I will attempt to answer that in as non-technical a way as I can, which may make some of it not exactly correct but more clear. First, I must say that it is probably better to read the basics than to watch videos.
You should start with trying to understand it in IPv4 (and later move to IPv6 if desired), which uses 32 binary bits to define an address -- some of those bit define the network and some define the hosts on that network.
It is now a part of CIDR (classless inter-domain routing), haviing evolved from an older way like everything in tech. If you are a precocious 5 year old read that entry (CIDR) on Wiki for an overview.
Subnetting allows you to alter a typical single 256 address address network into small subnets that break the 256 addresses into smaller groups (keeping a total of 256 total addresses between all subnets). Why 256 you ask, well everything here works in binary so 2^8 as a single network as you know it from home routers uses 8 bits to define the hosts.
Look at a typical home user IP address like 192.168.0.1 -- the groups divided by dots are called octets, and that is for a reason. Each octet is actually defined by eight binary digits, so the 4 octets times the 8 binary digits gives you 32 binary digits and in the home network (aka /24) 8 of the binary digits - the final octet allow 256 addresses. 192.168.0.0 to 192.168.0.255 (but of course you can only really use 254 as 2 are reserved for specific uses), and usually your router will be .1 and user addresses can range from .2 to .254).
Then look into supernetting, where you can "borrow" some of the 32 address binary numbers to "loan" them to the host numbers so that a single network can have more than 256-2 hosts and can actually have lots more. There are reasons for doing that but it may be less efficient for a number of reasons.
Good luck.
Facebook
Twitter
Instagram