Ferber with twins and multiples

Sleep training one baby is hard. Sleep training two is two of those at the same time. The protocol does not change much for twins and multiples, but the practicalities are very different. This guide is the practicalities.

The framing

Most twin parents assume the babies will wake each other. They sometimes do. In practice, twins who have always shared a room sleep through each other’s noise more reliably than parents expect, because the noise has been part of the night since birth.

The Sleep Foundation guidance on twins is straightforward: unless one twin is unwell, run the same protocol on both at the same time. Splitting the arc is almost always slower than running it together.

Same room or separate

Three options. Pick the one that fits your living situation.

• Same room, both cots. The most common setup. Run the protocol on both at once. Wakings are correlated more than parents think but not so correlated that you cannot manage. About a third of twin families say the second baby slept through the first baby’s Night 1 crying.
• Separate rooms. If you have the space. The arc runs cleanly on each baby independently. The check-ins double, which is the harder part. Plan two timers. With the app, that usually means one iPhone per baby.
• Same room, both cots, with a divider. A folding screen or a tall blanket-rack works. This is the worst-of-both-worlds option in practice: the babies still hear each other but the visual separation does not help much. Skip this unless you have a specific reason.

If you can pick, same room is usually fastest. The babies are both running the same arc, the rhythms sync within two or three nights, and you have one timer to think about.

Who cries first

Some parts of an arc happen at the same moment for both babies (you put both down drowsy but awake, you start the timer, you walk out). Other parts cascade (Twin A starts crying, Twin B was settling, Twin B now restarts).

The way to handle this:

• Run the protocol on whichever baby is in their first wait. If both are crying, the first interval applies to whichever started first. The other baby joins that timer.
• One check-in per visit, regardless of how many babies are crying. You walk in once, you check in on both, you walk out. Same words. Same duration.
• Restart the timer on the longest active wait. If Twin A is at the five-minute mark and Twin B has been crying for two minutes, the next check-in is in three minutes (Twin A’s wait), not five.

This sounds complicated and is, in practice, simpler than it reads. The app handles the timer logic so you do not have to think about it.

Mid-night wakings

The classic twin problem: Twin A wakes at 3 a.m. and starts crying. Twin B was asleep, now wakes too. You now have two crying babies and only one of you in the hallway.

Run the protocol on the baby who woke first. The check-in covers both. If the second baby falls back asleep before the first, that is fine. If both stay awake through the sequence, that is also fine. The intervals are the same.

If a twin gets a different baseline than the other (one settles fast, one takes longer), do not adjust the protocol per twin. Run the same waits. By Night 3 the slower baby usually catches up.

How long does the arc take

Most twin arcs run a night longer than a single-baby arc. Where a single baby’s turning point is Night 3, twins are often Night 3 or 4. Where the arc is normally seven nights, twins are usually seven or eight.

The Huckleberry summary notes the same pattern. The reason is straightforward: twins have more opportunities to disrupt each other in any given night, so the protocol takes one extra cycle to take.

That extra night is the only meaningful cost. The endpoint is the same: two babies sleeping through.

Two adults, one parent on duty

The standard rule applies, even more strongly with twins: one adult does the check-ins each night. The other does not enter the room. Mixed check-ins teach the babies that the protocol depends on which adult is on duty, which means the protocol is no protocol.

Trade nights if you both want a turn. Do not trade mid-night.

If you are a sole parent during the arc, twins are still doable, but you will want a friend or family member to be on call for the first three nights in case both babies need a real intervention at the same time.

Triplets and beyond

Same protocol. The cascade is bigger. The arc is usually one or two nights longer again. The check-in pattern stays one visit per interval, covering all babies.

The single biggest help with triplets is the cot layout. Place cots so you can see all three babies from the doorway and reach all three with a short, calm walk-around. Long check-ins are the failure mode.

What does not work with multiples

• Sleep training one twin at a time, sequentially. The wakings of the trained twin are still triggered by the untrained twin, and you end up retraining repeatedly. Run them together.
• Different protocols per twin. A “we’ll cry-it-out the easier baby” approach almost never works. The babies share a room and synchronise more than parents expect.
• Trying to soothe one while letting the other cry. This breaks the protocol for both. Run the protocol on both, even when it feels like one needs more help.

Two practical bits

• Stagger feeding by ten minutes during the routine. Feeding two babies in series in one bedtime window is fine, and it lets you do bath and pyjamas in series too. Keep the windows short.
• One white noise machine, in the centre of the room, at moderate volume. Two machines at different settings produce a confusing sound field.

What the app does for multiples

The Ferber app runs one timer at a time. For twins, the simplest setup is one iPhone per baby, usually one per parent. With Apple Family Sharing, a single purchase unlocks the full program on both phones.

The big help is that you do not have to hold two schedules in your head at 3 a.m. Each phone runs its own baby’s intervals. You walk in when the alert fires. You check in. You walk out.