You're on your way to the next level! Join the Kudos program to earn points and save your progress.
Level 1: Seed
25 / 150 points
1 badge earned
Challenges come and go, but your rewards stay with you. Do more to earn more!
What goes around comes around! Share the love by gifting kudos to your peers.
Keep earning points to reach the top of the leaderboard. It resets every quarter so you always have a chance!
Join now to unlock these features and more
The Atlassian Community can help you and your team get more value out of Atlassian products and practices.
Recently I've been more active in community and I started to notice that many new questions have already been answered in previous threads.
I think you have noticed the same (but tell me if that is not the case ;-) )
Since I'm a big fan of KCS and always looking for new ways to improve and optimize everything I do, I had to act on it.
What did I start doing?
I always try to include a direct answer to the question and a link to some Atlassian documentation (maybe an article from https://confluence.atlassian.com or a feature request or bug from https://jira.atlassian.com).
What is new is that when I see a recurring question, I now also add:
- a tag
- a link to the previous threads with the same tag. I found that the sorting by "views" is the one that shows the threads that are more relevant at the top.
By doing this, I think that I'm helping our community by sharing the most active discussions on the same topic (it is unlikely my reply will contain all the details previously shared) and also let new users join the conversation on the existing threads.
Here are two tags that I've created and used:
Add you comment below to let us know what you think about this, would this work for you and help to build an even better community?
And which tags would you like to use?
I've written essays on this in the past, as it's been annoying me since we were on OSQA.
People use us as a search engine, and it's frustrating. It's rude, but perhaps 25% of the questions are noise that could be answered with LMGTFY (I don't, but it really is tempting) and another 50% are exactly what @Caterina _or Kate_ Curti describes - subjects we've covered ad-infinitum, but people simply can't seem to find before they ask.
I'd rather have something where we can write a definitive answer to something (e.g. the Resolution problem) and then group up ALL the repeats underneath it, so that anyone asking the question again or searching and finding the same question goes to the same place. (I'd then hang discussions off it around possible new approaches etc)
Even a simple starting point would be a good stack of articles, and the ability for anyone to click "link to definitive article" as a quick answer. I'd want to monitor the use of that, as I suspect it would be rather high. Plus a "knowledge base suggestion system" would help - the title searches here aren't really much use in deflecting repeats.
Absolutely agree that I feel like i'm a search engine sometimes. Like Nic, I would guesstimate that about 20-30% of the time the user could have found their answer with a bit more searching in the Community and/or the Atlassian documentation.
I also agree that grouping would be nice but I'm unsure if it will be practical or not. It might be that the definitive answer is associated w/ an Article rather than a simple answer. One of the challenges for sure is that the precise answer often varies over time due to changes in the SW. This is most often true when providing step-by-step instructions which leads to follow up questions/replies, e.g. "this did not work for me", months down the road. This would mean the definitive answer needs to change over time or have different answers based upon version. While I feel grouping should be covered via tags this simply isn't the case. The other challenge w/ grouping is that it is going to be a lot of administration/housekeeping by the champions or someone to be moving/grouping questions.
@Caterina _or Kate_ Curti, for sure I often add links to other related questions and documentation. The idea is to teach users to fish. With that said, it takes time to do all that searching for them.
@Nic Brough -Adaptavist- and @Jack Brickey, thank you for your inputs (and also for all your contributions, I'm always impressed when I see how knowledgeable you are)!
I fully agree with you on so many aspects here!
My point of view is that, we should always have some users coming to the Community and that's actually good. I'm happy every time I see a new question, even if it was already asked in the past.
Even if everyone was using a search engine to find the content...
As much as I would like to also have a piece of knowledge somewhere that answers all the questions in a way that is easy to consume and will work for everyone, after some time in my role, I'm not sure that is possible. It is true that you could have a couple of variations to cover the various use cases, but when and how do you stop?
It is also true that we have a wide range of documentation and knowledge base articles already, are you maybe saying that there is a problem with them? Or maybe the articles are simply missing. I'm sure we would be interested in your feedback and would also do our best to make them even better. It would be great if you could share some examples for us to look at.
