Basic things to consider when writing an analysis:
1) Quite a few writers write across multiple metas but don't actually play all of them. While you can use references like the tier's viability rankings and roles compendium, they only help so much and don't benefit as much as being hands on with the tier. Often this will lead to writers creating very generic analyses that are clearly written from an outsider's perspective, leading to QCers needing to edit a large majority of it when in the Quality Control stage. If you want to write about a Pokemon, make sure you've used it in the meta you're writing for. Using a Pokemon three times will not cut it here!
2) Do you know what sets (or at least have an idea of what sets) are the best or most used for said Pokemon? If the answer is no, then you should probably either try to write for a Pokemon that you're more confident with or even ask a member of the Quality Control team about which set(s) to use; this should generally be done for verification though, not if you're completely clueless. Not asking may result in either a rejection due to not understanding the Pokemon or a complete rewrite to accommodate for a new set. If you're feeling especially uncertain, then you can post what set(s) you think the Pokemon can run in your reservation post.
3) Do you have prior experience with writing? If you don't, I'd highly suggest reading other analyses that are either in their final stages of QC or in GP, as they'll give you an idea of the level of quality most forums require for their writers. Your quality doesn't have to be on par with them, as they will have already been checked by other people, but it does have to be around that level of quality. You shouldn't be copying the format of analyses in QC 3/3 or in GP, as they'll be in paragraphs (unless it's an Ubers Pokemon), and you'll be asked to change it back to bullet points. You should be looking at how they convey their points, not just the points themselves; a good analysis will be able to explain them concisely.
4) You have
to ensure that you follow both the GP standards
and the standard writing format
. The GPers are quite lax towards writers, but a basic understanding of English is required when writing. All analyses will need to follow a set template to keep consistency.
5) Different tiers may do things differently. If you're unsure, then you should ask a member of the QC team about what standards they adhere to. For example, most tiers stop using bullet points after the second QC check, but Ubers keeps them throughout. Doubles OU also only has two QC checks, while all other core metas have three.
6) Please be considerate about how long you take to write an analysis. If you can't do it in a reasonable amount of time, then you probably should wait until you have more time instead of picking up then later dropping an analysis. If it's going to take a decent amount of time, PM a QC member so they can decide what to do. You don't get penalized for dropping however, so you don't need to worry about that.
7) If you're legitimately struggling, it's perfectly fine to go to a QC member. Actually, I actively encourage this, as it means that the initial analysis will already have some decent input.
8) As a general rule of thumb, I wouldn't refer to previous versions of the analysis you're writing, even if it's a revamp. This will often lead to either lazy writing, old information being transferred across, or just straight plagiarism. We take plagiarism VERY seriously.
9) You HAVE to follow this format
. Every USM analysis is formatted in this way, unless they're mini ones.
Basic things to consider when QCing an analysis:
1) Don't be lazy! QCers are the people who ensure, well, the quality of on-site analyses. Your job isn't to be nice or to churn QC checks out as fast as possible—it's to ensure that every analysis reaches a high level of quality. You should be taking your time as you go through them, questioning sets, details in each section, and the examples given. If you check something and only change one or two things, then the next QCer changes half the analysis, you haven't been thorough enough in your check.
2) You should be checking for overall comprehensiveness, the relevance of the examples and information to the meta it's being written in, the conciseness of the analysis, etc. You're helping to convey how the Pokemon performs and what it does.
3) Unless something is obvious, like extremely bad examples or just factually incorrect information, subjective changes should be discussed with the rest of your team. Basically every QC team will have a side chat in their Discord with which to debate on more subjective things like sets and Checks and Counters. Don't make large changes without discussion, like making a massive change to the actual set or revamping a large amount of something like the Usage Tips or Team Options. Small changes to any section will often not require team input, unless you want other's opinions.
4) If something is bad, you have every right to reject it. You’re trying to get the best analyses up for your metagames. If you don’t have enough writers, you may have to be a bit more lenient with your rejections, however. Some metas require the leader's approval for rejection, others require three rejections, and a few only require one; always check with your section leader beforehand and discuss it in your QC chat.
5) 99.5% of the time, THERE'S SOMETHING TO CHANGE. If you're doing “good job looks good n_n qc 3/3” then you're probably not looking thoroughly enough. Things that aren't changed in QC 3/3 are often left for uploaders to change, which is just passing off the responsibility and in turn makes your check pointless.
6) You're encouraged to help GP by nudging the writer to make small grammar- and spelling-related edits in your QC checks. Make sure to do it as comments, though, e.g. "remember to capitalize Pokemon names" or "I see this error a lot, please fix that so GP doesn't have to"; leave the fully fleshed-out markups to GPers. For a reference on what changes you're allowed and encouraged to make to help with the GP process, refer to this guide
7) There's very little point in reorganizing a list of examples. The order doesn't matter, the actual examples do. Don't needlessly make writers change theses around.
8) You HAVE to ensure the writers follow the analysis format. Do not let it move past your QC stage until they have, or it'll look poor on your end.
Basic things to consider when AM checking an analysis:
1) When starting out with amateur checks (AM checks), you're mainly looking for factually incorrect information inside sections. You shouldn't be changing sets or making clearly subjective changes, as these are for the official checkers to do.
2) Treat AM checking as a way of proving to others that you understand what you're talking about and that you can relay this information in a coherent manner. It may seem as if these checks go unnoticed, but QC does check these when QCing and will use them to consider you for a potential future spot.
3) Don't GP check if you're AM checking, same as with the QC section.