Usability in Web Applications

I wanted to sample the usability of Web Applications to actual business users and how they perceive value. In order to do the same, I went ahead and created a survey at LinkedIn : http://www.linkedin.com/answers/technology/enterprise-software/TCH_ENT/36530-4256932

 The advantage of a global community like LinkedIn is we always get high participation from people who really think and try to innovate. Without an exception, the answers show some real insights from real users of Applications. I think all of us developers might periodically learn something from the end-users and align our priorities. Hope everyone finds the survey as useful as I did.

Synopsis

This discussion has added so much value that I decided to share it over here...it has been truly global and I can't thank enough the few good men who participated in the discussion...so many horizons are opened when we get exposure to such a rich group of users...

My Question:

--------------------

When you use web-based business applications, what is the single factor you appreciate the most?

I would appreciate to-the-point answers, like "Speed" or "Less no. of
postbacks" or "Lots of tooltips/help links" etc. However, your comments
are definitely invaluable and could not understate their importance by
any means. So feel free to give a one-word reply and then expand it.

Intention : Being a business application developer, my sole
parameter for success is "Quality as perceived by the end-user". Want
to increase that score.

 
Ed Lass,  Albany says

OK, I'll bite:

Results for the user--the app gets the job done better than if it didn't exist.

Roshan Shah, Canada says 

Minimal Clicks to get to what I want and fast loading of page and small forms to fill if any.

Devesh Dwivedi, Washington DC says

User friendliness: The application
should be very user freindly and by that I mean easy to comprehend and
apply/use as well. It should be a breeze and not a plop of technical
junk. I love more advanced features and functionalities but if it comes
at the cost of user friendliness, I won't buy that. - My two cents:)

Alok Jain, Washington DC says

If this is your sole prameter, I would suggest putting some time understand users and their needs.

The importance of any factor depends on users and context. For e.g. for
a call center app used by young guys speed is important because their
success parameter is dependent on number of calls they can handle.
While in Internet banking feeling secure and having precise
control/clarity is much more important than speed.

You could apply various techniques to find users' priorities, best would be to just go and observe them and talk to them.

My Take -

I totally agree with that "everyone can't
like chocolate ice cream"! Different segments, age groups and job roles
will definitely have different priorities and they cannot be judged by
a one-size fits all approach. However, what i am trying to accomplish
with this question is getting the opinion of different users in
different professions and see if I can really find out any common
subset - a pattern which would then give us a basic set - after which
the specialization would come.

Shane O'Neil ,Canada says

Integration ... Software as a
Service (SAS) companies that "get it" develop an API. Importing and
exporting data is a necessity; however, an API allows the customer the
ability to integrate the web based app with other applications for more
flexible functionality, back end integrations and reporting.

From a software provider perspective, APIs can form the basis of a
partnership and create stickiness. There is an investment from the
client perspective to code to the API. To change vendors means new
development cycles and hence a new investment so you'll have to really
suck at what you do to make the client go through that pain. Clients
will be more forgiving.

An API will also allow the client to develop functionality against
your app that may be important to them, but off-strategy or a lower
priority to the you.

Two cents. Hope this helps.

Vinod Kumar, Greater Boston Area says

I would extend upon Roshan's answer and say - Usability.

Usability has 2 parts - Intended & Perceived. Intended usability is
quantitaive, can be measured by some of the factors that Roshan
mentioned - e.g. no. of clicks it takes to go from one place in the
application, to another. Perceived usability is subjective - It is how
the user views the software as being user friendly. 

Paulo Arancibia, Argentina says

A minimum number of click and postbacks by each action and a good balance of color.

Thierry Thomas, France says
Portability: it should really run the same way on any browser.
And speed, of course!

