It is sometimes unclear in which way a question should be handled. What can I do to check why excactly it doesn't meet our standards? What could I do to inform the user and improve it to be of higher quality?
This concise summary serves, under the form of a step-by-step guide, as a technical reference to current moderators (not only the diamonds, but also those who are voluntarily taking part of ensuring high quality among Super User), as well as people flagging and editing (even suggesting edits).
When is a question acceptable on our community?
While closing and flagging can become a habit, remember that we are trying to help the user. Handling questions doesn't only imply closing and flagging those that you don't like; but should also involve leaving a note how a question can be improved when possible, or be improved by yourself to make it match.
Getting your first question closed/deleted on this site is not pleasant, please consider first to improve it...
Is the question within the scope of Super User?
Our scope is defined in What kind of questions can I ask here?. The essence is that questions that match that paragraph are within the scope of our site – including conformance to the Terms of Service.
Unless a question clearly matches one of the not about points of our FAQ, there is no need to migrate. In fact, before migration you should first ensure that the question is of sufficient quality: Don't migrate crap.
Is the question constructive, real and useful?
What questions should I not ask here makes it easier to check this out:
The emphasized part covers both "not constructive" and "real", whereas "too localized" is based on how applicable the question is, because extreme conditions diminish the usefulness for everyone.
This doesn't necessarily mean that you have the close the question for being subjective, you could turn it into a great question instead; see "Turn recommendation questions into questions that ask for advice".
Furthermore, questions should be about solving an actual problem and not about solving a complete task. For example, if you're asking for a step-by-step how-to, you're probably doing it wrong.
Exactly. People need to do their homework because there is no point in doing their homework for them. Questions like these aren't real questions, simply because people don't learn how to solve problems on their own.
Is the question a duplicate?
Duplication Question Suggestion Boxes allow you to see a list of questions at the top, making it easier to scan the list of possible duplicates. Upon clicking the link, it opens the close box for you.
Sometimes, using a search engine could help as well. But remember that the main concern here is to prevent people from wasting time by not having them answer the question, it is not our goal to have all duplicates closed – but if you want to try, feel free to go ahead.
What can I do to improve the question?
Turning a badly written question into a question of high quality can help getting an answer for the question, as well as making our site look better. The point of a Q&A site that can be edited is that you don't have to scroll through boring forum pages. Instead, it should take you minimal time to read the question and the best answer, while learning the most from it in the meantime. This is often summarized as Quality > Quantity...
For example, turn recommendation questions into questions that ask for specific and expert advice instead.
Q&A is bad, let's go shopping outlines that while recommendation questions are not welcome, they could be improved by changing them to be asking for advice. While some questions are not salvageable, it is really worth it making an attempt to save questions who would result in really valuable advice.
Consider how "What do I need to consider when buying hardware to meet my needs?" is a constructive question and is instead extremely useful advice, not to forget that you can link to that advice the next time such not constructive question has been asked.
So, when you come across "Do I buy gaming mouse X or gaming mouse Y?", you know what to do...
Turning the title from a topic / task into a question about the problem.
Sometimes, the title of a question really doesn't explain its context.
Consider these three:
The first one really doesn't tell me anything, the second one has the habitual "add 'how to' or 'how do I' in front and we'll be alright" behavior which adds exactly nothing. Both of these two are focusing on the topic / task / goal, but none of these two are about the problem.
Exactly, in the third one we get rid of the topic / task / goal and turn it into the problem which clearly lets you see that the user simply can't find the setting.
Here is another effective one:
Only the last one clarifies the problem, and is also more SEO-friendly as a side-effect.
Make the body easier to read.
Removing excessive parts from the question as well as improving its look can greatly help:
Categorize the question into the right place.
Besides what's asked in the title, you can also help the reader by adding tags to the question. This allows the favorite tags system to function better, eventually getting the right persons to the question.
Make sure you add as many tags as possible, but avoid the use of meta tags, and avoid creating new tags unless you really have to.
Leave an edit note when an edit is not obvious.
While some edit actions in the questions are trivial, not every action would come over as clear to the poster. If their post is suddenly a lot shorter and you don't explain why it could be seen as an offensive action and/or confuse him. Not everyone reads the FAQ and/or Meta from start to end before asking their very first question.