Erbit wrote: I've already explained it. If it is encrypted traffic (e.g. HTTPS) then no. Another move, unfortunately, and as I wrote Wireshark is the answer.
Of course it can. First of all, Colleagues forget that "network traffic" or "network activity" does not mean the necessity to eavesdrop on the CONTENT of packets. Often - in OSINT cases it is enough (/ gives quite interesting results) the mere analysis of CONNECTIONS (i.e. who, when with whom), web searches (e.g. "how to make a bomb with home methods + attack a neighbor"?) And,
2) which may not surprise everyone - analysis of queries to DNS servers. The vast majority of these are open (they fly in an open, unencrypted connection).
3) Sometimes the analysis of connections (those already encrypted) can also tell a lot, or rather their metadata (unencrypted), i.e. packet headers, often a real mine of useful information ... Of course, for further processing and planning a wider ... reconnaissance

4) If the company uses traditional e-mail (traffic to / from non-web clients), then another thing - over 95% of transmission using e-mail protocols is unencrypted (the use of PGP in Polish companies is really marginal). *) We discussed with the rest not so long about this wonderful / sad fact wider on Niebezpiecznik on a few recent mishaps (including software for box routers or prank with the President in the background)
5) Etc. / etc
It all depends - so on what - and how much determined 'opponent' we have before us. in other words, WHOM we let into our internal network ...
Personally - if I could attempt any advice in this respectable group - I would suggest separating the two matters.
1) internal network for the company and its critical interests (such as servers, employee workstations, communication, e.g. VoIP type, etc.). the same for the home network - a man who wants to keep some confidentiality.
2) internal networks for customers / guests.
How to do it ? There are several (a dozen) ways, but the simplest one is to separate both networks (corporate and for customers) into separate subnets, then set separate VLANs for them and watch over who we connect where / to whom we give access, where. More radical (but inherently safer) and more costly is the physical separation of infrastructure to support both of these networks. So separate switches, APs and everything connected with some good (sensible) mikrotik / qbiquity class router. Then, performing any network traffic diagnosis / analysis, although still possible, is much more difficult.
best regards
*) - of course, you cannot forget about the STARTTLS standard, which is more and more often enforced on e-mail clients by hosting companies