Feeding

Baby-led weaning or purées: what actually differs

Both approaches work. Neither has been shown to produce better weight, iron intake or less fussy eating than the other, so the honest difference is practical rather than nutritional. Readiness signs, the six-month starting point and the choking rules are identical either way — and most families end up combining the two within a month.

Published 17 August 2026 · 11 min read

The question is usually asked as though one approach is correct and the other is what uninformed parents do. It is not that kind of question. Both get a baby from milk to family food, both are endorsed by health bodies when done sensibly, and the differences that survive scrutiny are practical rather than nutritional.

What follows is what each one actually involves, what the research supports, and the parts that are identical either way — which turns out to be most of the parts that matter.

What each approach actually means

Spoon-feeding starts with smooth purée on a spoon, thickens over a few weeks, then moves to mashed and lumpy textures and on to finger foods somewhere around seven to nine months. The adult controls the pace and, to some extent, the amount.

Baby-led weaning skips the purée stage. From the start, food is offered in soft pieces the baby picks up and puts in their own mouth. The baby controls both what goes in and how much. Milk continues as the main nutrition through the first year, exactly as it does with purées.

The word “weaning” is unhelpful in both cases. Neither approach involves reducing milk in the early months — the milk amounts stay roughly flat until solids are established.

What the evidence shows, and what it does not

Comparative trials are smaller and shorter than anyone would like, and families who choose baby-led weaning differ from those who do not in ways that are hard to adjust away. With that caveat, the picture is consistent:

  • Growth and weight. No reliable difference. Concerns that self-feeding leads to underweight babies, and claims that it prevents overweight ones, are both unsupported.
  • Choking. No higher rate has been demonstrated for baby-led weaning when foods are offered in suitable shapes and textures. The food itself is the risk factor, not who is holding it.
  • Iron. Both approaches can fall short, and both can be done well. This is the one nutritional issue worth actively managing.
  • Fussy eating. The most-repeated claim and the weakest evidence. Any association is small and confounded.

Which leaves the practical differences, and those are real: mess, how much a childcare setting can accommodate, whether a parent can sit through a baby gagging without intervening, and how much control the adult wants over the amount eaten.

Readiness is the same question for both

Around six months, and three signs together rather than any one alone — the AAP and NHS describe the same set:

  • Sitting upright with steady head control, in a supported seat
  • Reaching for food and bringing it, or a toy, to the mouth
  • Losing the tongue-thrust reflex, so food goes back rather than straight out again

Baby-led weaning is slightly less forgiving of an early start, because it requires the sitting and grasping to be there rather than only the swallowing. A baby who is ready for purée at five and a half months may not be ready to self-feed for another few weeks — which is a reason to start with a spoon, not a reason to wait.

Age here means corrected age for a baby born early, and the readiness signs outrank the date in both approaches.

What each track looks like by age

The two weaning tracks, on one axisSpoon-fed and baby-led, and the month they stop being differentThe two weaning tracks, on one axisSpoon-fed and baby-led, and the month they stop being different6789101112monthsSpoon-fedBaby-ledSmooth puréeMash, some lumpsChopped family foodSoft sticksto holdSofter pieces+ loaded spoonsChopped family foodthe tracks meet hereThe choice governs about three months. By nine the tracks overlap; by twelve they are the same meal.
Drawn on one axis because that is the argument: the two approaches differ for roughly a quarter of the first year and then converge on chopped family food. What matters most — readiness, iron, the choking rules — is shared from the first day.
AgeSpoon-fed trackBaby-led trackShared
6 monthsSmooth, thin purée; single foodsSoft steamed sticks, held in the fistOnce or twice a day, after a milk feed
7–9 monthsThicker mash, some lumps, finger foods alongsideSofter pieces plus loaded spoons offeredTwo to three meals; iron-rich foods daily
9–12 monthsChopped family food, self-feeding takes overChopped family food, cutlery introducedThree meals and a snack; milk starts to drop
12 months+Family meals, no purée leftFamily mealsSame food as the rest of the table, without salt

The rows converge, and that is the point most method arguments miss. By nine months the two tracks are hard to tell apart, and by a year they are the same thing. The choice governs about three months of a child’s life.

Gagging is not choking

This is the part worth being able to tell apart before the first meal, because the correct response to each is the opposite of the other.

GaggingChoking
What it isA protective reflex moving food forwardThe airway is blocked
SoundNoisy — retching, coughing, splutteringSilent, or a high-pitched squeak
FaceRed, eyes watering, tongue pushing outPale or blue, distress or panic
What to doStay calm, stay close, let them work it outAct immediately — first aid, then emergency services

The NHS describes the gag response in exactly these terms: watering eyes, the tongue pushed forward, retching to bring the food back. It is loud and it looks alarming, and it is the system doing its job. A baby’s gag reflex sits further forward on the tongue than an adult’s, which is why it fires so often in the first weeks and why it fades over the following months.

Iron matters more than the method

A baby’s iron stores from birth run down around six months, which is one of the reasons solids start when they do. This is the one place where a poorly executed version of either approach causes a real problem, and it is worth more attention than the spoon question.

