Editorial standards

We do not publish a statistic we cannot source. Resume advice circulates a great deal of confident, unattributed numbers — that recruiters spend exactly six seconds on a resume, that applicant tracking systems automatically discard three quarters of applications — which are repeated so widely that they read as established fact while tracing back to nothing verifiable.

Our guides do not use them. Where such a claim is common enough that leaving it unaddressed would itself mislead, we name it and explain what is actually known instead. Percentages that appear in our articles are illustrations of how to phrase a resume bullet, and are written so that their status as examples is unmistakable.

Claims about our own product are held to the same rule. We describe templates as designed for ATS readability, never as guaranteed to pass, because we have no access to any commercial ATS and cannot substantiate the stronger claim.

Product guidance — what a score means, why an issue was raised, what to change — is derived directly from the rules the engines actually run. When the code changes, the explanation changes with it, because the explanation is generated from the same findings that produced the score rather than written separately alongside it.

Longer-form guides are written by us and edited before publication. Where AI assists in drafting, a person is responsible for every claim that survives into the published text; we do not publish unreviewed model output, and no article is bylined to a person who did not write it. You will notice we do not attach author names or credentials to our guides at all — that is deliberate, since an invented byline or an inflated credential would be the easiest fiction on this whole site.

Advice is aimed at helping you present work you actually did. We will not suggest padding a resume, and the AI in this product is constrained so that it cannot do so silently — see how Truth Guard works.

These are the rules, and they are enforced in the code:

• Only genuine submissions are ever displayed. Placeholder records used during development are filtered out before the page renders, so a seeded example cannot reach a visitor even by mistake.
• When there are no real reviews, the section does not appear. There is no filler, and no rating is shown. As of now, that is the state you will find the homepage in.
• No review is labelled verified, because we do not operate any mechanism that ties a review to a confirmed purchase. Applying the badge anyway would be claiming a check we do not perform.
• We do not publish an aggregate star rating in our search-result markup. Sites that show rating stars in Google results on the strength of invented reviews are a well-known abuse; we emit no rating data at all.
• We do not pay for reviews, offer anything in exchange for one, or edit a review’s substance. Reviews are not filtered by how favourable they are.

Stated plainly, so you can hold us to it and so it is unambiguous internally:

• Invented users, customers, or usage numbers.
• Company logos implying a customer or partner relationship we do not have.
• Fabricated reviews, ratings, or success stories.
• Endorsements attributed to recruiters, hiring managers, or employers who never gave one.
• Made-up credentials, awards, certifications, or press coverage.
• Guarantees about hiring outcomes or about passing any specific applicant tracking system.

We would rather show nothing than show something invented — the About page explains the reasoning.

Guide and blog pages carry advertising. Advertisers have no input into what those pages say, no advance sight of them, and no influence over our recommendations. We do not run sponsored posts presented as editorial, and we do not rank or recommend anything on the basis of a commission. The builder itself, and the documents you produce with it, carry no ads.

We get things wrong. When we do, we would rather fix it visibly than quietly.

• A factual error in a guide is corrected as soon as we can confirm it.
• A change to how a score is calculated that would move your number is recorded on the changelog, not applied silently.
• Where a correction changes the substance of published advice, we say what was wrong rather than editing the sentence and moving on.

To report something inaccurate, email support@mbresumebuilder.com. Pointing at a specific sentence and telling us what is wrong with it is the most useful form this can take.