Question and answer content has been around on the internet a lot longer than the web. From the early days of usenet (and before that, on prehistoric technologies that came before my time) people have been asking experts for answers. You can't fail to have noticed, however, that it has resulted in some of the lowest quality content on the web. For far too many questions, the ranking answer is Yahoo! Answers (or similar) that looks as though it was written by someone who found writing YouTube comments too intellectually challenging:
Incidentally, note that there is a Yahoo! Answers API which can be a great source of data for keyword research and content inspiration.
There is a clear need and opportunity for this kind of content. Many users search in question format:
unfortunately not quite enough to support the venerable Ask Jeeves that encouraged this behaviour
There are clearly a lot of commercially-oriented queries in there. You only have to look at many of the kinds of questions people ask on Twitter to see this:
You can see the attraction to search engines of indexing Q&A content. While they have made leaps forward in natural language processing, they are still dumb text query engines at heart and having both the question and the answer in plain text on the page clearly supports them in providing efficient answers to many natural language queries.
My favourite example of the right way of doing things is Stackoverflow. If you have ever tried to do anything related to programming, you will have hit annoying issues very early on. At Distilled we have labelled most programming as "copy, paste, swear, fix typo".
If you're anything like me, RTFM might as well stand for JFGI these days and you will Just Google It(TM) straight away. As soon as you do this (certainly in recent months), odds are you are going to land on Stackoverflow. The answer you find there is likely to be helpful, authoritative and on you go. But how did it get that way? Stackoverflow is clearly good for search engines, but it got that way by being great for users. If anything, the experience of asking a question and getting it answered is even more impressive than just seeing the repository of brilliant answers that already exist.
Even really annoying basic questions get quick, patient answers
Lessons from Stackoverflow
So what did they do so right?
Avoid many versions of the same question
The issue of multiple almost-identical threads is so prevalent on most Q&A sites that Google has evolved a UX pattern specifically for this:
Stackoverflow gets around this with a many-pronged attack:
- Start with a culture of curation - when users know they are creating a reference, they behave differently to threaded forum discussions
- Use great power wisely - with curation as a justification, Stackoverflow editors can edit, move, lock or close replies to questions to encourage desirable behaviours
- Scale curation and editorial - I'll write more about the gamification, but the increasing site editing ability that comes with the earning of karma enables a small core team to enlist the help of a large passionate team of editors
- UX hints point people in the right direction - as you start asking a question, similar previous questions appear "google-instant" style encouraging browsing before submitting
Get good answers, fast
From what I've read the biggest KPIs for the Stackoverflow team are proportion of questions with accepted answers and the speed of answer. Indeed these are high on the list of metrics to consider before opening a bring q&a site out of beta. I love the data-driven attitude and transparency they show - this is a post about bringing the home improvement forum out of beta:
It's interesting to think about how they have designed a site and a community to achieve great results on this front. In my opinion, a large part of it stems from having nailed the incentives - in particular:
Gamification
It's no secret that people love points, awards and power. The game mechanics built into stackoverflow bring all of these things:
- Points - with evidence ranging from gathering twitter followers to foursquare points, we see that give people a number and they will work to improve it even without an obvious reward. The setup of stackoverflow rewards both quick answers (high # points / time) in general and correct / insightful answers to hard questions (that get voted up). This nicely aligns with the goal of "good answers, fast"
- Awards - you get badges (see below) as you complete tasks around the site. These are nice for their own sake - especially if you get access to the rarer ones - but they also bring you:
- Power - the points and badges you acquire unlock special powers ranging from the ability to rate other people's answers all the way up to full admin rights to the site with the power to delete, move and edit pretty much anything. This funnel of power aligns user incentives very effectively
I've only got a handful of stackoverflow badges so far. Maybe I'm immune to their wily ways?
If you can, build from a passionate community
In the case of Stackoverflow, they built from a bunch of overlapping groups of passionate users (as I understand it, based largely on the personal clout of Joel Spolsky and Jeff Atwood). In the case of the SEOmoz Q&A forum, it's obviously benefiting greatly from the community on the blog etc.
If you want to read more about the intersection of Stackoverflow and SEO, they have had a couple of posts about it: one, two and a HN thread.
