When answering the question "what is good", there are many factors that need to be taken into consideration, some of which this article will attempt to address.
Proper contextualization
What does it mean to contextualize? Servants can only be considered good or bad if they are compared within the context of a specific situation - something that could be good in one situation may be bad in another.
Why isn't there an agreed upon "general" gameplay
If we look at it from a logical standpoint, the phrase "general" gameplay refers to farming. Why is that? A standard two week event uses 4032 natural AP, during which we may clear a challenge quest once for the reward and occasionally do a few boss encounters during the story quests. The rest of the event will usually consist of farming trash mobs for event currency and the like. It wouldn't be far off the mark to say that the ratio of farming to bosses is 99:1.
However, not everyone considers farming to be general gameplay. If one considers farming to be easy, they could be of the opinion that challenge quests are all that matters. On the other hand, some people may think the opposite, as almost every challenge quest that is not NPC-locked can be soloed by a friend support. Yet others may be of a differing opinion, and consider farming to be as important as high difficulty content.
Even within a specific context such as challenge quests, there is no "general" gameplay that can be easily agreed upon, as people often utilize different strategies, all of which may be suitable for the situation at hand. For example, one might consider Jeanne to be the best soloer for the Saber Alter Memorial Quest due to her being able to completely trivialize Salter's damage output. She makes an otherwise challenging clear very safe, if a bit slow.
But there is also the option of soloing with Cu Alter, albeit with a lot more RNG, as evidenced by the savescumming in the following video.
tremprovin
Based on the two solos, would it be better to say that Jeanne is the better soloer for challenge quests? The answer is no - Cu Alter is currently the only servant, to my knowledge, that has soloed both the Dark Round Shadow CQ and King Hassan CQ.
元首の复仇
Would it be better to judge "general" gameplay viability based on the number of CQs that a specific servant has soloed? Or would it be more reasonable to only evaluate their performance in a subsection of CQs that people consider to be highly difficult (e.g. King Hassan CQ)? Maybe even the amount of RNG needed to successfully solo a CQ in the first place? These are decisions that cannot be easily made. What we can say is that there are options that are better suited for different situations depending on their circumstances and resources (such as the Saber Alter Jeanne solo requiring much less RNG and, by extension, real time investment than the Cu Alter solo).
Regardless, it is clear that even if specific factors and circumstances are taken into consideration, people can still have different perceptions on what constitutes "general" gameplay. Even when looking at the well-defined category of "soloing challenge quests", there are many different factors that can be taken into account when determining the value of a specific servant or strategy. If we evaluated a broader category (e.g. simply clearing challenge quests, not just through solos), we’d find it significantly more difficult to categorize servants based on gameplay. There are potentially thousands of ways through which challenge quests could be cleared, if we don’t limit ourselves based on an evaluation metric - ease of min-turn, solo capability, low rarity are all criteria that come to mind. Should all CQs be considered equal? If not, how should their difficulty be ranked?
If someone was to talk about "general" gameplay without giving details, they could potentially disagree with (or even mislead) other people due to differing priorities. If someone were to use only min-turns as the standard for their gameplay and give servants, strategies and setups based solely upon that criteria, they would obviously disagree with someone who values a different strategy. Hopefully I have made it clear how difficult it is to hold a discussion on even ground without establishing a clear standard.
Setting the goalposts
So how can one determine what kind of gameplay would be "good" for them? Sadly, there is no right answer. You have to set your own requirements. People can give you recommendations or provide sample setups but ultimately, it is on you to decide how you want to play the game.
A common requirement a player might have could be that they would like to clear content with the least possible amount of investment. To meet this standard, they could easily look up friend solos for whatever node they are running, since it wouldn’t cost them much, and would allow them to clear content quickly. You can see many story solos in this reddit thread. Furthermore, there are many community threads, discussion forums, and Discord servers in which people are willing to add newer players to their friend lists to assist them with their gameplay.
Even if your goal is to clear content without being carried by friend supports, there are many viable setups that would meet your requirements, such as low rarity clears. Regardless, the most important thing to do is to first define your requirements, standards, and specifications.
There's more than meets the eye
Something that one should keep in mind when evaluating different gameplay styles is the amount of work and effort that goes on behind the scenes for executing these setups.
This especially applies to runs where RNG is present - whether it be a solo, min-turn, low rarity only, etc.
For example, here is a really high RNG run:
赤瞳的斯卡哈
And here is a really low RNG run:
肉松-子
To someone who doesn't know how to notice when RNG is present, they might think the Merlin solo is more difficult than the Xiang Yu solo as it takes more video time. But the fact of the matter is that the Xiang Yu run required several resets and took a lot longer. In contrast, the second video is just long but without much RNG at all and can likely be replicated in 1-2 tries easily.
Before attempting to emulate similar strategies, one should always try to understand the RNG involved in them. One method in which this can be done is by evaluating how much RNG is required for being dealt good servant cards, or beneficial enemy actions. This, however, is a topic for a more comprehensive article.
While one might initially come to the conclusion that a lot of servants can solo the aforementioned CQ, it becomes evident that some runs are not what the majority of players should attempt to do as the purpose behind it is to showcase the limits to which a servant can be pushed and takes a tedious amount of time, unless the player is prepared to put in all that time in order to replicate it or do something similar.
People have different goals
After seeing the variation in methods, it's not hard to imagine how players can have different restrictions. While some players might not want to be carried by a friend’s servant, others simply wish to clear the story boss encounters in the cheapest way possible. In the latter case, those players would certainly go with a friend solo, as their restrictions are different from those of the former.
If you are a player who values taunt/buff stacking in their content clears, you might consider CEs such as poster girl and outrage to be good investments. To other people, who are satisfied with simply clearing the content in question as cheaply as possible (e.g. pressing AAA with Jeanne for 1.5 hours in the Saber Alter Memorial quest), the aforementioned taunt CEs would appear worthless.
To players who attempt to min-turn challenge quests and, to that end, rely on buff stacking, a Jeanne solo method would have no value due to the differing means in which they aim to achieve their goals, even within the same context (challenge quests).
One should always try to use the right tool for the job. This does not mean that if someone doesn't utilize a certain tool, they are incapable of using it at all. If someone owns a screwdriver and rarely uses it, that doesn't mean they don't know how to use a screwdriver - it just means they haven't had any reason to use one. In this case, it would be pointless to discuss the value of a screwdriver with them as they have no need for one.
Another thing to understand is that even if one is abiding by their own specific restrictions and requirements, it is possible (even likely) that they do not have the most optimal comp possible.
People have different benchmarks
Because people may have varying benchmarks for what is good, it is pointless to compare them unless there is a clearly defined task or context in which they are set. Some people might consider certain servants good because they are capable of min-turning/soloing various CQs while others might consider them bad because they are not capable of doing the same in specific CQs. It is entirely possible for someone to consider a servant to be "bad" while you think they’re "good" due to the difference in benchmarks, even within the same context.
One notable example of this is Kama, who has a charm on her NP. While some people might consider her to be bad for minturns due to the charm preventing the enemy unit from attacking, others might consider her to be one of the best as she is the only servant at the time of this writing to 5 turn clear full gauge Kama without command spells or refills (although the run did take ~100 hours).
Pinh丶
Conclusion
How good or bad a servant, strategy, or setup is depends on your personal goals. It is important to do research before deciding on the best gameplay style for yourself - don't let others decide how you should play the game in your stead.