Baby Milestones Month by Month, and When Screening Actually Happens

Baby milestones month by month, built around the two things that actually matter: how wide the normal ranges are, and when the AAP schedules formal screening. Both taken from primary sources, because the version of this article we published before got the screening ages wrong.

Month-by-month milestone charts are useful for knowing roughly what is coming. They are bad at the thing parents actually use them for, which is deciding whether to worry — because they present single ages for things that have windows several months wide.

So this article gives the ranges rather than the ages, and puts the screening schedule at the top rather than the bottom.

The schedule, which is the part that matters

Three distinct things happen, and they are routinely confused with each other. All three quotes below are from the AAP's developmental surveillance and screening page.

Developmental surveillance — every visit. The AAP states: "AAP recommends conducting developmental surveillance at every health supervision visit, with special attention at the 4- to 5-year visit as the child prepares to enter elementary school." This is the clinician watching, asking and noting, and it is continuous.

Standardised developmental screening — 9, 18 and 30 months. "General developmental screening tests should be conducted at the 9-, 18-, and 30-month health supervision visit." This is a formal tool, not a conversation.

Autism-specific screening — 18 and 24 months. "Autism spectrum disorder screening tests should be conducted at the 18- and 24- month health supervision visits."

Note that the 24-month visit carries autism screening but not general developmental screening, and the 30-month visit carries general screening but not autism screening. That is why they are listed separately rather than merged.

The ranges, from the WHO

The WHO Multicentre Growth Reference Study publishes windows of achievement for six gross motor milestones. The windows themselves rest on 816 children in five sites: Ghana, India, Norway, Oman and the USA. Brazil was the sixth country in the wider growth study, but the motor component was added after fieldwork there was already under way, so Brazil never collected it.

Two qualifications the WHO attaches, which matter more than any single number in the table. The children were selected as healthy ones "living under conditions that are highly unlikely to constrain growth", whose mothers breastfed and did not smoke. And the study was "designed to produce a standard rather than a reference" — a picture of how children can develop when very little is holding them back, rather than a survey of how babies in general do. So read what follows as the spread among healthy, well-supported children, which is a narrower claim than it first looks. WHO's own use for the windows is narrow too: they are "recommended for descriptive comparisons among populations, to signal the need for appropriate screening when individual children appear to be late in achieving the milestones, and to raise awareness about the importance of overall development in child health".

One thing to get straight before reading the table. Each window runs from the 1st to the 99th percentile of the ages at which those children reached the milestone — the left number is the age by which the earliest one per cent had managed it, the right the age by which all but the last one per cent had. Neither end is a typical age, and neither is the same kind of number as the CDC's 75% threshold quoted further down.

Milestone	Window (months)
Sitting without support	3.8 – 9.2
Standing with assistance	4.8 – 11.4
Hands-and-knees crawling	5.2 – 13.5
Walking with assistance	6.0 – 13.7
Standing alone	6.9 – 16.9
Walking alone	8.2 – 17.6

One disagreement between WHO's own documents, worth naming because this article links all three. For walking with assistance the windows table gives a left bound of 6.0 months, while the percentile table for the same study and the published paper both give 5.9 (181 days). The figures above are the windows table's. The other eleven bounds are identical across all three documents, and nothing in this article turns on the difference.

Read the overlaps rather than the milestones. A baby can be sitting unsupported at four months or at nine. Crawling can start at five months or not until thirteen. Standing alone opens at 6.9 months and its window does not close until 16.9.

Two babies of the same age can be five months apart on the same milestone and both be entirely unremarkable. That is why "is my baby behind" is usually the wrong question and "what did the last check-up say" is the right one.

Roughly what happens when

Ranges rather than ages, and a chart of this kind is a rough guide. Where the WHO publishes a window we use it. Where we give an age for a milestone outside the WHO's six, it comes from a CDC or AAP page and is linked to it; the rest of each band is general description, not a dated milestone.

Months 1–3. Head control builds. Focus improves. Smiling back at you is on the CDC's list of what most babies do by 2 months — the checklist item is "Smiles when you talk to or smile at her" — and the CDC's own definition on that page is worth carrying with you through the CDC and AAP ages below: developmental milestones are "things most children (75% or more) can do by a certain age", not things all of them do. Sitting without support is already possible from 3.8 months at the very early end of the WHO window.

