How to Start Solids

One of my core beliefs — you will not be surprised to know — is that data is key to making our parenting decisions effectively. Often, however, data is most useful to answer the big questions but falls flat on the details.

Take sleep training: There is reasonably good evidence on the benefits of sleep training and its short- and long-term effects. But if you decide you do want to sleep train, there isn’t really any evidence as to what the best way to do it is.

In many of these cases — and this is essentially the central thesis of Cribsheet — the lack of data mostly reflects the fact that it doesn’t matter very much what you do on these details. There are a lot of good ways to sleep train, some of which will work better for some families than others. You want to stick to one, but which one isn’t that important.

That argument is simultaneously freeing and frustrating. It’s reassuring to know that you probably will not mess up too badly. But for some of us, it is paralyzing to try to move forward without a plan.

I would argue that perhaps nowhere is this more the case than in the introduction of solid foods. Data gives us a few hints as to dos and don’ts but virtually nothing on the how. Yet it is the how that is daunting. The first food we gave my daughter was rice cereal. I remember giving her some, taking a few pictures, and noting the milestone. And then slowly realizing that, like, we were supposed to do this all the time now? How was that even going to work?

Below, I’ll start with the (limited) data-based guidance on starting infants on solids and then offer some perspective — and some resources — for making a plan when the data isn’t going to make it for you.

Take me to the bottom line

Basic guidelines for solid food introduction

There are a few guidelines for starting solid foods that are supported by the data. Here’s a simple list.

Babies are ready for solid food when they can hold their head up, sit with support, and start moving their hands toward their mouth. General guidelines are 4 to 6 months, although increasingly, advice from the American Academy of Pediatrics, the CDC, and others pushes toward 6 months. Solid food shouldn’t be started before 4 months.
Before about 12 months, babies will still get most of their calories through breastmilk or formula. The saying “food before one is just for fun” captures this idea, even if it is an oversimplification (some babies do eat more solid foods earlier, etc).
Allergens — peanuts, eggs, dairy, wheat — should be introduced early. Early allergen introduction has been reliably shown to lower the risk of developing long-term allergies. These foods shouldn’t just be introduced early; they need to be eaten with some frequency.
Early foods should have iron in them (babies at this age begin to need more iron than is in breast milk or formula); many cereals are fortified with iron for this reason, and meat also provides a lot.
Do not introduce choking hazards (e.g., hard candies, nuts) to babies.
Tastes are formed by repeated exposure to flavors early on (there is a fascinating study about a taste for carrots being formed by carrot-juice-drinking moms through breast milk, for example). This means that exposing kids to the same foods over and over, even if they do not like them the first time, is a good idea.

In addition to these, there are some logic-based guidelines. Babies do not have many teeth, so it is a good idea to introduce softer foods first.

There are also a lot of things you might be told that the data doesn’t support. In the puree versus baby-led weaning discussion, for example, people sometimes argue that baby-led weaning lowers obesity rates. On the flip side, others argue it leads to more choking. Neither claim is well supported by the data.

There is also talk about early food choices contributing to picky eating. Picky eating is a tricky problem for many families, and there are some data-oriented approaches to addressing it, but any measured impact of early food introduction on this behavior is extremely limited.

You’ll hear that rice cereal is the “perfect” first food or that oatmeal is. The reality is that across the globe, there is an enormous variation in what “first foods” are, and no reason to say one is better than another. Good baby food marketing is probably responsible for the rice cereal recommendation, for example. And, in general, there is an element of cultural randomness in what is considered the ideal first food.

There’s also no strong evidence to support the prohibition on using salt in infant food (in moderation, obviously). And this idea that you need to introduce only one food every three days? That’s more tradition than data-based logic. While it’s true that early on it’s useful to pay attention to new foods in case your child has a reaction, you can do that with a much faster food introduction schedule.

Finally, when we think about the question of what a healthy diet is, as I talk about ad nauseam, our evidence in general is simply quite poor. If you’re looking for data to say your child (or you!) should eat only organic food, grass-fed food, or avoid sugar overall, you aren’t going to find good support in the research.

