South Africa publishes a great deal of data about itself, and most of it goes unread
Government departments, regulators, industry bodies and universities release thousands of figures about South Africa each month. MzansiData turns a selection of them into charts, explainers and tools for a general reader.
In a typical month Stats SA publishes inflation, retail sales, mining output and tourism arrivals. The Reserve Bank publishes interest rates and exchange rates. Eskom reports electricity demand by the hour, the Department of Water and Sanitation reports dam levels every Monday, naamsa reports new vehicle sales, and the IEC publishes election results down to each voting district.
Much of this is hard to use. Some of it appears only in long PDF reports, and some in spreadsheets whose column names make sense only to the people who compiled them. Most releases get a single news story on the day and are then overtaken by the next one, although many series go back decades and say more as a history than as a single release. This site is an attempt to make more of that material readable.
Mission and vision
We turn South Africa's public data into clear, non-partisan stories that people can use. Every chart is built from primary sources, explained in plain language, and open for anyone to check, share and learn from.
A South Africa where everyone can read the numbers that shape their lives, and where good data and data skills are a public good, not a privilege.
What we commit to
- Primary sources only, named on every chart.
- Analysis first: what the numbers mean, not just what they are.
- Non-partisan: no sides, only evidence.
- Open: method, sources and corrections in public.
- Free: no ads, no paywall.
What we have found so far
Before publishing anything, we compiled a list of what is released, by whom, how often and in what form. The first pass found
Source: MzansiData, source catalogue of sources with a regular release, grouped by topic (released 30 October 2026).
The quality varies. Some sources publish clean, documented files on a fixed schedule, and a few have an API. About a third publish only PDFs, some require a request form or a login, and several long-running series have been paused or renamed in the past two years. Much of the work behind each post is getting the data into a usable state.
What we will publish
Posts will appear regularly and will be of four kinds.
Regular updates
Short posts on figures that affect most households, such as prices, employment, fuel, the rand and interest rates. Each has one or two charts and an explanation of what changed since the previous release.
Longer analyses
Pieces that follow one question through several datasets, for example who wins government tenders, where people are moving within the country, how load-shedding changed electricity use, or how a ward has voted over thirty years. These take longer to produce and appear less often.
Tools and trackers
Interactive pages for looking up your own area or circumstances, such as municipal finances or a personal inflation rate. Every tool is listed on the tools page.
Notes on the data
Short notes on individual sources: what a dataset measures, where it falls short and how to obtain it. These are for readers who want to check our work or use the data themselves.
The topics will include the economy, politics and government, crime and justice, the environment and climate, markets and money, media, sport and culture. Anything South Africa measures regularly and publishes is a candidate.
How we work
Each chart has a source line giving the publisher, the dataset and the date of the release it uses. Where we have calculated a figure rather than copied it, the post says how.
All the data comes from the body that publishes it, mostly official statistics agencies and established research bodies. When a series has a known weakness, such as a change in method, a gap or a revision, we say so where it affects the reading.
We describe what the data shows and how reliable it is. Conclusions about policy are left to readers.
The code that fetches and processes the data is kept with the site, so any chart can be rebuilt from its original source.
MzansiData is made by people working with AI. AI tools help find and process the data and draft each post. A person checks the figures and approves every post before it is published.
Contact, press and privacy
We welcome suggestions for datasets or questions worth looking into, and corrections. Corrections are noted in the post they apply to.
- Contact: suggest a dataset or a question, send a correction, or ask about using our charts.
- Press: the credit line, links to our charts and tools, our sources and their terms, and the logo.
- Privacy note: what we collect when you read the site, subscribe or write to us, and how to have it deleted.
- Topics and all stories.