Great moderation
Stackoverflow solves the moderation problem in one way. Quora is tackling it in a different but fascinating way:
- A quiz for new users (that wasn't in place when I joined!) to ensure that everyone registering for the site at least understands the way that the business wants the site to end up structured
- A whole bunch of product features designed to nudge people towards desired behaviours
- An encouragement to "think in the Quora way" throughout - see the FAQ for a number of examples
You should want these benefits on your site
We have been doing a lot of thinking about the possibilities in this space - and what happens when you get it right. When Tom was at SEOmoz (before heading to NYC), one of the things he pushed hard was to add many of these features to the Q&A forum:
Ask an Owner
I have also been working with one of our UK clients, Reevoo, on a new feature they call "Ask an Owner" that enables retailers and manufacturers to allow potential buyers to ask questions of those who already own a product. By allowing those retailers and manufacturers to expose that content to search engines, we hope to access some relatively untapped areas. Reevoo already provide review functionality for many top retailers and brands and they are seeing some phenomenal stats on the new Ask an Owner service in terms of questions being answered, % of good answers etc.
It is amazing how many relatively sensible questions still have no content indexed e.g. "Does the HP probook 4320s support skype video calling?" you can work out the answer from many of the resulting pages if you know what to look for, but wouldn't it be great if there was a page with that title and body content including something like: "Yes. There is an in-built webcam that works very well in reasonable lighting conditions. As with many laptops, it's built into the top of the screen which makes for natural conversations as you automatically look roughly at the camera as you speak. The built in microphone is also good enough in quiet conditions. For more serious use, you should consider a stand-alone microphone." That's the kind of content that should be generated by "ask an owner" style functionality.
If you happen to want to know more about ask an owner and our general views on UGC in retail, you can check out the whitepaper I wrote on the subject (registration required).
Q&A and the investment community
The investment community got all excited about Q&A sites last year and pumped loads of money into Quora and the like. The reasons they got excited are similar to the reasons I believe there is untapped SEO potential here, but I also think there is significant value for many smaller businesses even in things that wouldn't get investors hot under the collar.
Today is a US holiday and I'm also not in the UK office. As a result I may be slow to jump back into the comments below, but don't let that stop you sharing your thoughts. I'll join in when I can!
Hey everyone, this is Casey from the SEOmoz Marketing Team, I thought I would take this time and add some stats from the SEOmoz Q&A Forum. Below are the stats from March 28 to July 3, 2011:
- Question
- 4,256 Total public questions asked by 1,941 members.
- 3591 questions are looking for a specific answer.
- 665 questions are open discussions.
- 18 Hours - the average until first answer was received.
- 2,044 answers marked as helpful.
- 4,256 Total public questions asked by 1,941 members.
- Traffic
- 1,011,500 pageviews.
- 5,254 keywords to 2,841 pages.
- 14,656 unique visitors from search traffic.
- Members
- 2,422 active members in Q&A.
- 52,722 MozPoints earned.
- 100 Users earned more than 100 MozPoints
- 2,422 active members in Q&A.
Thanks to all our members who have spent many countless hours in the Q&A Forum answering questions for other members! The amount of knowledge that gets shared in there is amazing!
Thank you for sharing, Will.
For those who want to look into research on Q&A development, I can recommend a couple of academic papers that help design q&a sites:
Viewing this from the other pespective of "Q+A sites rank well", we've found that answering questions on Q+A sites can have two benefits:
1) If the client has a generic brand name (e.g. quality insurance) then answering questions that contain their brand name as a generic term, such as "what is good quality insurance?", helps you have some control over what appears on page one for the brand - as these pages tend to rise high quickly.
2) If the client site is not very strong, and you want to rank for question and answer topics found in keyword research, you can try and answer these questions on Q+A sites. Use a branded avatar as the company and if people like your answer they may then search for the brand. If you can add a signature link then even better. The better the answer you give, the better your brand appears so don't spam it!
Yahoo answers is a joke for the most part. I would imagine that building an effective and truly useful Q&A service is extremely challenging!
