Why the school Wi-Fi dies at 8:45 and works fine at 10

At 8:45 the Wi-Fi is unusable. By 10 it is fine, and by the time anyone with a budget walks down to look at it, there is nothing to see. Somebody suggests a bigger internet connection, and a year later the same thing happens at the same time of day.

The morning peak is a density problem. Bandwidth and density are limited by different things, and that distinction is the whole article.

Count clients per access point, not megabits

An access point can only hold a conversation with a finite number of clients, and how much internet sits behind it changes nothing about that ceiling. Thirty iPads in one room on one AP behave very differently from the same thirty spread across three, because every device has to take its turn on the radio and the queue is per AP. Ask your controller or MDM for the peak client count per access point at 8:45, not the daily average. The APs carrying a crowd at the bell are your answer.

Slow clients cost everyone else airtime

A device connected at a low data rate holds the radio longer to move the same amount of data, and every other client on that AP waits while it does. One aging laptop, an old printer, or a projector sitting on 2.4GHz can absorb a share of the airtime out of all proportion to what it is actually doing. This is why replacing a handful of old devices sometimes fixes a wireless complaint that no network change did.

The morning burst is authentication, updates, and sync all at once

At 8:45 every device in the building wakes, associates, authenticates (usually 802.1X against a RADIUS server), checks for iPadOS and app updates, and syncs whatever cloud storage the school uses. That all lands inside the same ten minutes, on top of the lesson starting. By 10 the same fleet is quiet in the background and the network looks healthy, which is exactly why the fault is so hard to demonstrate to anyone.

More bandwidth does not fix a density problem

Doubling the internet circuit changes nothing about how many clients an access point can serve or how airtime gets shared in a classroom, and schools pay for that upgrade regularly. Bolting extra APs onto the ceiling without a plan is the second version of the same mistake, because badly placed radios on overlapping channels make co-channel interference worse rather than better. What settles it is a site survey with measurements taken in the actual rooms, which tells you where the access points go, how many you need, and which ones to move.

If you only do one thing

Get the peak client count per access point at 8:45 before you buy anything. That one number tells you whether you have a density problem or a bandwidth problem, and they have different bills attached.