What's the recommended approach when you ask a question and you don't actually get valid answers?

It'd normally leave the question as unanswered but, of course, that leads to getting a low acceptance rate, which seems unfair. Asking in forums is normally my last resort so I tend to ask difficult questions and I understand some of them do not actually have a positive answer (for instance, when you ask whether there's a tool to do X and the only answer is "there isn't, you'd have to write it yourself").

  • 7
    "when you ask whether there's a tool to do X and the only answer is 'there isn't, you'd have to write it yourself." -- That's an answer. – user141160 Feb 28 '10 at 13:27
  • 5
    @DN: When someone answers 'There is no tool', I see 'I don't know of any tool'. How can someone be sure? – LukLed Feb 28 '10 at 17:26
  • This has been asked many times before. – Ether Feb 28 '10 at 17:56
  • 1
    @LukLed That's part of the beauty of the web. If someone comes along and corrects them, or if the situation changes, the answer can be changed. And that may never happen, but then, that's a problem with incorrect answers, too: meta.stackexchange.com/questions/7572/… – user141160 Feb 28 '10 at 17:58
  • @Ether: I don't want to be rude but the go-ask-somewhere-else type of answers do not really contribute to any forum. – Álvaro González Mar 1 '10 at 10:49
  • @LukLed: In some cases, there indeed is no tool: "what is the program to check if my code will finish?" is a restatement of the Halting Problem, and is undecidable in general; "what is the original string of this hash?" is another example. Contrary to the popular illusion, not everything can be solved with 200% more enthusiasm. – Piskvor left the building Mar 3 '11 at 21:18

The first thing to check is that you've asked the question with all the appropriate information. Do any of the answers suggest you've given the wrong impression? Is there any new information you can add?

I wouldn't personally worry about a low acceptance rate - not if anyone looking at your questions will see that you're an active participant who cares and is unlikely to abandon a question.

  • 1
    I normally update the question as soon as I find answers to a different problem. But you're probably right: I should really care, there may be badges and all but it's not a contest. – Álvaro González Feb 28 '10 at 12:25
  • 1
    ... not a contest?? Oh, damn... – Mark Henderson Mar 1 '10 at 4:14
  • Of course, I actually meant "shot NOT really care". – Álvaro González Mar 1 '10 at 10:50

The acceptance rate is overrated!

Ignore acceptance rates of other users, ignore your own acceptance rate, ignore users who do not ignore your acceptance rate. They cannot provide a good answer anyway.

If you did not get a good answer, do not accept one. The green check mark indicating the answer which helped you is more important for future readers than this stupid accept rate.

  • 6
    +1 you're totally right. – Pekka Feb 28 '10 at 13:11
  • 1
    I wish you were true, but I was berated by the self-appointed SO police more than once to improve my acceptance rate before asking new questions. See e.g. stackoverflow.com/questions/2026546/override-get-but-not-set – Graviton Feb 28 '10 at 15:45
  • 2
    @Ngu: Ignore them! Also flag their comments. – Ladybug Killer Feb 28 '10 at 15:55
  • To ignore them and/or flag their comments is anti-social behavior. SO/etc are respective communities, and part of participating in them means respecting community practices. Continuing dismissive behavior will only guarantee it will happen again. – OMG Ponies Feb 28 '10 at 19:37
  • 2
    @OMG: And forcing someone to accept non-answers is not anti-social? And it is a respecting practice? OMG Ponies, you are a weird person! – Ladybug Killer Feb 28 '10 at 20:46
  • @John Smithers: I see you like to exaggerate and misconstrue, so where's the gun to the head, forcing us to select answers? For the people who feel persecuted about their acceptance rate, it sounds like you already know the answer - which only makes sense if you are asking disingenuously. Your definition does not match what the community believes - which is your prerogative. Continually flaunting that is effectively anti-social, and self damaging. I'm not weird, but you apparently feel the need to talk about something you know nothing about. – OMG Ponies Mar 1 '10 at 2:47
  • @OMG: Oh, I know nothing about it! Very interesting! And the other 7 guys upvoting me, have no idea either. My little pony, you are not only weird, you are a troll! Or you just like to offend people. Well, I don't care, just go on. – Ladybug Killer Mar 1 '10 at 7:44
  • The reputation system is a nice incentive to contribute but it can sometimes make you too competitive. Thanks your sharing your opinion. – Álvaro González Mar 1 '10 at 10:51
  • The problem I have with low accept rates is that it suggests to me that the poster is going to walk away and pay no attention, which gives me a lot less personal incentive to help. On SO, I've got way more than all the rep I need, so I don't care about my answer being accepted, but I'd like to to be read and considered. – David Thornley Mar 1 '10 at 15:20
  • 4
    @David: well, there are certainly users who give this impression... even before the accept rate was displayed, it was easy enough to pick up that vibe: users would refuse to clarify their vague questions, ignore or argue with comments asking for additional information, etc. IMHO, these are usually lousy questions anyway - we'd all be better off if they were just down-voted and ignored. Still, if you think you have a great answer then you should post it: even if the OP never acknowledges it, you'll still have put it out there for other readers. – Shog9 Mar 1 '10 at 21:05
  • @Shog9 +1 for the last sentence. :) – Alenanno Jan 21 '12 at 13:11
  • "ignore users who do not ignore your acceptance rate. They cannot provide a good answer anyway." well, that's like, your opinion... I DO pay attention to low accept rates, and if it's very low it discourages me from putting the effort and write an answer. for me, it says that the PO either didn't bother to check the answer or to validate its correctness. even if none of the answers helped the PO - I would expect her/him to come back later on when they've found the answer and write it down and accept it - for the community. if they didn't bother to do that as well - I won't bother to help them. – Nir Alfasi Jun 29 '12 at 18:12
  • @alfasin: And who cares if you bother to help or not? Right, no one. – Ladybug Killer Jun 29 '12 at 19:20
  • @LadybugKiller if you need help, and no-one else replies, and I CAN help, then I guess you WILL care. but that's only like, MY opinion :) – Nir Alfasi Jun 29 '12 at 20:35

