I turn messy data into decisions people act on.

Naledi Dube — data analyst in Johannesburg. Six years in retail and public health. Open to new roles from January 2027.

About

Numbers, then people

I am a data analyst who would rather sit with the people who use a report than polish it alone. The best dashboard is the one somebody opens on a Monday morning.

What I do

I clean and join messy data, build models that answer one question well, and turn the result into a dashboard or a one-page brief that a manager can act on.

Where

Four years at a national grocery chain, forecasting stock for 120 stores, and two years with a public-health NGO measuring clinic waiting times in Gauteng.

A team that cares about getting decisions right, not just reports out. Hybrid in Johannesburg, or remote.

Experience

Six years, two sectors

2019

BSc Statistics

Graduated with distinction; thesis on forecasting with sparse data.

2019

Junior analyst, national grocery chain

Weekly sales reporting for 120 stores; automated a 2-day manual report into 20 minutes.

2021

Demand analyst, same chain

Built the stock forecast for fresh produce; waste down 14% in the pilot region.

2023

Data lead, public-health NGO

Measured waiting times at 30 clinics; the redesign cut the median wait by 38 minutes.

Now

Open to new roles

Senior analyst or analytics lead, from January 2027.

The tools I use every week, and what I use them for.

Skills

SQL & Python

Cleaning, joining and modelling data — from a spreadsheet to a warehouse.

Dashboards

Power BI and Looker Studio reports that people open without being reminded.

Forecasting

Demand and capacity forecasts, honest about their uncertainty.

Fieldwork

Clinic visits, store walks and interviews — the data behind the data.

Writing

One-page briefs a busy manager can read on a phone.

Languages

English, isiZulu and Sesotho.

Projects

Two pieces of work I am proud of

Less fresh produce thrown away

A store-level forecast for fruit and vegetables at a national grocery chain. Buyers got a daily order suggestion instead of a gut feel; waste fell 14% in the pilot region and the model went national the next year.

Shorter waits at 30 clinics

Two months of arrival times, collected on paper with clinic staff, showed that most of the wait happened before 08:00. Staggered appointments cut the median wait by 38 minutes.

References

Full contact details on request.

Naledi made the forecast trusted, not just accurate. She spent a week in the stores before she wrote a line of code.

Head of planning National grocery chain

She turned a pile of paper registers into the clearest report our board has ever seen — and then helped the clinics act on it.

Programme director Public-health NGO

Contact

Get in touch

A role, a project or a question — I reply within two working days.

Message sent

Thank you for reaching out. We’ll be in touch shortly.