And yes, even if there is an article... I learned that it is always much more effective if I link to it but then, well... copy the content in my reply!
I always try to look for the big picture and the long term strategy (community managed articles? auto-linking of articles? what am I missing here?), but I always like to find something that works in the meantime (tagging).
I think the biggest hurdle is finding the most relevant information when searching through knowledge base articles, as Jack and Nic discussed previously.
It's one thing to think about community answers from the Atlassian perspective, and quite another when considering the volunteer hours being spent to connect all the dots.
If the champions at very least could increase their efficiency in responding and spend less time looking for answers in previous posts, it would be a huge win.
@Caterina _or Kate_ Curti, certainly agree with your points here. This is a great topic, centered around how to make the community a better place. I wonder if it would be worthwhile having a sit down discussion during the summit between Atlassians and Champions to brainstorm further. Just a thought.
Interesting idea about , I agree that a brainstorming can help in defining a vision and some strategies around this topic.
Side question (but maybe also related): are you all going to be at Summit?
And no, I did not check if I your name is there on Monique's post.
@Meg Holbrook your contribution to the community is very much appreciated and I do myself spent a lot of my free time on community, but yes I understand that working for Atlassian and doing support all day puts me in a privileged position.
At this stage, sometimes I (and my Chrome history) just "know" which pages are the most common one to share.
Just a thought: maybe I could share my favourite/most used resources!
I think when I started I did a lot of research myself, and yes... I wish there was some automatic finding and linking ;-)
I think the most difficult part, as a person answering the repeat questions, is maintaining a link database of those 'root' FAQs. So I spend a lot of time digging for links back to the previously asked questions, which is also not a great use of my time.
Was discussing this with @Jack Brickey the other day, and I suggested building out a text expander for frequent links - I use Phrase Express. This still puts a lot of work on the person answering the questions.
I love the idea of creating this big bad web of link-ability, but wish that it wasn't so much work and backtracking.
Interesting usage of Phrase Express, but I totally agree with you that a common solution would be much better for everybody (including the community users not just the champions).
Maybe we can all come up with some options and vote on the preferred approach?
And in the Atlassian style, everybody can follow... or not!
I am with Nic on the KB-like approach. Although I'm quite certain that it might take quite some effort to build this into the platform as it is, maybe it might even help surface most relevant answers to similar questions after a while too. Automagically! By collecting feedback in the "did this article solve your problem" style.
That approach (though maybe theoretical) might go beyond the accepted answers and votes we have now, as I would guess they are mainly used for 'active' / unanswered questions. Users that find help through a KB style article and confirm they have been helped by it, may push certain answers up the ranking ...
Hi @Walter Buggenhout _ACA IT_,
I like the "Automagically" approach, but yeah... I don't know myself if that's feasible or not.
I'm catching up with all the super valuable inputs here, let me know if there are still some items that I did not cover in my previous/other replies.
Agree with Nic, there should be something like a good knowledgebase. There is a Jira knowledgebase with articles, but the list is long and unclear (https://confluence.atlassian.com/jirakb/all-jira-knowledge-base-articles-901485449.html).
A similar list is there for Confluence (https://confluence.atlassian.com/confkb/confluence-faq-154001.html) . This is the first hit, if you search for Confluence FAQ with google, but it is old, the last time it was updated in February 2016. If you look further, you'll find an endless list of knowledgebase articles (https://confluence.atlassian.com/confkb/all-confluence-knowledge-base-articles-229181713.html) updated one year ago. I can understand that a user prefers asking a short question in our Community before looking through hundreds of knowledgebase articles... It doesn't cost anything, usually, you get a fast answer from a friendly person, everything's fine. That's how it should be, but on the other hand, this costs a lot of time and a lot of us are answering these question in their spare time.
There was a similar discussion here already: https://community.atlassian.com/t5/Mod-Chat-discussions/Should-we-use-Articles-for-How-Tos-for-commonly-asked-que/td-p/608535 but I think, nothing happened after (or maybe I missed something).