Another thing that is good practice for questions for which you have yet to receive a good answer is responding to the answers given.

I too rarely ask for help unless it is truly something that I just cannot figure out. This usually means questions that are of the WTF? and who-even-thinks-of-that type. As a result (at least on SU as opposed to SO), I have a relatively low accept-rate.

However, I make sure to respond to every answer to explain why it does not work for me. That way, I make it clear that the question is not abandoned (I do have an interest in it being resolved), and by responding to the non-answer replies, I end up providing more information that would be helpful to others who would offer an answer (without having to constantly update the question with possibly superfluous details that may only apply to specific answers).


Have you ever tried Bounty


What if I don't get a good answer?

In order to get good answers, you have to put some effort into the question. Edit your question to provide status and progress updates. Document your own continued efforts to answer your question. This will naturally bump your question and get more people interested in it.

If, after 2 days, you still don't have an answer you like, you can offer a bounty. Slice off a bit of your own hard-earned reputation -- anywhere from 50 to 500 -- and attach it to the question as a bounty. We'll even throw in 50 reputation to sweeten the deal. The bountied question will appear with a special icon in all question lists, and it will also be visible on the home page Featured tab.

Once initiated, the bounty period lasts 7 days. If you mark an accepted answer, your bounty is awarded to the answerer (do note that accepted bounty answers are permanent and cannot be changed). If you do not accept an answer in 7 days, here is what happens at the end of the bounty period:

The highest voted answer created after the bounty started with at least 2 upvotes will be automatically accepted. Half the bounty will be awarded to the owner of that answer. If there is no answer meeting the above criteria, and you as the question owner have not opted to accept an answer, you forfeit your right to ever accept an answer to your question. In any case, you will always give up the amount of reputation specified in the bounty, so if you start a bounty, be sure to follow up and accept the best answer!

Of course, bounty awards, like all accepted answers, are immune to the daily reputation cap and community wiki mode.


When you're sure you've done everything in your power to present a good question, and it turns out to be so difficult and/or specialized that it simply doesn't get a qualified answer, you are perfectly right in leaving it open for future generations (i.e. not accepting an answer).

On the other hand, when asking questions, you will be judged somewhat by your accept rate, especially if it is below 50%. This social dynamic is encouragement to close a question (e.g. by an "I guess this can't be done" answer of your own) where it might have been better for the system to keep it open, and see whether anything comes up later.

That is a bit of a problem in the system IMO. Maybe it should be possible to "give questions away" to the community user if they don't find an answer, but might some day.

  • You are right, that's probably something that could be fine tuned in the future. – Álvaro González Mar 1 '10 at 10:52
  • Yep, I've definitely felt the pressure too to accept some invalid answers to keep my accept rate high enough. – Björn Lindqvist Nov 9 '12 at 12:02

Not the answer you're looking for? Browse other questions tagged .