Looking at what the data does and doesn’t tell us, it should be clear that the information we have would be consistent with a huge variety of approaches to the introduction of solid foods. You can adhere to the data with a traditional approach of starting with rice cereal and working your way through single-ingredient jarred baby foods (with, say, a product like Ready, Set, Food! or Lil Mixins to cover your early introduction of allergens). Or you could adhere to it through a full-on baby-led weaning approach where your child eats a version of what you eat from very early on. Or you could be somewhere in the middle. Or, like me, you could do it completely differently with the first and second child.

The question, though, is where to go from there — how to choose, if the data doesn’t choose for you.

Decide what your approach will be

You, essentially, have two steps: thinking about where you want to go and what’s comfortable for you, and then making a plan. With something like potty training, it is clear where we want to go: to a magical place with no diapers. With food introduction, it’s sometimes less obvious.

Here’s a thought experiment from the standpoint of someone with a 5-month-old: What do you want mealtimes for your child (and you) to look like at 9 or 10 months? Do you imagine eating meals all together? Do you imagine feeding your child separately? If you imagine eating together, do you imagine your child eating what’s on your plate (more or less)?

Answering this doesn’t really speak to the details of how to do it, but it is a useful frame for (say) thinking about baby-led weaning versus other approaches. This question also helps to visualize where you are going in a way that means you, hopefully, will not be surprised, as I was, that solid food introduction doesn’t end when you complete the first rice cereal feeding.

At this point, you should also consider: eventually, you’re going to be providing three meals a day; how do you want to get to that stage? Most people start with one meal a day and then work up, but it’s worth thinking about the next steps.

A closely related thought experiment is simply what feels comfortable and practical to you. This has a million components. Some approaches to food introduction are much messier than others. Personally, I embrace the mess (as my spouse will tell you). But this isn’t how everyone feels, and your own reaction is something to factor in. There are questions of practicality to consider.

Making your own baby food in a food mill may seem great in principle but may not be feasible in practice. There is a lot of room here for making it feel good to you.

Make a plan for food (and some resources to do so)

Just because the data doesn’t make a plan for you doesn’t mean you cannot make a plan. If thinking about the details of introducing foods makes your skin crawl with anxiety, the way out may be to recognize that any plan that adheres to the basic data-based principles will make you feel better.

I’m not talking about anything too elaborate. I’m picturing … a Google Sheet, with the date, what they will eat that day, maybe at what time? I would imagine that you could all make a much nicer, color-coded version than I would. If it were me, I would probably start by planning out a week or two, and then somewhere in the sheet, keep a running list of foods that had been tried. But, again, I suspect you can all do this better.

Or you can rely on someone else to do it for you. There are some very excellent resources that can help, and I’m going to detail a few below.

I want to be clear: several of these are focused on baby-led weaning, and they all come (as do we all) with their own perspective. These resources make sense to me, and I think they can be helpful if they work for you. But I’d suggest you approach them as assistance to your own planning, rather than some type of rigid guidance that you must follow to produce a successful eater. The whole point here is that there are lots of good approaches that are consistent with what we know from data.

Solid Starts: This was the clear winner when I polled people on Instagram. It has a great (free) food database and very detailed (paid) plans for food introduction. If this approach works for you and you want someone to provide you with recipes and a detailed guide, this ticks all the data boxes for me. There’s also an app.
Zayne’s Plate: Adorable and helpful Instagram content, a cookbook, and a free 28-day meal plan (accessible here).
Feeding Littles: Online courses designed to walk you through an introduction to baby-led weaning.

No data doesn’t have to mean no plan. It just means you get to choose your own adventure. And whatever adventure you choose, it will end with oatmeal and bananas smeared around your house in a sticky, glue-like paste. Enjoy!

The bottom line

There are a few guidelines for starting solid foods that are supported in the data; babies are ready for solid food when they can hold their head up and sit without support, generally around 6 months, allergens should be introduced early and often, and early foods should have iron in them.
Beyond that, many common recommendations lack strong data support, including the idea that there’s a “best” first food, baby-led weaning versus purees, salt restrictions, and the “one food every three days” rule.
In reality, the evidence supports a wide range of approaches to starting solids, which means there’s plenty of room to make choices based on what works for your family.