The term 'concurrent users' refers to the number of users connected to and using your application at any given time. For example, you could have 2 users connected via a peer-to-peer connection in 50 different 'rooms', or 20 concurrent users connected to an interaction in 5 different rooms. Either way, there are 100 concurrent users connected to your application at that point in time.
For users of the Free Plan on the Skylink Platform, the maximum number of concurrent users supported refers to the maximum number of simultaneous connections between peers that can be maintained at any given time.
If you're familiar with WebSockets, you can also think of a single peer (user or endpoint) as being one socket connection to the Skylink Signalling server. Using the example above, then, you would say that there are 100 concurrent socket connections.
There is an assumption in the explanation above, which is that a connection equates to a single physical user. However, it is also possible that in your application, a single "user" could use more than one connection, and therefore represent more than a single concurrent 'user'. For example, if a user opens multiple browser windows or different browsers and connects those to the same 'room', that user would actually have two concurrent connections.
Other articles you may be interested in: What are Peer Connections and how are they measured?| How is TURN Bandwidth measured and billed?| How is MCU/SFU Bandwidth measured and billed?| What happens when I exceed the included amount of Peer Connections or Bandwidth in my plan?