Months 4–6. Rolling, reaching, mouthing everything: "Rolls from tummy to back", "Reaches to grab a toy she wants" and "Puts things in her mouth to explore them" are all on the CDC's list of what most babies do by 6 months. Standing with assistance opens from 4.8 months. Unsupported sitting does not belong in this band, whatever a monthly chart says: 3.8 months is the early edge of the WHO window, not a typical age, and no CDC or AAP page places sitting here.

Months 7–9. Crawling for some — its window opened at 5.2 months and stays open until 13.5. Standing alone becomes possible from 6.9 months. This is where unsupported sitting actually falls: the AAP says that by six to eight months, if you position a baby upright, they will stay sitting without leaning forward on their arms, and the CDC lists sitting without support among what most babies do by 9 months. The AAP attaches the line worth acting on: "If she is not sitting by herself by nine months, let your pediatrician know." The 9-month visit carries the first standardised developmental screening.

Months 10–12. Cruising, pulling up, possibly first steps — walking alone opens at 8.2 months, so some babies are already walking well before the first birthday and many will not for months yet. On the WHO's percentile table for the same study, the median for walking alone is 365 days, which the table converts to 12.0 months. The first birthday is roughly the halfway mark, not a deadline.

Months 13–18. Most of the WHO's children were walking before this stretch was over: on that same percentile table the 75th percentile is 13.1 months and the 90th is 14.4. The window nonetheless does not close until 17.6 months, so a child still not walking at 15 or 16 months is inside it rather than outside it. The 18-month visit carries both a general developmental screening and an autism screening.

Months 19–30. The 24-month visit carries autism screening; the 30-month visit carries the third general developmental screening.

How to use a milestone chart without it making things worse

Use it to know what is coming, not to grade what has happened. Knowing that cruising precedes walking is useful. Knowing the average age of cruising is not, given the spread.

Track your own baby against themselves. Direction over time is more informative than position against a chart, and it is also what surveillance at each visit is doing.

Take specifics to the check-up rather than a general worry. "She isn't bearing weight on her legs at all" is actionable. "She seems behind" is hard to work with.

Do not skip the 9-, 18- and 30-month visits. Those are the three carrying standardised screening, and they are the mechanism this whole article defers to.

Ignore comparisons. Given windows five months wide, another baby of the same age is a sample size of one drawn from an enormous range.

Where a book fits, and where it does not

We make illustrated hardcover books from families' own memories, so read this as interested rather than neutral.

If you are here because you are worried, we are not the answer and no purchase is. The schedule at the top of this page is the answer, and after that your child's doctor. That is the entire honest response to the reason most people arrive at an article like this.

The narrow case where a book fits is the opposite mood: looking back at a first year rather than forward with anxiety. The milestones keepsake page covers that version and the first steps page covers the single moment. The walking article goes deeper on that one milestone.

The case against, plainly: a keepsake made to a chart is a keepsake that ages badly, because the chart was never the point and the ranges make it a poor record anyway. Make it about what happened, not about when it happened relative to average. If that is the version you want, the first birthday book is the same idea built around one day instead of a timeline.

Frequently asked questions

When does the AAP recommend developmental screening?

Standardised general developmental screening at the 9-, 18- and 30-month health supervision visits, and autism-specific screening at the 18- and 24-month visits. Separately from both, the AAP recommends developmental surveillance at every health supervision visit.

What is the difference between developmental surveillance and screening?

Surveillance is the ongoing process at every visit — watching, asking, noting. Screening is a standardised test at three specific ages. The AAP describes screening as supplementing and strengthening ongoing surveillance rather than replacing it. Confusing the two is what produced the wrong ages in the earlier version of this article.

How wide are the normal ranges for baby milestones?

Very. On the WHO's windows: sitting without support 3.8–9.2 months, hands-and-knees crawling 5.2–13.5, standing alone 6.9–16.9, walking alone 8.2–17.6. Two babies the same age can be five months apart on the same milestone and both be unremarkable.

Should you worry if a baby misses a milestone in a monthly chart?

A monthly chart gives single ages for things that have windows months wide, so "missing" one often means very little. Individual concerns belong at a check-up, which is exactly what the surveillance-at-every-visit recommendation exists for. Take a specific observation rather than a general worry.

Does a baby have to crawl before walking?

