incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Weir <apa...@robweir.com>
Subject How about a next-generation approach to user support?
Date Sat, 25 Jun 2011 13:33:16 GMT
>From a user's perspective, maximum satisfaction comes from not having
a question in the first place.  The product just works intuitively.
They don't need to interrupt their task to find out '"How do I do X?".
 If they need to search for an answer to a question, the product has
already failed them, at least in a small way.  The next best thing is
to have an answer readily available in the in-product documentation,
in the help system, which should be indexed and searchable.  If that
fails, then the user typically searches the web.  If they cannot find
an answer at that point, then they end up posting a question on a
support list or support forum.  If they get an answer 24 hours or 8
hours later, or even 1 hour later, then their task has been
interrupted.  The product has let them down.  Of course, this is far
better than the customer just giving up and not telling us what was
confusing to them.  It is better than them moving to another product
because they cannot figure something out with OpenOffice.  But it
still represents a long string of lost opportunities.  The user
doesn't want support.  They want to work on their task.  Support is
the last resort.

So we should think of a support site as **a way for users to get
questions answered**.  This does not necessarily mean that they need
to post questions.  A good support site will make it possible for
users to find answers to many or even most questions without having to
ask them.  This is done via FAQs, good search facilities, etc.  A good
support site should also be SEO-savvy, and follow best practices for
indexing by Google, Yahoo and Bing, and others.  Most users with
questions will start from the search form in their browser than will
start at whatever support site we create.

So this leads us to want to maximize the value of questions and
answers previous asked, to build a repository of wisdom and
experience, especially for the most-asked questions.

I think it is critical to keep in mind the accumulative value
generation aspect of lists and forums, how the questions and answers
build up a repository of information that helps us (the project
members) understand the common user problems in the product. But it
also feeds the search engines, so users can find the information they
seek without having to post a question.  With commercial software
vendors, you often see this feedback loop.  The support department
identifies the top issues for new product releases and feeds that
information to development.  They also look at most common questions
and write them up as tech notes or FAQ's.  Developers take turns
listening in on support calls, so they have a greater appreciation for
the problems that real customers are running into.  The ultimate goal
is user/customer satisfaction.  Not volume of support calls, or number
of questions asked, or number of forum posts.  A large volume of
support calls can often be an indication of product quality problems.
It is a bad thing.  You want to maximize user satisfaction.

I fear that if we fragment user questions into both forums and lists
then we reduce the value that we could have from a unified repository
of questions and answers.  In fact, I think we should be bold at this
point.  The start of an Apache project is a natural break in the
evolution of the OpenOffice project.  We don't need to copy the
methods that were used before.  Maybe we should look at the best
practices available today, and leap ahead to something more powerful.

Someone mentioned before the StackExchange style of community led,
collaborative/social Q&A site.  If you are not familiar with these, I
urge you to poke around a site like http://superuser.com/

These types of sites are not designed for long discussion threads.
They are optimized for Q&A.  They bring in features like question
tagging, rating of questions and rating of answers, a wiki-based FAQ
system, etc.  Community members earn reputation points and badges for
asking and answering question.  These points earn additional
privileges related to scoring of questions and answers, moderation,
etc.  So the site is self-policing.  This point system also makes
participation quite additive, and that is a good thing.  The tagging
of questions, and merging of duplicate questions adds a
self-organizing aspect to the information as well, which is very
valuable.  You can also assign some of your points as a "bounty" or
reward for someone who answers a particularly hard question.

I think something like this represents the state-of-the-art for a user
support forum.  If we had this for OpenOffice.org, it would give our
user community a coolness factor as well increase our support
effectiveness.

Forming a new StackExchange-based site is free, but does require
demonstrating a high degree of community support and commitment.  It
appears that you need a solid group of around 100 people willing to
commit to participating in the site in order for it to be created.
That should be possible for OpenOffice.

The process is defined here:

http://area51.stackexchange.com/faq

Regards,

-Rob

Mime
View raw message