Monk isn't bad for just doing casual content, and it's totally useable for Savage and Ultimate. While I'm definitely not the best Monk in the world, I've done all of battle content in Shadowbringers, and I've done it with Monk before any other job. I can say with absolute certainty that none of the problems people have with it are a matter of it's performance, there's a handful of fights this tier that it's weak on but there's as many it shines on so it even out. The problem is that it's entirely because of its design being unsatisfying to play. It's that it's extremely lackluster to the point where it feels like a beta version of Samurai. Despite being a standout in ARR, the skills and traits that have been added from expansion to expansion since have been varying degrees of flawed and despite player feedback being provided, most of the problems have never been addressed. Part of this is because of the dev's belief that the job was the most "complete" of the jobs made in ARR which is a fair assessment. However how complete it was in ARR has no bearing on how satisfying or complete it is now three expansions later. Indeed much of what has made it work in ARR has been systematically undercut by shifts in overall design that have steadily chipped away at what was once a strong core and weakened the job as a whole.
Now here's a slightly edited version of an essay I posted on Reddit on the subject of peoples dissatisfaction with Monk.
To get into why people are frustrated with Monk you really need to look at it as it's evolved from expansion to expansion and how the devs have handled it, or specifically how they've failed to handle it properly. Part of the frustrations people have with it are just as much an aspect of how the devs are exceptionally slow to implement feedback with the job or the devs seemingly not taking feedback at all than they are with new changes in direction.
In ARR Monk was arguably the best designed job in the game with GL, form based combos, and extra GCDs that didn't break its dots. It had some problems like button bloat in Fist Stances that you only ever wanted to use Fire Stance for the DPS increase, Haymaker (attack on a dodge), Featherfoot (Low chance to Dodge. Would pair well with Haymaker but Monk wasn't a tank so they never saw use), and One Ilm Punch (A dispel resisted by everything) but otherwise it was probably the most well designed job in the game. In Heavensward you basically see the devs draw inspiration from Monk for other jobs by creating BotD on DRG, Enochian on BLM, and giving NIN Armor Crush to extend Huton. All of this is fine, but when it came to Monk's additions in Heavensward it was fairly lacking, with a number of skills whose use cases were extremely limited Purification as a TP regenerator with requirements that made it useless, and Tornado Kick as a means to dump GL3 but with potency so low it was rarely justified. On top of that the job was plagued by problems like TP running dry faster than other jobs, no means of limiting aggro, a lack of useful party buffs (We had Mantra, but in Heavensward it was a joke and the other Melee could cross class a lesser version of it), poor recovery due to the extremely long cool down of Perfect Balance, it's burst damage was nonexistent, and it had extremely low personal DPS beside (by the end of the expansion everything beat it in personal numbers other than Ninja, and everything beat it in rDPS). None of these problems were fixed by the end of the expansion and the devs at some point said something to the effect of "We're done making changes in Heavensward and wait till next time." And Monk players were hopeful. Stormblood was going to have Monk as the poster job and the biggest draw to it was supposed to be an action rework that removed useless or overly Niche actions, something that Monk had in greater numbers than any other job. It felt like the expansion was tailor made to fix Monk's problems.
Come the Stormblood media tour and Stormblood proper and things actually looked fairly grim for Monk players. While certain problems were fixed, they were really only fixed as parts of broad shifts in gameplay like aggro management being de-emphasized or TP becoming a non-factor in single target. However they absolutely failed in removing useless actions, we kept the Fist Stances, Purification (a TP regenerator tied to Chakra in such a way that it was unusable), and One Ilm Punch, an action so useless the devs themselves had mocked it. What they did remove was Touch of Death, Fracture and other skills we actually made use of to buffer our nonlinear combos or if we needed to miss a positional. These skills were critical to the feel of the job because being able to use them without breaking our combos were critical to the feeling of a free flowing martial arts job. It also turned out that certain things we though would be fixed like Aggro management, actually weren't. Monk struggled with Aggro problems until 4.3, where they gave Monk an aggro cut... by tying it to Purification and requiring Monk to sacrifice a full gauge to use it which was an enormous DPS loss when every other job got their aggro cut for free (except Samurai, which cost a Third Eye proc, which basically meant it was free).
Meanwhile the additions to the kit were pretty much widely derided by the playerbase, with Riddle of Fire having a slow that went against the fast GCD of GL3 and disrupted the job's flow. In general it just felt like a worse replacement for Blood for Blood which we lost with the removal of cross class actions. Deep Meditation was implemented as a means to use Forbidden Chakra and as our gauge, but more, except its double gated RNG nature was poorly received for breaking the consistent flow. The Chakra Gauge itself was extremely underdeveloped with only a single target attack (Forbidden Chakra) and Purification (the TP regenerator on a 2 minute cooldown), but we were missing any other uses like say, an AOE use which was critical to have in dungeons. Brotherhood had the same RNG complaints as well as additional problems, it tied monk to 8-man physical compositions for its window to provide enough Chakra to use a Forbidden Chakra, because Forbidden Chakra can only be used at a full gauge. It was also prone to overflow in the event that RNG was actually good due to its cooldown, so you'd just lose procs. As a result, the skill basically is never satisfying. Either you get too little to do anything with in most solo, 4-man, or 8-man duties with bad comp or bad RNG, or you have good RNG and it overflows. There was also the much mocked "Tackle Mastery" which attached bonus effects to Shoulder Tackle based on the Fist Stances of which only one was useful, the one that increased damage. The other two, knockback in Earth Stance or a gap closer on Fists of Wind but only within 5 seconds of using Shoulder Tackle the first time, were the subject of much mockery. We also got Riddle of Earth, a skill to refresh Greased Lightning in one niche instance that was prone to failure because it required taking a hit. So it could get blocked by a strong enough Shield by a scholar or just a lack of time on the GL timer between disengage and the AOE going out, and it didn't work when there was nothing to be hit by or you were supposed to dodge. Monk's changes were near universally hated at launch and the devs said in the 4.1 Q and A they'd fix certain problems if people complained enough, but even though people complained the devs... didn't fix anything they introduced in Stormblood at all beyond Brotherhood overwriting other Brotherhoods.