Iron-rich foods that work in both approaches:

  • Well-cooked meat, offered as strips to hold or minced into mash
  • Lentils and beans, mashed or as thick patties
  • Iron-fortified infant cereal, on a spoon or thick enough to load onto one
  • Egg, cooked through, as omelette fingers or scrambled
  • Something rich in vitamin C alongside, which increases absorption

Iron intake tends to be lower in the early weeks of baby-led weaning simply because less food is swallowed. That corrects itself as intake rises, but it is a reason to offer iron-rich foods daily from the start rather than starting with fruit and vegetables for a month.

The combined approach most families land on

Ask parents what they actually did and the answer is usually both. It is not a failure to commit; it is the arrangement that fits how babies eat.

  • Loaded spoons. Purée on a spoon the baby takes and puts in their own mouth. Self-feeding, but the food does not have to hold its shape.
  • One of each on the tray. A stick of something soft plus a bowl of something thick. The baby picks; you can help if they want help.
  • Purée for iron, fingers for practice. The nutritional job and the skill job split between two textures.

The one combination worth avoiding is spoon-feeding into a full mouth or chasing a baby with a loaded spoon. Responsive feeding — offering, waiting, stopping when they turn away — is what both approaches actually share, and it does more for how a child eats at three than the method ever will.

What to avoid either way

The list is identical for both approaches, and it is short:

  • Choking-shaped foods — whole grapes and cherry tomatoes, whole nuts, popcorn, chunks of raw apple or carrot, chunks of meat or hard cheese, sausages in rounds. Quarter, flatten, cook soft, or leave out.
  • Honey before twelve months, because of the risk of infant botulism.
  • Added salt and sugar. Cook the family meal without salt and season the adult plates.
  • Rice drinks as a milk substitute under five, and unpasteurised dairy at any age in the first year.
  • Delaying allergens. Egg and peanut, in age-appropriate forms, are introduced early and regularly in current guidance rather than avoided.

When to ask

Speak to your paediatrician, health visitor or a paediatric dietitian if your baby gags on every texture well past the first weeks, if they consistently refuse all solids at seven or eight months, if weight gain flattens as solids increase, if swallowing seems effortful or comes with coughing at every meal, or if there is any reaction that involves swelling, breathing or widespread rash — that last one is urgent rather than a question for the next appointment.

Questions parents ask

What is baby-led weaning?

Skipping purées and offering soft finger foods the baby picks up and eats themselves, from around six months. The baby controls what goes in and how much. Milk continues as the main source of nutrition through the first year either way.

Is baby-led weaning better than purées?

Not in any way that has been reliably demonstrated. Trials comparing the two have not found a consistent advantage for weight, iron status or later fussiness. Choose on practical grounds — how your baby responds, how much mess you can live with, and what your childcare setting can support.

Does baby-led weaning cause choking?

Studies have not found a higher rate of choking when finger foods are offered in appropriate shapes and textures. What does increase risk, in either approach, is the food itself: whole grapes, nuts, popcorn, chunks of raw apple or carrot, and anything hard and round. Gagging is common and is not choking.

Can you do baby-led weaning and purées together?

Yes, and most families do. A loaded spoon of purée alongside a steamed stick on the tray is not a compromise between two methods — it is what feeding a six-month-old normally looks like. Nothing about mixing them undoes the benefits of either.

When can a baby have finger foods?

From the start of solids at around six months, provided they can sit upright with good head control. The NHS describes a finger food as a piece big enough for a baby to hold in their fist with some sticking out — a baby of this age cannot use a pincer grip yet, so very small pieces are harder rather than safer.

Do babies need teeth for baby-led weaning?

No. Gums manage soft food perfectly well, and the first teeth are incisors, which cut rather than chew. What matters is that the food squashes easily between your finger and thumb, not how many teeth are through.

Does baby-led weaning prevent fussy eating?

It is the most common claim and the least supported. Some studies suggest a modest association with better acceptance of textures, but the effect is small and confounded by which families choose the approach in the first place. Repeated, low-pressure exposure to a food is what predicts acceptance, in either approach.

How much will a baby actually eat at first?

Almost nothing, for weeks, and that is normal. Early solids are practice at handling food rather than a meaningful source of calories. Milk feeds should not drop much before eight or nine months whichever approach you use.

Is baby-led weaning safe for premature babies?

It can be, but the timing follows corrected age and the readiness signs rather than the calendar, and preterm babies more often have feeding or oral-motor issues worth assessing first. This is a conversation to have with your paediatrician or a dietitian rather than a decision to make from a chart.

What should I never give a baby under one?

Honey, whole nuts, whole grapes, popcorn, hard raw vegetable and fruit chunks, chunks of meat or cheese, added salt and added sugar. The list is the same whether food arrives on a spoon or in a fist.

Sources

Read next

General information, not medical advice. Every child develops on their own timeline and the ranges here are typical, not targets. Pixy is a tracking tool, not a medical device. For anything about your child’s health, ask your paediatrician or health visitor.

Track it instead of remembering it

Pixy logs sleep, feeds and routines from newborn to age 12, and predicts the next nap from your child’s own wake windows. That part is free.