Editorial policy

Why this page exists

Google's guidance on assessing content quality asks three questions about anything published: who made it, how it was made, and why. Most agency blogs answer none of them, which is part of why most agency blogs are worth so little. This page answers all three for everything we publish.

Who writes this

Articles are published under the SBPO Consulting name and the organisation is accountable for them. Where an article is attributed to a named individual, that person genuinely wrote or reviewed it and has agreed to be named.

We do not invent author personas. A great deal of content marketing is published under fabricated bylines with stock-photo headshots and invented credentials; we would rather attribute honestly to the organisation than manufacture a person. As named contributors are added, they will appear with real, checkable credentials.

How claims are sourced

• Primary sources, deep-linked. Where we state how a platform behaves, we cite that platform's own documentation and link to the specific page, not a domain root.
• Every source on the page. Articles and service pages carry a visible references list. If a claim is not sourced there, treat it as our opinion or our practice rather than an external fact — and we try to make which is which obvious in the prose.
• No statistic we have not opened. We do not repeat figures circulating in other blog posts. If we cannot reach the original study or documentation, the number does not go in. This is why our articles quote fewer statistics than most.
• No invented case studies. We do not publish client outcomes, project metrics or named clients without written permission, and we do not illustrate a point with a composite client who does not exist.

How AI is used

We use AI tools in drafting and research, in the same way we use a search engine and a spell-checker. We think stating that plainly is more useful than the alternative, which is that almost everyone does it and almost nobody says so.

What that means concretely:

• A human decides what an article is for and what it will argue before drafting begins.
• Every factual claim is verified against a primary source by a person. An AI tool's assertion is treated as a lead to check, never as a citation.
• Every article is edited by a human for accuracy, argument and voice before publication.
• A human is accountable for what is published, regardless of how the first draft was made.
• We do not mass-generate pages to occupy keywords. Google's spam policies treat scaled content abuse as a violation whether it is produced by a person or a machine, and we agree with the reasoning: the problem is publishing content that exists to rank rather than to help.

Dates and updates

Every article shows its publication date and, where it has been revised, its update date. We change the update date only when the substance changes — not on a rebuild, and not to make old content look fresh. A date that moves without the content moving is a lie told to a crawler.

Technical content ages quickly. Where an article describes a platform behaviour that has since changed, we would rather revise it than delete it, so that anyone who relied on the old version can see what changed.

Corrections

When we get something factually wrong, we correct the article and add a dated note saying what was changed. We do not silently edit a claim out of existence.

To report an error, use the contact page. We will respond either way, including when we think the original was right — in which case we will explain why.

Independence

• No paid placements. We do not accept payment, affiliate commission or free product in exchange for coverage or recommendation. Nothing on this site is sponsored.
• No guest posts. We do not accept guest articles or link insertions, and unsolicited offers of either are deleted unread.
• No undisclosed relationships. If we ever have a commercial relationship with a platform we recommend, we will disclose it in the article. We currently have none.
• Recommendations are not sales. Where the honest recommendation is a tool or approach that means less work for us, we publish that.

What we will not publish

• Content whose only purpose is to rank for a term we have nothing useful to say about.
• Statistics we cannot trace to a primary source.
• Client work, results or names without written permission.
• Comparisons written to make a competitor look bad rather than to help a reader decide.
• Anything about other companies with names similar to ours. Their business is not our subject.

Feedback

If an article is unclear, incomplete or wrong, telling us is genuinely useful — it is how the good ones get better. See our about page for who we are, and our approach for how the same standards apply to client work.