We turn PDFs, price lists and paper records into data you can use.

Mhlopheni Group is a South African data company. We collect and structure hard-to-reach information, analyse it, and build the datasets, dashboards and tools that put it to work.

Based in Ekurhuleni, Gauteng. Working with clients across South Africa.

Weekly specials!!

While stocks last, Mon to Sun

  • Maize meal 10kg ...... R89,99
  • White sugar 2.5kg only R52.99!
  • Sunflower oil 2L R64,99 (was R72,49)
  • Long grain rice 2kg ... R38.49
  • Brown bread 700g R17,99
Item Size Price (R) Was (R)
Maize meal 10 kg 89.99
White sugar 2.5 kg 52.99
Sunflower oil 2 L 64.99 72.49
Long grain rice 2 kg 38.49
Brown bread 700 g 17.99

Illustrative sample, not real store data.

Collection comes first

Most data problems we see aren't modelling problems. The information exists, in scanned reports, weekly flyers and spreadsheets that don't agree with each other. It just can't be queried yet. We start where the data actually is.

  1. Collect

    Pull from PDFs, scans, websites, spreadsheets or people in the field, with a record of where every value came from.

  2. Structure

    Clean it and fit it to one schema, so units, dates, names and prices are consistent.

  3. Analyse

    Find the patterns that matter to your decision, with the method written down so it can be checked.

  4. Deliver

    Hand it over as a dataset, an API, a dashboard or an app: whatever your team will actually use.

What we do

Three kinds of work, often combined in one project. Every engagement has a fixed scope and a fixed price.

Data collection and extraction

Getting information out of wherever it's stuck.

  • Turn PDFs, scans and legacy archives into clean JSON, CSV or a database
  • Build field-collection apps for teams gathering data on the ground
  • Build APIs over your data so your other systems can use it
  • License datasets we've already built

Good fit for

Publishers, edtech companies, research units, NGOs and app developers.

Analysis and dashboards

Making collected data answer a question.

  • Price tracking and basket comparisons across retailers and regions
  • Household affordability and cost-of-living analysis
  • Dashboards that open in a browser with nothing to install
  • Commissioned reports with a documented method

Good fit for

Retailers, FMCG brands, unions, NGOs, research institutes and newsrooms.

Document automation

AI applied to documents and structured data.

  • Extract and classify large document collections
  • Generate structured content to a fixed format: question banks, catalogues, reports
  • Search and ask questions across your own documents
  • Automated validation, deduplication and reconciliation between sources

Good fit for

Firms with document backlogs, publishers, edtech and financial services.

We also build the software around the data: Flutter mobile apps, web dashboards, admin portals and APIs running on Cloudflare’s edge network. And we run practical training for teams who want to do more of this themselves.

What we’ve built

Our own products and datasets. They’re where we test ideas before offering them to clients, and they show the kind of work we do.

In development

Shop Savers

Grocery price comparison for South African households. It finds the cheapest way to buy a whole basket across nearby stores, alerts shoppers when prices drop, and lets stokvels place group orders.

The basket is grounded in published household affordability research, so it reflects what families actually buy.

Flutter, Firebase, basket price optimisation

Available to license

K53 road sign dataset

65 South African road signs with descriptions and high-resolution images, sorted into regulatory, warning, information, guidance and road-marking categories.

JSON, CSV and a browsable viewer. Ask about licensing at licensing@mhlophenigroup.co.za

Built

ExamZA

TVET exam papers and memorandums side by side, so students can work through a paper and check answers without switching files.

Flutter, Cloudflare R2, D1 and Workers, React admin

Built

ClearRoad

A learner’s licence study app with 52 lessons, saved progress, and a practice test that unlocks once the material is covered.

Flutter, Firebase

Internal tool

Currency analytics dashboard

Volatility, return distributions and risk-return profiles for five currency pairs, fed by our own API and refreshed automatically.

Cloudflare Workers API, JavaScript, Chart.js

How a project runs

  1. Show us the data

    Send a sample: a few PDFs, a spreadsheet export, or a description of what you collect and what you need from it.

  2. Get a fixed proposal

    Within five business days you receive a written scope, price and milestone dates.

  3. See it working at each milestone

    You review real output as it’s built, not only at the end.

  4. Own what we hand over

    The data, the code, and a record of every source and transformation are yours.

We reply to every enquiry within two business days.

Built for POPIA from the start

Personal information is part of most data work. We design for the Protection of Personal Information Act from the first line of code rather than adding it later.

  • We collect only what a stated purpose needs.
  • We ask for consent in plain language, and record it.
  • Each client’s data stays separate. Competing clients never see each other’s information.
  • Every dataset comes with a record of its sources and what was done to it.

Questions about how we handle information go to our Information Officer at privacy@mhlophenigroup.co.za.

Research lab

Alongside client work we run an internal research programme on market data, quantitative modelling, payment systems and digital money.

It keeps our simulation, risk-modelling and data-engineering skills sharp, and it’s where product ideas are tested long before they reach a customer.

Our research is internal. Mhlopheni Group is not an authorised financial services provider and does not offer financial advice, investment or portfolio management, trading signals, trading software, or crypto asset services.

Work with us

We’re a small team that works with skilled contractors, project by project. If you’re good at one of these, we’d like to hear from you.

See roles and apply
  • Contract Flutter developerRemote
  • Contract data engineer (Python)Remote
  • Price collector, part-timeGauteng

Tell us what data you’re working with.

Send a sample or a short description. You’ll hear back within two business days.