ABOUT ME
Lena Novak
Data analyst turning messy data into decisions people can
actually trust.
Open to freelance & full-time
Chicago, IL
Remote-friendly
7
Years in analytics
120+
Dashboards shipped
92%
Still used after 90 days
−35%
Ad-hoc requests since self-
serve
CONTACT
lena@novakanalytics.com
github.com/lenanovak
linkedin.com/in/lenanovak
Chicago, IL · CST
TOOLKIT
LANGUAGES
SQL
Python
R
BI & VISUALIZATION
Tableau
Looker
Power BI
Excel
PLATFORMS & PIPELINES
BigQuery
Snowflake
dbt
Airflow
Fivetran
Git
CERTIFICATIONS
✓
Google Data
Analytics
Professional
Certificate
✓
dbt Fundamentals
· dbt Labs
✓
Tableau Desktop
Specialist
·
Tableau
CURRENTLY
READING
The Data Detective — Tim Harford
I'm Lena Novak — a data analyst who thinks a dashboard nobody opens is
worse than no dashboard at all. For the past seven years I've worked
across healthcare, e-commerce, and logistics, turning messy production
databases and conflicting stakeholder priorities into numbers people
actually trust.
The career started sideways. I studied economics and fell into analytics by
building a pricing model for a friend's secondhand bookstore. The
spreadsheet became a Python script, the script became a side project, and
somewhere in between I realized that the most interesting questions in any
company are answered by the same thing: good data, asked honestly.
Today I work on the analytics team at Northgate Health, where I build
reporting pipelines, investigate strange metric movements, and spend a
suspicious amount of time in the data dictionary. I care deeply about
documentation, reproducible analysis, and explaining variance to people
who would rather I just say "it's fine."
HOW I APPROACH DATA
01
Start with the decision
Every analysis starts with one question: "What are you going to do
differently with this?" If the answer is nothing, the analysis can wait
until it isn't. It keeps me from building beautiful reports nobody acts on.
02
Make numbers auditable
Clean naming, documented transformations, visible assumptions. If a
stakeholder can't trace a number back to its source query, the number
doesn't mean anything. My rule: the data dictionary is a deliverable,
not an afterthought.
1 / 3
