Our work

Numbers first. Then the story.

What we have built, what it does every working day, and what it took. Every engagement started with the audit.

Case study 01

Client

Business Bagel

South African business news publisher. Newsletter, web, social, podcast and video.

The problem

Output capped by headcount

Three people, and every new format meant another salary or got dropped.

What Veld built

The production engine

Turns one editorial decision a day into a full day’s publishing across seven platforms.

Status

In production

Five runs a week since June 2026. Built, run and monitored by Veld.

One decision a day.

Every publishing day Business Bagel puts out a newsletter, three social posts, six or more web articles, a podcast read and up to two video scripts. A person makes one call in that cycle, which stories run. The engine does everything after it, and publishes the result across seven platforms.

5

Publishing runs a week, Sunday through Thursday

~15

Finished, ready-to-ship deliverables per run, across seven platforms

1

Point in the run where a human decides anything

06:00

Newsletter in subscriber inboxes, every run

A run, end to end · South African time

How the day actually runs.

Capture and selection are kept apart on purpose. The one human step is set in the time it takes to drink a coffee.

07:35

Radar

Sweeps the business desks for what broke overnight and ranks it by how little of the market has covered it yet. The brief is in the editor’s inbox by eight.

10:00

Discovery

Captures the full candidate list from a fixed set of feeds. Deliberately neutral: no weighting, no ranking, no framing.

+ 2–6 min

Picks Human

The editor marks up the shortlist: the roundup stories, the two main posts, the backups, the standalone articles. That is the whole input.

Unattended

Build

Sourcing, drafting, source-tagging, fact-checking, layout-ready finals, long-form articles, the newsletter, the podcast read, the video scripts and the platform descriptions, through to scheduled and ready to ship.

05:45

Articles go live

Each picked story’s web article publishes overnight, stepped fifteen minutes apart.

06:00

The newsletter sends

Fifteen minutes after the last article, so every “read the full story” link already works when it lands.

10:00 →

The day fills in

Standalone articles release every two hours. Thursday’s run schedules across the weekend, so the site keeps publishing on days nobody is working.

One run

What comes out the other side.

All of it comes off one sourcing pass and goes out across seven platforms. A fact is verified once, then travels to every platform that needs it.

1

Newsletter, emailed to the full subscriber list

1

Roundup carousel carrying the day’s four stories

2

Main post carousels, five slides and a description each

6

Web articles, one per picked story, 350–500 words

1

Podcast script, a spoken read of the edition

0–2

Video scripts, each with three hook options

1

Description set for YouTube and Spotify, plus five hooks

+

Standalone articles scheduled out across the coming days

Consistency

The value isn’t in what the system can write. It’s in what it isn’t allowed to.

Rule · Sources

Nothing written from memory

Every figure, date, name and quotation exists first as a verbatim quote captured from the source. A quote with no exact match on file stops the run.

Rule · Corroboration

One outlet isn’t a story

A story resting on a single uncorroborated report is removed from the run and a backup takes its place. The system would rather publish less.

Rule · Checks

No model marking its own work

No AI auditors, no self-grading. The checks are countable: quote coverage, word counts, option counts, link integrity.

Rule · Recency

Freshness is on the news, not the fact

A story qualifies only if something actually happened in the window: a filing, a result, a deal, a statement.

Beyond the run

All of it runs inside a team of three.

3

People, total

5

Advertising clients, each with a monthly client feature the engine produces

5/mo

Post-campaign analytics reports, one per client, generated automatically

2

Human touchpoints in the video pipeline: present, finalise the edit

“Veld has been the reason we’re able to perform like a 20-person media business with a headcount of only 3.”

Christian Maidman · Founder, Business Bagel

More work

The same approach, in very different businesses.