If you ask me about connecting answers with tags, I would say, it depends... I'm not a fan of creating a lot of new tags just to combine answers. In my opinion, tags should not divide anything into small pieces. I would use them for bigger topics only.
Regarding tags, I would always ask myself, if a user would enter this tag into the tag-search-field. If the tag is too detailed, I don't think, that it makes sense. Combining a tag and a search phrase (looking for the phrase within tagged topics) is the best way in my opinion.
We should have something like FAQ topics and articles and a good structure for them. And someone who takes care of them, updates them (what if a new Jira version behaves in another way, should the article be removed, or should be there a second one - soon we'll get the same problems as the odlder FAQs before...)
I'm with Nic and Thomas. I like the idea of a FAQ. Part of the major problem is that even though a user can search and hit upon a number of community answers, and can even explore them, there is a fundamental problem that most of them are not "solved". And if they are solved, it's a long trail to figure out what did actually solve it. When we are limited by just comments, it's tricky.
Very good discussion. Let me also jump to it :)
The problem is that you want to fix all problem in one shot. That is impossible. I am working on a support for many years and I can say that problems on community are more global that we think...
People will always have problems searching for answers. There are group of people (like most of us) that can quickly and easily find answer and usually do not ask many questions (only when we are stuck or find something not clear). On the opposite side we have people that like to ask for help even they are able to find the answer by themselves even there are many good articles, FAQ, KB, and other things that help..
If we group questions that are already asked then again.. some people will explain everything in very precise small details, include even a screenshot and say what they tried already and other people would simply ask "having error after upgrade please help" - and be a God and try to know what they are thinking of.. you need to ask ton of questions starting with "is this server or cloud" or "did you check logs" and maybe after some time (after exchanging few posts) when user is responding quickly (and on all of your questions) we can try finally help him.
In which group people are that depends.. It is genetic thing, how we were growing up (in which environment) and where we live (what culture we have). Even we would try hard .. we would not change that. There were and would be questions that would be asked again.. and again.. and again.
So in general we would try to just simply reduce the number of duplicated content. But .. we need to know that content could be very quickly be outdated due to many releases and solutions (server/cloud/data center). "How to archive a project in JIRA?" asked 10 years ago might have completely different answer than now. Knowledge base articles and documentation is many times also outdated and hard to find the correct solution. Just to many things going on currently.. And that is what we cannot easily fix.
I think on community we can start with 3 things to optimize "traffic":
* A better visual way of filtering existing content - not only tagging which mostly is wrong and need to be edited later
* Required step to read FAQ with all important information for all new users BEFORE posting their first post (how to accept an answer, what is a difference between question and discussion.. why it is important to describe the problem in details .. etc.) - we are sure that people are educated before posting and it might also technically reduce spam (additional step for spam machines)
* Going through the "intelligent and interactive wizard" instead of choosing a type of a post from 3 icons. - that would reduce number of mistakes and let people think before they post something and give us more information. How does it work? It is super simple ... User is answering few simple questions instead filling a form. Example questions:
1 Question.. Do you have a problem with an application, want to ask for feedback or share knowledge with someone? (let assume it is a problem)
2 Question.. Is it related to Cloud, Server, both or you are not sure? (user choose Server)
3 Question.. Did you checked already existing questions and/or searched for solution in Atlassian documentation? If yes, maybe you can paste URLs to some of them that did not work.. (Yes, I tried this and that.. )
4 Question.. What is your experience with using Atlassian products? ( I am new to Atlassian products)
Overall previous user answer generates another.. What we get after that?
After filling that kind of wizard and posting we are sure that people will create questions when they really have questions, discussions when they as for feedback (no need to convert them...), do not need to fill tags (those would be filled by them automatically based on their answers). We will know did someone tried to find a solution and how much experience he is having already. Thanks to that our answers might be more precised without additional noise and as a bonus we might have links to articles that people already tried (no need to point them again to the same posts or articles). A WIN/WIN for everyone only by improving the way posts are created.
Am I right? :)