We've certainly been finding that. We're a PR6 site (not that it matters of course!) providing resources for website owners as well as directory listings for New Zealand websites (NZS.com). Not too long ago we added a Q&A service designed to focus on a local audience. While some questions and answers have been great, some others have often been questions either directed at our staff regarding a user's own listing, or they have been from a user that believed we were one of the companies featured on our site.
It's much harder when you haven't created a stand alone purpose-built question and answer website, but rather you have added the service onto your own well-established site.
There is a good thread on stackoverflow, ironically, about stackoverflow-like clones in various language flavors.
https://meta.stackoverflow.com/questions/2267/stack-overflow-clones
I have played around with a bunch of them and my personal favorite for PHP is Question2Answer. It has two shortcomings 1) there is only one person developing it, and 2) out of the top listed ones above it does not support OpenIDs out of the box.
Would you suggest putting time in on 3rd party services like Quora, or, for medium-sized consumer companies, running Q&A right on their sites?
Also, have you seen any B2Bs make good use of it? Large B2Bs could easily do Q&A and use it as a sales tool, and SEO tool, and a general branding tool. But most are too afraid of it to try.
I've gotten a bunch of value from contributing to Quora personally. When Lisa Barone wrote about it on Outspoken, I commented there on the very cool things that can happen as a result. I'd suggest that anyone with expertise in their field at least give it a try on a relevant Q+A site or industry/niche forum. BoardReader's a pretty good tool for finding the latter.
I have personally used this for the CPG brand i manage and this has proved to be greatly successful. In 2 months, referrals from QA sites increased 120% and through answering questions and consistently achieving 'best answer' status, it has proven to drive links.
Also has helped in organic SEO rankings. Conversions from the extra traffic has been limited but in the research, i have seen it produces eyeballs that stay for a long time (~3 minutes)
I think Quora provides its own non-SEO-related benefits, but - unless you're a big deal, like Rand - there isn't much help in the SEO department beyond maybe catching the interest of a peer or two.
If your company is big enough to be a topic, though, then answers from someone at the company can really be worth the time.
strong agreement on both points.
This is how I feel also. I use Yahoo Answers and Quora to help increase my known Expertise, but having a recognized Brand seems to instantly make a difference (at least on Quora, since it's a bit more social/modernized compared to YA). I've tried a variety of Q&A sites, those two are my favorite so far.
On a side note, onsite Q&A boards are excellent traffic source, even if you aren't a big brand. It's surprising how many questions people ask Google these days.
I run a very successful Q&A site. It gets a lot of long tail traffic and attracts ad clicks nicely.
On my site though, I am the only one answering questions. I could see that it could explode as far as traffic is concerned if I opened the questions for others to answer, but I want to produce quality stuff and not end up like yahoo answers!
Thanks for a good article.
Are youthe only one because you decided to be the only one answering? Or it is because the site is failing in incentivizing the answers by the others? Because if it is this second reason, IMO the site is failing its purpose, which to create a community, and with that all its potentiality is just that: potentiality.
lol.... Maybe it is all about quality control?
For some subjects you don't want a lot of ignorant information up there.
If an idiot poses as an SEO and somebody uses his bad advice the worst that can happen is the person using the advice loses a pile of money... but if a forum is for brain surgeons giving advice to other brain surgeons you don't want any posers in there.
Just saw this comment now! It is actually a choice to just have me answer. I am a professional in a particular field and I consistently produce good quality content for my site by answering questions. I considered opening up questions for the public to chime in on...but realized that the quality of my site would suffer.
Great post. Q&A is the majority of our site by any measure (PV, landing UU, content) and we have over 193,000 answers posted by Dermatologists, Plastic Surgeons and Cosmetic Dentists with thousands more added each week.
I met with Joel Spolsky earlier this year (we are both Yale CS grads), and he commented that he was surprised by our traction in medical because sites with a very low reciprocity rate (% of users who both ask and answer questions) never worked out as a StackExchange site. Our recipirocity rate is 0% because we prevent doctors from asking questions (although editors can post it on their behalf) and prevent non board certified doctors from answering questions. We launched without those restrictions but had to put them in to avoid self-promotional questions and answers.
Adding Q&A to an existing community makes sense -- the key will be figuring out the incentive for people who answer questions. The "just add points" form of gamification is probably not enough incentive so additional incentives are often necessary. For our doctors, national exposure and attracting potential clients are big incentives.
