Editorial standards

Most writing about education technology is either marketing in disguise or a chart with no source under it. We hold ourselves to a different standard, and this page is where we write it down so you can hold us to it too.

The short version: we publish under the Learner First masthead and source every load-bearing claim to a primary reference, we review products against a rubric you can read, we keep our own products out of those reviews, and when a check hasn’t happened yet we say so rather than imply it has.

Who’s accountable for what we publish

Today we publish under one masthead — Learner First — not individual named bylines yet. That’s a deliberate, honest choice while the bench is small: rather than dress the work up as a newsroom it isn’t, we put the publication’s name on it and hold that name to the standards below.

Named individual authorship, and a separate credentialed reviewer for each piece, are what we’re building toward as we grow. We’ll show them on the page the day they’re real — not before.

How we review products

When we review a tool or a service, three rules hold every time.

• We review against a published rubric. The tests a product is scored on are written down and the same for everyone — not adjusted to fit the conclusion we wanted.
• We exclude our own products. Learner First is published by a company that also builds education software. Where one of the publisher’s products could appear in a review or a directory, we disclose it and leave it out of the scoring. We never rate our owner’s work favourably and call it independent.
• We’re plain about money. Where a link earns us a commission, we say so on the page, and the commission never changes the score. We don’t run affiliate links on our own products.

The label under every piece

Every piece carries one honest label under that masthead, so you never have to guess how far it has been checked.

1. Claims sourced to primary references

   The default. Every load-bearing claim traces to a primary source, linked at the foot of the piece. It is descriptive, not a badge — it says where the facts come from, it does not claim an independent audit we haven’t run.

2. Opinion

   An argument rather than a sourced report. We label it so a position is never read as if it were a finding.

3. Reviewed by [name]

   Once a credentialed subject expert has checked a piece, their name goes here. Until it does, we don’t imply it has happened.

A credentialed reviewer is the strongest trust signal a publication can carry, and almost no one in this field uses it — so we’re building that bench rather than claiming it early. If a reviewer’s name is there, that review happened. If it isn’t, it hasn’t yet.

How we judge our own writing

Before a piece publishes, it’s scored against five tests. A piece that doesn’t clear the bar gets rewritten or dropped.

1. Provenance

   Every claim is backed by a primary source, real data, or first-hand use. This carries the most weight, because we write about decisions families and institutions make with money and time.

2. Specificity

   Concrete numbers, names, and dates. “RM 99 a month” beats “affordable.” “Penang” beats “Southeast Asia.”

3. Structure

   Answer-first, scannable, and built on the right template for the piece — a review, a how-to, and an analysis are not the same shape.

4. Voice

   Plain, specific language. No buzzwords, no “world-class,” no urgency theatre. If a sentence could appear on any site, it doesn’t belong on this one.

5. Utility

   The reader leaves able to decide or do something they couldn’t before. A piece that informs but doesn’t change what you can do hasn’t earned its place.

How a piece gets made

Every piece runs the same path, start to finish.

1. A brief, before anything is written: who it’s for, what they’re trying to decide, the claim we’re making, and the evidence it will take to stand it up. No brief, no piece.
2. Research in parallel: the question behind the search, the primary sources and real figures, and how an answer engine would read the result.
3. An angle gate: a piece needs original data, operator experience, recency, specificity, a contrarian read, or genuine depth — one of those, or we don’t write it. A human signs off the outline before drafting starts.
4. Three drafting passes: get the argument down answer-first, strip the filler, then set the voice.
5. Visuals that carry their weight: every chart is also written out in words beneath it, so the point survives whether or not the image loads.
6. A scored quality gate: the piece is graded against the five tests above and must clear the bar; flagship pieces also take an adversarial read designed to break them.
7. Live checks at desktop and mobile, sign-off, publish — then we watch how it actually performs and revisit it.

Corrections

We get things wrong sometimes. When we do, we fix the piece and note what changed and when, rather than editing it quietly. If you’ve found an error, please tell us — we’d rather hear it from you than leave it standing.

A note on AI

We use AI to help research and draft, the way a newsroom uses any tool. It doesn’t change the standard on this page: a named person still owns the argument, the claims are still checked against primary sources, and nothing publishes that a person hasn’t read and stood behind. The judgement is ours.