Introduction
The solution I will present is something I posted about in WoW forums in 2012. Back then it was a solution adressing PvP balance being neglected since the game designers did not want to affect PvE. But, the solution directly affects PvE balance as well. During this time, a very few selection of classes dominated both PvP and PvE. Just like today (haha), but to a much higher extent). After the change, class balance improved dramatically in comparison to the past.
The system was ignored until late 2017 until blizzard came to the exact same conclusion as their dedicated hardcore playerbase. That the only way to make balancing their 36 "classes" less of a nightmare, they had to seperate PvP and PvE balancing. Now don't panic! This is not as preposterous as you might think! This can be done in many different ways, and I believe it can indeed be done in ways that pleases almost everyone.
Below i will try to type my heart out to combat the class balance/variety dilemma as best as I can. I will try to solve the problem by looking at what worked "best" for World of warcraft during their balancing battles between 2006-2021. It is always important to learn from others mistakes and succeses. I would love to hear your thoughts and I think IS would too! Just try to keep it respectful and civil.
The solution will mainly adress problems that arise when trying to keep the game balanced in PvE while also trying to do the same with PvP (and vice versa). It directly solves the issue when PvP/PvE balancing affect each other in negative ways. It also removes any reason for the game designer to worry about how specific changes will affect PvE/PvP.
Background
As many others here, I have introduced a relativly large bunch of people to the game. Mainly friends and guildmates from WoW and RS. Everyone I introduce quickly jumps to class related questions. This is understandable since they are a vital factor in an MMO.
Almost in all cases, immediate concern is voiced about balance and/or variaty. 64 classes sounds like quite a lot after all, especially if the game wants to be balanced and/or varied. And they would not be wrong. Afterall, when you break it down, you really have 8 classes with 8 specializations each. And for a WoW player, alarm bells immediatly go off. Ill explain why.
If you would try to compare WoW and Ashes, you could say that
Ashes of creation has 64 classes
World of warcraft has 36 classes
Now, I say WoW has 36 classes because that is technically what they have. You are not a "mage", you are either a frost, fire or arcane mage. You can't be nothing. In AoC terms, WoW has 12 archtypes and 36 classes. Even the dual class system is there. A paladin is just a priest and warrior and so on. It is currently impossible to know
Why is all this boring stuff about WoW important? Well, to put it bluntly, because blizzard proved that by using the standard MMO balancing strategies and systems (which will be used in AoC), it is probably impossible to balance an MMO if it is to be varied. WoW is especially important to look at and learn from because they have had the largest competitive MMO PvE and PvP scene. This gives us a lot of data to work with. While you can never achive complete balance and still have a varied MMO, I think the solution I will present may help a great deal to make it as varied and balanced as possible, while not affecting the game for anyone in any negative way. I also believe this will relieve great stress and mitigate typical balance problems game designers almost always face.
What is the problem, why is it so hard to balance PvE/PvP?
When a game offers PvE and PvP content with the intention of keeping them both high quality, problems and dilemmas unveil themselves. The root of the problem lies in that PvE and PvP are two completly different types of gameplay and systems. You can not, and should not use the same TOOLBOX when working on two completly different machines.
Just like in real life, a car mechanics toolbox is vastly different than a carpenters. Because they are working on two different things. Now, tell me. Why would a game designer, use the same toolbox when working on PvP and PvE class design, when they are completly different things. Even within the same game, PvE and PvP is so different on all aspects of gameplay, that they could even be considered their own games inside the other "metagame".
Just tell me exactly what the damn solution is already!
The solution, that has vastly improved the class balance in other games for both PvE and PvP is to balance them seperatly. Use two different toolboxes for two different machines if you will

What i mean by that, is that the game designer should make abilities, auras, passives work differently in PvE and PvP. They can also make abilities that have no optimal use in PvE, but are vital in PvP (and vice versa). Yes, I am well aware that Ashes will already do this. In the form of PvP auguments and enchants. This has been confirmed by Steven. But, I am not sure these PvP augments and enchants have the purpose to make the game more balanced.
How would this work exactly?
*Example 1:
Assume there is a class with a powerful damage reducing ability. The ability puts up a barrier that reduces damage taken by 25% for everyone in your group standing inside it. It could look like this:
Dome of light (3 min cooldown)
Instant cast
Summons a barrier to protect all allies at the target location for 10 sec, reducing all damage taken by 25%
This is a very powerful ability that is also very fun to use. Raid wide damage that would be critical, will no longer be critical. It has many uses in PvE, but also in PvP.
Assume that this unique ability is what makes this class esecially good, unique and desirable in PvE boss fights. It's their ace. Almost mandatory for groups in hard, telegraphed AoE fights if you want to do them smoothly. Even though this class may heal about 5-10% less than other healers, this ability still makes it important, unique and viable. This class is regarded as top tier in PvE.
Now also assume that this class is bottom tier PvP. It has almost no representation in arena and sieges. In many cases, it is simply seen as a worse version of other healers.
After about a month of data gathering from Sieges, and arenas, in combiation with the PvP community feedback, you think you have found the issue. To be extra sure, you also consult with your official PvP community council to verify.
The game designers, togheter with the PvP community concludes that the issue is that:
1. The class does not provide enough group utility to make it worth bringing in serious PvP.
2. You also conclude that the class dies way to quickly even when using the tools it has to its disposal.
3. It is also bad at keeping it's friends alive during enemy burst windows.
This class, does indeed, need a buff on these areas to become viable again! But, the developers quickly realize they simply can't buff this! Why?
The infamous dilemma
Because buffing it for PvP purposes would make it even better, and maybe too strong in PvE. This applies in a world where PvE/PvP stats and abilities are exactly the same in PvE/PvP
This class is already working great and as intended in PvE. People feel it is unique, fun and based on statistics, well balanced in comparison to other healers. It would be very stupid for the developers to change it's PvE performance at this time. Since the community loves it so much and it is well balanced.
If only the developers could make some adjustemnts for the PvP side of the class, without completly breaking its PvE side!
The two toolboxes
The game designers decide to break previous philosopies and ideas and make a change that will only affect PvP. PvE players would not be affected in any single way what so ever. While it would make the class more balanced in PvP.
And they do it, by making an ability specific PvP auguement that changes how the dome of light ability works only in PvP. It would look something like this:
Dome of light PvP augument (Call it something cooler)
Reduces the cooldown of Dome of Light by 50%.
Dome of Light reduces damage taken by an additional 25%.
In the upcoming patch test it works. The patch testing community notice that previous problems are mostly solved. The class is in a much better state PvP vice. And it does not affect PvE players in any way at all! Everyone is happier. Hell, it even made the game designers life easier! The game designer did not even have to think how this change would affect the PvE content of the game!
Conclusion
By using the game design system I layed out. You can fine tune PvP without the game designers having to think about how it will affect PvE balance. You can work on these two entities seperatly, making them into their best versions without affecting the other in negative ways. This system can obviously be used the other way around, where certain auguments do not work in PvP, but do so in PvE.
My question to those that disagree
Since this way of doing things will improve PvP balance, without affecting PvE balance, content and players in any way shape or form, potentially opening up more variety between classes, why would you stand against this?