No. The WHO lists hands-and-knees crawling as one of six gross motor milestones with its own window, not as a prerequisite, and in that study 4.3% of the children did not exhibit hands-and-knees crawling at all. The AAP puts the alternatives plainly: "A few children never crawl. Instead, they use alternative methods, such as scooting on their bottoms or slithering on their stomachs."

What is the most useful thing to know about milestones?

That the windows overlap so heavily that comparing your baby to another of the same age carries almost no information, and that the three screening visits at 9, 18 and 30 months are the actual mechanism for catching problems.

How this guide was made

The AAP screening ages were taken from the AAP's own developmental surveillance and screening page and are quoted rather than paraphrased. The WHO windows were extracted from the WHO's published table with pdftotext, not from a page describing it. The same was done for the percentile table and for the WHO's own paper on the motor study, and the CDC and AAP pages behind the non-motor milestones were opened and read on 28 August 2026.

The screening ages and the WHO walking figure were both wrong in the version of this article we published before. The statistics audit of our own blog found it claimed the AAP recommends formal developmental surveillance at 2, 4, 6, 9 and 12 months — which conflates surveillance with screening and gets the screening ages wrong — and that it cited the WHO's walking figure as 9 to 18 months. It also asserted that the 9-month visit is used to screen for autism, which is not what the AAP says.

A second check of this page, against the pages themselves rather than against our own earlier text, found four more things to fix. They are worth naming, because the correction is the evidence that the check was done. The windows were described as covering six countries; the WHO's own paper says the motor data came from five sites and 816 children, with Brazil excluded. Sitting without support was placed in months 4–6 as something a large share of babies do; no CDC or AAP page that opens says that, both put it later, and the claim has been dropped rather than softened. Walking was described as becoming typical across months 13–18; the cited table carries no median at all, and the WHO's percentile table puts the median at 12.0 months, so that sentence has been replaced with the percentiles themselves. And the line saying babies who shuffle or roll instead of crawling are a known pattern was dropped in favour of the AAP's own wording, because the AAP describes scooting and slithering as the alternatives and treats rolling as a style of crawling rather than a substitute for it.

Several other claims from the older version have been dropped rather than replaced: a figure for how many parents the CDC's milestone checklist reaches annually, and a JAMA Pediatrics statistic about first words. Neither could be traced to a source we could open. No substitutes were invented, and this article says nothing about first words as a result.

Sources

• American Academy of Pediatrics, "Developmental Surveillance and Screening" — source of the screening ages (9, 18 and 30 months), the autism screening ages (18 and 24 months), and the recommendation for surveillance at every health supervision visit. aap.org
• WHO Multicentre Growth Reference Study Group, "Windows of achievement for six gross motor milestones" — Acta Paediatrica Supplement 2006;450:86–95. Source of every window in the table above, read from WHO's own published table. who.int (PDF)
• WHO Motor Development Study, the paper behind that table — 816 children in five sites (Ghana, India, Norway, Oman and the USA), assessed monthly through the first year and bimonthly to 24 months. Source of the five-site sample, the healthy-children selection, the standard-rather-than-reference framing, the statement that each window is bounded by the 1st and 99th percentiles, WHO's own recommended use for the windows, and the 4.3% who did not crawl on hands and knees. who.int (PDF)
• WHO, "Percentiles and mean in days and months for six gross motor milestones" — the percentile table from the same study. Source of the walking-alone median (365 days, which the table converts to 12.0 months), of the 13.1-month 75th and 14.4-month 90th percentiles, and of the 5.9-month figure for walking with assistance that disagrees with the windows table's 6.0. who.int (PDF)
• CDC, "Learn the Signs. Act Early." milestone checklists — source of the social smile at 2 months, of the definition of a milestone as something most children (75% or more) can do by a given age, of rolling, reaching and mouthing on the 6-month list, and of sitting without support appearing on the 9-month list. cdc.gov (2 months) · cdc.gov (6 months) · cdc.gov (9 months)
• American Academy of Pediatrics, HealthyChildren.org, "Movement: 4 to 7 Months" and "Movement: 8 to 12 Months" — source of the six-to-eight-month sitting description, the nine-month flag to raise with a pediatrician, and the description of babies who never crawl. healthychildren.org (4 to 7 months) · healthychildren.org (8 to 12 months)
• Bright Futures / American Academy of Pediatrics, "Recommendations for Preventive Pediatric Health Care" — the periodicity schedule, which sets out the well-child visit ages the screenings above attach to. aap.org (PDF)