OK - here's a bit of a crash course on Wi-Fi fundamentals:
Wi-Fi is two way radio like walkie-talkies, not one way radio like television. All Wi-Fi devices (phone, tablet, computer, printer, router, Access Point, anything that says "Wi-Fi,") are both a receiver and a transmitter. The vast majority of Wi-Fi is deployed using an Access Point and a group of client devices Associated with the AP (there's another version, but it's much rarer.) Each AP and it's group of Associates Wi-Fi clients could be considered a single "cell" (this is a rather inaccurate characterisation, but it will do for the purposes of this explanation/debate.)
There's a Wi-Fi AP built in to a SOHO "Router" but often the terms "AP" and "router" are used interchangeably, though it is also rather inaccurate. This isn't just hair splitting over nomenclature - in data networking an "AP" and a "router" are very different things. The get-you-on-the-Internet omni-box most people have at home that they call a "router" (AKA HomeHub, SuperHub, et al) has a Wi-Fi AP built in along with a load of other stuff.
In each Wi-Fi cell there is a prevailing rule that "only one thing at a time can transmit." The more "things" there are, the more data they need to ship, the more competition there is for some "air time."
For a simple metaphor, think of it in terms of sound - imagine a dinner party with the same rule that "only one person at a time can speak" - the more guests, the more chatty they are, the more air time competition there is and longer any individual conversation between any two given guests will take as everyone's conversation has to be "interlaced" with everyone else. Same for Wi-Fi (except all conversations are client-AP or "guest-host" and never "guest-guest" - "guest-guest" conversations have to be "relayed" through the "host" - "guest-host-guest" as it were.)
How could we improve the "throughput" of the party..?
One way would be to split our diners into two groups and sit them in separate rooms, thereby each "group" (cell) can "speak" concurrently with the other: There's still an "only one person at a time can speak" rule within each group, but now that each group is smaller, the air time competition within each is reduced and because both groups converse independently of each other, and concurrently with each other, the "throughput" of the whole party is increased (roughly doubled.) Though of course there's no reason to split the groups into even numbers - one group could be larger than the other with obvious consequences for the throughput of each group.
We do the same for Wi-Fi. On a big site with hundreds/thousands of users, we split it up into many cells (lot's of AP's.) As you might imagine, it's a bit of a black art having to "guess" how many devices are likely/desirable in each cell and thus how may AP's (cells)) we put up.
Back to my dinner party metaphor - another way we can "split" the party in two is instead of placing two groups in separate rooms with walls (sound barriers) between, instead we do so in the frequency domain. We have one group talk in a high squeaky voice (high frequency - let's call them 5GHz) and one group speak in a low bassy voice (low frequency - let's call the 2.4GHz.) If we arrange our voices and ears to "tune in" to the appropriate frequencies and "tune out" the other, we effectively "cannot hear" the other group (think of bats and people - sorry my metaphor is getting a bit out of hand here.) So we've managed to again split up the party into two groups that can converse concurrently without breaking the "only one person in the group can speak" rule, but without having to physically separate them. Guess what - same for Wi-Fi.
So the "advantage" of naming your 2.4GHz and 5GHz groups differently is that you can elect which group to join in with from your Wi-Fi device. If you name the groups the same (SSID) - you don't get the choice, the device will effectively decides for you (subject to a few rules and protocol compliance.) With same SSID, there are still two "groups" of devices, a 5GHz group and a 2.4GHz group, and each group is "conversing" concurrently with the other - it's just that you've got no choice as to which group you join in with - the client chooses for you.
There are some nuances...
If all my dinner guests decided to join the "5GHz" group, there's no throughput improvement as the whole party is in there just as it was without separate groups. Whereas, if we persuade half the party to join the 2.4Ghz and half the 5GHz group, then we get some throughput improvement overall.
Of course, if your network (party) is rather light on conversation, it'll make little if any difference as there was not much, if any, "air time" competition in the first place. But if there's lot's of traffic, and air time competition becomes a "problem" then splitting up my party into smaller groups can help ease (but not eliminate) the air time contention.
Of course, being complex technology, with real Wi-Fi there's a few "complications." Sometimes you get Hobson's choice over which group to join. As I've mentioned, certain Wi-Fi protocols only work in certain frequencies, so you have no choice as to which group you join. So try as hard as you might, a B/G Wi-Fi client device will only ever join the 2.4GHz group as such as device cannot "hear" and "speak" 5Ghz. Vice verse for A/AC devices (though A/AC (5GHz) "only" devices are practically unheard of.)
So some people like to try and "distribute" their client devices between the wavebands (groups.) For example, high speed data hungry devices like video units in the 5GHz band, (protocol, coverage, etc. permitting) and low speed devices in the low band. There's plenty of use cases one can think of.
To be able to select which client goes in which group, you need to name the SSID's differently for each waveband. If not and they are both named the same, then the clients will automatically choose which group to join using criteria established by the devices designer (how to choose is not mandated in standards.) Often they try to get the "fastest" they can, and fall down to slower (older) protocols if they can't get a faster one.
Most people can't be bothered with the "faff" of choosing themselves, so ISP kit tends to ship with SSID's the same and rely on automation to make the choice. But that may not lead to the "ideal" distribution of client devices between the wavebands, so some people like to control it themselves.
I'm going to be offline for a while, but post back if you want further discussion - if I don't pick it up, there's others here that know this stuff and will doubtless continue the discussion.