turnover, pay, and mobility — with the wider labor-market statistics used to understand those outcomes in aggregate.
I have worked across predictive HR analytics, employee surveys, enterprise reporting, and the data infrastructure needed to produce defensible measures. I work primarily with R, SQL, Power BI, Snowflake, and dbt, alongside Python and Tableau.
My current independent initiative is the Open Workforce Analytics Standard (OWAS). This website is my personal research and project portfolio; OWAS maintains its own standards and methodology mission.
I am drawn to questions about people that are easy to count but difficult to explain. Working life has consequences far beyond an organization’s HR dashboard, and many of the interesting effects only become apparent when we look across firms, households, and labor markets.
Outside analytics, I’m usually experimenting with food from different places, traveling, playing golf, or improving my Turkish.
- Labor economics and organizational decision-making
- Econometrics, forecasting, and statistical modeling
- Workforce measurement and data architecture
- International travel, cooking, golf, and Turkish language
-
MS in Economics
University of North Texas
One labor market, two scales of analysis
Inside the firm: hiring, separations, pay, job changes, employee tenure, and decisions made under organizational constraints.
Across the economy: employment, labor turnover, business formation, wage growth, and movement between employers.
These are connected subjects, but they are not interchangeable datasets. Aggregate trends do not automatically identify individual or firm behavior; enterprise data cannot by itself establish economy-wide mechanisms. Much of my research begins by asking where the two perspectives illuminate one another — and where they do not.
Selected projects
Open Workforce Analytics Standard (OWAS)
Measurement methodology · data architecture · reproducible analytics
An independent reference for workforce definitions and implementation. It separates externally sourced standards and published evidence from proposed OWAS conventions.
Workforce economics: a macro data model
BLS/Census · R · Parquet · Power BI
A public-data pipeline and semantic model for studying employment, compensation, worker flows, and business dynamics across sources with different populations and temporal structures.
Enterprise workforce laboratory
Synthetic data · employment events · survival analysis
A privacy-safe, in-development environment to examine firm-level employment events, denominators, and predictive questions.
Selected professional applications
Enterprise workforce data architecture. At Post Holdings, I worked on multi-business-unit workforce reporting and the pipelines connecting HR systems to Snowflake, dbt, and Tableau. My work included data quality checks, turnover and headcount logic, talent-acquisition integration, and predictive modeling development.
Statistical workforce analysis. At Nokia, I applied predictive analytics to workforce questions and helped develop and report employee pulse-survey research.
Analytics delivery and governance. Across Catapult Learning, Terminix, and other employers, I have developed HR reporting and Power BI solutions, including row-level security implementations.
This is a selective account of applications, not a complete employment chronology. Employer data and proprietary materials are not reproduced here. Read my full resume.
Methods & tools
| Analytical methods | Data systems | Communication & delivery |
|---|---|---|
| Econometrics and statistical inference | R, SQL, Python | Power BI, Tableau |
| Survival and attrition modeling | Snowflake, dbt | Semantic models and dashboards |
| Time series and workforce forecasting | ETL/ELT and API integration | Technical documentation |
| Survey measurement and evaluation | Data validation and dimensional modeling | Stakeholder-facing analysis |
The methodological priority is consistent across projects: define the population and event, respect the time grain, test the data, then interpret the result.
Beyond the datasets
I gravitate toward data about people: what they do, where they work, why they leave, and which constraints shape decisions that look voluntary from a distance. Labor is threaded through household choices, migration, education, health, and nearly every part of economic life. I’m interested in the questions hidden in routine administrative data — particularly when the firm-level and macroeconomic stories seem to disagree.
Outside work, I enjoy golf, travel, cooking experiments that cross culinary borders, and studying Turkish. Living and traveling internationally has made me more curious about which features of work are universal and which depend on institutions, culture, and incentives.
Books I return to: Mostly Harmless Econometrics, The Armchair Economist, and Weapons of Math Destruction. They reflect a preference for careful identification, strong intuition, and a healthy suspicion of tidy explanations.