Editorial Process

Last reviewed 20 July 2026 by Sukie Gao.

Infant sleep is a topic where bad information does real harm, and where almost nobody shows their work. This page shows mine — where the numbers come from, where AI is involved and where it is not, what I refuse to write about, and what happens when I get something wrong.

Start here: I am a parent, not a clinician

This is the load-bearing fact and it belongs above everything else on this page. I have no medical training. I am not a pediatrician, a nurse, a midwife, a lactation consultant, or a certified sleep consultant. No part of this site is reviewed by a clinician before publication, and I am not going to pretend otherwise by using words like “expert” or “medically reviewed” that I have not earned.

What I can do is read primary sources carefully, represent them faithfully, and be relentlessly clear about the boundary between “here is what the AAP published” and “here is what I think.” Everything below exists to hold that boundary.

Where a page comes from

Topics are chosen from real questions parents are searching for — not from a keyword tool alone, but from the shape of the question behind it. “4 month sleep regression” is not a request for a definition; it is a person at midnight wanting to know whether this ends. If I cannot work out what the person behind a query actually needs, I do not write the page.

From there the sequence is: pull the primary sources, extract only what they actually say, build the structure around answering the question rather than around hitting a word count, draft, then edit hard. The edit is where most of the value gets added and most of the words get cut.

Sourcing standards

Sources are ranked, and the rank determines what a source is allowed to do on the page.

Three rules govern all of it. Every statistic carries an inline link to the organisation that produced it — if there is no link, the number does not go in. I never invent a figure to sound authoritative, so you will not find a single “73% of babies” on this site. And where sources genuinely disagree, I say so rather than picking the tidiest one; published wake-window bands, for instance, vary meaningfully between reputable pediatric sources, and pretending they are settled would be a lie of convenience.

The anchor sources are the AASM pediatric sleep duration consensus for total sleep, and the AAP plus NICHD Safe to Sleep for anything touching the sleep environment.

Disclosure: how AI is used here

I use AI tools as a research assistant and a drafting aid. Being vague about that would be its own small dishonesty, so here is the actual division of labour.

AI helps with: gathering and summarising what published sources say so I can go read the originals; suggesting an outline; producing a rough first draft of explanatory sections; catching clumsy sentences and repetition; and generating alternate phrasings when I have written myself into a corner.

AI does not decide anything. No AI output is published unedited — not one paragraph. Every number in a draft gets checked against the primary source by me, by hand, because language models fabricate citations confidently and this is exactly the subject where that would be dangerous. Every link gets clicked before it ships. The judgement calls — what to include, what to cut, what to refuse to answer, how to phrase a caveat so a tired parent does not misread it — are mine. And nothing personal is ever AI-written: my own experience appears on this site only where it is genuinely mine.

The short version, which also appears in the footer of every page: articles on My Baby Sleep Schedule are researched with AI assistance and edited, fact-checked and published by Sukie Gao.

The first-person rule

Most parenting content is padded with invented intimacy — anecdotes that never happened, dated to make them feel real. I will not do that, so I hold myself to a hard limit: I only write in first person about things I actually did.

Concretely, that means I will tell you that I felt lost with a newborn and spent an enormous amount of time searching for answers, because that is true and it is why this site exists. I will tell you a sleep sack was the one piece of gear I genuinely used with my daughter, because it was. I will not manufacture a specific night, a specific week, a weight, or a number of minutes to make a point land harder. When I write about sound machines, blackout curtains, monitors or bassinets, the framing is always what guidance says and what parents report — never “I tested this”, because I did not.

What I will not write about

• Anything that functions as medical advice. No diagnosis, no dosing, no declaring a practice safe for your baby.
• SIDS risk reduction as primary content. It is the most consequential topic in infant sleep and I am not qualified to be the page you land on for it. Pages reference AAP and NICHD guidance and link straight out.
• Prescriptive sleep-training methods. The method question is personal, contested, and belongs with your pediatrician. This site does arithmetic on age bands; it does not sell a philosophy.
• Anything requiring a clinical judgement I cannot make — reflux, tongue tie, growth concerns, sleep-disordered breathing. Each of these gets a “call your pediatrician” and no more.

Review cadence and dates

Every page carries a real publication date and, where it has been revised, a real modified date. Those dates are not regenerated on every deploy to look fresh — a build does not change a date, only an actual edit does. If a page says it was updated in a given month, something in it genuinely changed that month.

Every page is reviewed at least annually. Reviews check that outbound links still resolve, that cited guidance has not been superseded, and that the page still answers the question people are actually asking. When a review results in substantive changes, that is noted visibly in the page body, not just in metadata.

Correction policy

I would much rather be corrected than quietly wrong, so corrections are the fastest-moving thing on this site.

• Factual errors — a wrong number, a misread source, a claim the citation does not support — are fixed as soon as I can verify them, usually within 48 hours, and the page carries a dated correction note. Silent edits to substantive claims are not acceptable.
• Superseded guidance — a source publishes an update — triggers a rewrite of the affected sections and a new modified date.
• Dead or moved links are repointed at the current canonical source, or removed with the claim if no equivalent exists.
• Typos and clarity fixes are made without a note.

To report anything: email sukielovesupport@gmail.com with the page and the sentence. You do not need credentials and you do not need to be diplomatic. Clinicians who spot a misrepresentation of guidance in their field are especially welcome, and will be credited on request.

Independence

This site currently carries no affiliate links, no sponsored posts, no paid link placements, and no guest content. Nobody pays to be mentioned here and nobody can. The intended revenue model is display advertising through Google AdSense, which is served by Google’s systems and has no influence on what any page says — I do not know which advertiser appears next to which article and could not change it if I wanted to.

If affiliate links are ever introduced, it will be disclosed on every page carrying them and in the privacy policy on the day it happens. The editorial standards on this page will not change to accommodate it.

Holding me to this

None of the above is worth anything if you cannot check it. So: click the citations. If a link does not support the sentence it is attached to, that is a bug and I want the email. The rest of the site is on the homepage, the story behind it is on the about page, and the terms you are reading it under are on the terms of use page.