Enter 4.2, the devs finally decided to fix one problem that we'd complained about...two years prior in Heavensward. That our recovery was terrible because Perfect Balance was 3 Minutes long (Imagine Blood of the Dragon or Enochian being on that long of a cooldown basically). They reduced PB to 60 seconds and added an effect to Wind Stance Tackle Mastery that granted a GL stack. Monk finally had decent recovery, but then people realized this recovery could be used to turn Tornado Kick into a source of huge burst damage and a new rotation unintended by the devs appeared; the TK rotation. Some people really liked it because of how fast paced it was and that it used every part of the kit on some level. Others didn't like it because they found it clunky and against the playstyle they wanted from the job, but at least appreciated that TK saw use for the first time since it was added to the game. The majority of people didn't even know it was feasible because it was extremely unintuitive. From there on Monk didn't really see any change until 4.5 where it was given a token buff of 1% to Fists of Fire to try and coax people to play it and more promises that they'd make major changes and rework it for Shadowbringers.
So just as a tally of problems Monk had coming up on Shadowbringer's launch. From ARR/HW there were an abundance of useless skills like the Fist Stances or overly situational skills like Tornado Kick/Riddle of Earth. From Stormblood we had Monk's basic design, with Riddle of Fire's slow and an abundance of double layered RNG and comp dependency.
So what happened in Shadowbringers after two consecutive expansions of promises they'd make changes later?
They in essence, re-released Stormblood Monk, and not even a cleaned up version of TK Monk, but the version that everyone hated on launch. The only addition to the actual moment to moment gameplay was the addition of Greased Lightning 4 because people kept mentioning it in interviews I guess. Riddle of Fire kept its slow, the double layered RNG of Deep Meditation remained and got a new trait that made it 70% chance on a crit instead of 50%, Brotherhood's RNG and composition dependency remained. They kept the Fist Stances and added two new traits to them that still didn't fix the basic problem of there being one dominant stance that increases your DPS the most. There wasn't even a new skill we could use in our rotation, just more skills designed for Niche use. Our existing kit also suffered, we lost additional skills we used regularly like Internal Release which mitigated some of our Crit Struggles, Howling Fist and Steel Peak. They nerfed Perfect Balance and killed Monks ability to recover, making dungeons miserable to play. Tornado Kick kept its effect and was once again useless.Dragon Kick, a core part of our rotation, was tweaked because physical resistance buffed were removed and gained the Leaden Fist effect which put a disproportionate amount of potency on Bootshine and emphasized using Perfect Balance to alternate DK>Bootshines instead. The best additions we got were GL4 which was tainted by association with the Fist Stances, and Enlightenment, an AOE gauge spender which was tainted by basically being a repackaged Howling Fist we couldn't use in single target. The rest was... bad. Six Sided Star is an okay disengage tool which refreshes Greased Lightning but is about as lackluster as Shadowbringers Launch Shoha, and Anatman not only created the most cancerous opener the game has ever seen but its intended use was about as niche as Riddle of Earth for Stack upkeep since you couldn't move during it. Anatman doubled as a GL recovery skill probably to justify the PB nerfs, but because it built one stack every 3 seconds so it was by design bad at that too.
Naturally it was the worst received of the DPS by a long shot with people who liked the TK rotation, people who didn't like it, and people who didn't know about it all being unhappy. They had promised major changes to fix our problems (even mentioning a rework in a few interviews IIRC), and what we got were those same problems we complained about in Stormblood. All repackaged alongside ridiculous statements like Yoshi saying his favorite job to work on was Monk and that it showed the dev team's growth during the media tour and specifically identifying aspects people didn't like from Stormblood and why, but still not fixing them (in the Larryzaur interview). I don't think he actually believed that of course, it was pure PR speak, but the damage was done. A number of hasty changes were then made in 5.05, they removed Riddle of Fire's slow 2 years after people asked them to being the biggest, and they gave Monk basic GL upkeep equivalent to BotD or Enochian Upkeep with Formshift refreshing GL, but that had the side effect of making the many GL upkeep skills pointless. The changes had a very short term effect on people playing the job in Eden's First Tier of Savage, but it faded very quickly because the devs didn't fix many of the fundamental problems people had with it.
Case in point two patches later, Monk was the least satisfied in the recent poll on satisfaction, it was the least leveled job according to the Lucky Banchou census, and according to FFlogs reports at least it's the least played job in all content. While I think it would be remiss to claim that Yoshi's statements in the media tour suggest anything about what they thought of the job due to the previously mentioned PR talk. It does show a severe disconnect between the developer perception on how to evolve it and what the playerbase actually wants. Typically game developers don't want aspects of their jobs to be abject failures in the way that Monk for all intents and purposes is and has been, and by extension I think the devs actually believed that what they did for Shadowbringers would actually make people happy. They were just completely wrong.
Most of the problems we've had from Stormblood or earlier still remain with an abundance of overly niche skills, composition dependence due to crit RNG and Brotherhood, the RNG for both Deep Meditation and Brotherhood being disliked to begin with, the existing design that was once the most complete has been systematically weakened by how they've adjusted skills, and now there's an extreme distrust of the devs because they've proven multiple times that when they say they're taking feedback for Monk they actually aren't.