It's easy to compare Yahoo asnwers with Stack Overflow and get Stack to rise up as a winner. Yahoo is a broad Q/A site that tries to answer any question from philosophy to parody, so there are millions of stupid questions and due to the number it is hard to moderate. Stack overflow on the other hand is a very specific community targeting technology, so people going there won't ask stupid questions (they mostly don't), they will ask specific questions related to a programing issues, which makes it a valuable Q/A community. Even Quora is having a bit of issues with that, but since it's new the moderation works well for now.
I guess Q/A sites also have to be targeted, just like we narrow down our market and keywords to get better exposure and sometimes go for a certain specific niche, the same rules apply here as well. The idea is simple, find the appropriate Q/A site to engage long term, and result will definitelly come from it, use Yahoo and 5 more other Q/A sites and most likely you will fail to get what you want...
Really great point. I hate the fact that Yahoo Answers comes up because a lot of the answers are complete BS. You can really build up some great content around questions. I'd like to build a Q&A niche site but I think it would be a lot of work.
One of my first clients was toptipsforgirls.com The owner spun the user generate online tips out into a series of succesful books https://www.amazon.co.uk/Top-Tips-Girls-Kate-Reardon/dp/075534314X
If you do real quality QA then there are other avenues that open up
https://www.commandlinefu.com/commands/browse Not Q&A in the strictest sense, but tips for developers - based off the same mechanics
Great post, i'm developing a Q&A site right now and it's a big challenge, i'm loving it :)
Interesting article. I did a study of the most searched for questions by subject here https://searchinsights.wordpress.com/2011/09/01/the-most-asked-questions/ and the likelehood of someone phrasing a search as a question according to what content they're after
I think the interesting thing here is that not only does this have SEO benefits, it's also a great way for companies to find out how users are using their products and what they want to know about them at the pre-sale stage. Sometimes it's hard to gauge it properly without feedback, particularly in cases where a user is using the service or product on a regular basis and sees room for improvement. - Jenni
We recently started a local 'New Zealand based' question and answer service on NZS.com (I'm the editor):https://www.nzs.com/answers/It seems to be working fairly well. We have some great popular questions that rank fairly well for those terms (albeint mostly on a 'pages from NZ' level, such as 'how do I convert web traffic).
We noindex the questions that don't yet have answers as per Google's recommendations.
Any feedback would be super-appreciated!
Will, here's a link to a wildly successful Q&A site that answers questions for Oracle programmers. This guy is doing a lot of things right.
https://asktom.oracle.com/
"that looks as though it was written by someone who found writing YouTube comments too intellectually challenging"
Haha, thank you so much, that made me laugh hysterically!
Great topic, I agree Q&A websites hold great potential not only for getting your brand out into the market yet also from a linkbuilding front.
I agree sites like Quora are fantastic yet the only issue with them is that the demographic is limited, when you get more into Yahoo Answers an old school Q&A site the type of content is way more in depth accross a whole bunch of niches, which I love.
But yeah I currently use a huge number of Q&A website, really reccomend people to get onto them if you are not on them.
I would love to see some of this stuff on Quora.
I thought about building a niche Q&A site about 1 year ago, but I abandoned the idea because there wasn't a good off-the-shelf technical solution for it.
Maybe things changed, so does anybody know a good script, preferably well designed and written in PHP to kick-start a new Q&A site?
There is a platform called Qhub, I've played around with it in the past, and it works well. Very similar to Quora.
This post would have been great...5 years ago!
You joined just to make that comment... bitter about something??
I joined just so I could "lol" that. Q&A sites do seem somewhat antiquated though.
Getting involved on Q&A sites is a great way to build your brand and authority as an industry expert. Having a Q&A section on your site is also a good way to develop new content and build up your customer service in the process. If your site ranks when someone asks a question, you have just become a resource for that user.
I must confess I am a Q&A addict - don't often contribute however regularly read them, the more I read them the more I seem to be turning into a hypochondriac - maybe I should stick to questions about search marketing!
I'm looking forward to earning 500+ Moz points so I can contribute to your Q&A board in the future!
[link removed]