Fred Held, Greater Los Angeles Area says (I love this discussion!)
Sorry for the wordiness, but I love talking to developers.
1) Gets the job done faster, easier, 24/7 availability, remote capability
2) Help really works and does not send you to FAQ. I have NEVER found an FAQ that met my needs. Why that is done is amazing to me.
3) Chat help is incredible. I don't care if it is outsourced to the Congo as long as the person helps me and it is available 24/7
4) Mobile capability, over 40 percent of IBM Global services employees work from their home office small office or mobile locations. This trend is accelerating globaly.
5) Reliability and work is not lost due to a system glitch. An example of "cave man" designs is the clarification tool on linkedin. It crashes and gives a 1970 type meaningless insulting error report and the work is lost. Remember the now famos windows error message. " a fatal error has occured" WTF does that mean.
6) Intuitive usability, I can not say too much about user testing for usability. When at IBM we spent a lot of time on the system we were developing for our client on this. We created a user group who we tested. One such group were truckers. That was fun going to truck stops.
7) Downward compatability. MS and IBM are known for not providing this and forcing the user to upgrade making all the work done in the past unusable. This is bad marketing since the investment in this work far exceeds the cost of the upgrade.
8) How do you measure "Quality as perceived by the end-user". here is an example of IBM's Intranet ran by Mike Wing. They measure the soft stuff very well. A key question answered by the user population is the following. "What percentage of the time do you used the Intranet in your work to find out information versus co-workers and your supervisor" In 1994 they started measuring this the exact same way every year. They gained a lot of information in the survey and used the process of continous improvemnt. They then set a paremeter of succes. The paremeter was when 50% of the time the employee used the Intranet ot get the necessary information or help, they have achieved success. In 2000 they crossed 50% and started accelerating towards much higher numbers.

Once again I apologize for such a long answer but I am passionate about helping developers with rigid requirements documents, small budgets and tight time frames create software that is loved not loathed.

My Take -  

 I agree with you - "Don't
make the customer think" should be the mantra for every web
developer....because if I do, someone else won't and the customer would
go to him....and web 2.0 is really a revolution.....its not just a
technical overhaul....but it signifies the control going back to the
customer...the web merging with "customer is the king" philosophy...I
would go as far as to term as "The biggest revolution so far in the
largest democracy (the web)"

what we really lack is to know that "Am I making that guy happy to
whom the tool is being rolled out by the management?" Because at the
fag end of the day, that is success to me.....to enable the user smile
a bit more...as the system makes his life easier...proactively shares
some of his real life work load...I don't want to make him grim with a
200 page user manual :) ...and customer satisfaction is THE GOAL for
me....not knowledge of all the technologies or mesmerizing funda over
recursive algorithms (though I would be more than glad to have both :)
)......
even Simon and Garfunkel sang "Keep the Customer Satisfied" [;)]

Only thing I don't agree with you is the FAQ think....I have often been greatly helped by them:).... 

Fred said

I applaud your objectives. There are so many recommendations I can give
you and want to give you on how to make sure your users are truly
finding your new system a help to what ever they are doing.

Remember these very powerful objectives :

1) Intuitive, it is obvious what I have to do with the system to
get my work done. This is the hardest objective of them all. Think like
a 45 year old with few IT skills and you will succeed. If can't find
one and interview that person for a long time. Do not ever
underestimate your users skills.

2) Reliable, obove all systems must be reliable. Users go crazy with out six sigma reliability.

3) Recoverable, if something should happen, how do the users know
what to do to recover from where they were of back few steps. Lost work
is really bad.

4) Extensible, After a year the really talented users will push the
system to its limits in some areas you never even thought about. You
should be able to extend the capability in that area easily and
rapidly.

5) Secure, you know all about this already.

That is about it.

Sachin Palewar, India says
I appriciate simple interfaces much like google and straight-forward navigation so that its intuitive and visitor doesn't feel lost and confused.
In technical terms use of Web 2.0 technologies and Usability concepts play an important role.

Alexander Samarin , Switzerland says
I think, first, you have to define "Quality" in your sentence.
My favourite - flexibility of business applications.
My experience shows that the business people like when their separate requests for change are quickly implemented in existing systems/applications. These changes are typically small (from the point of view of the business) and unpredictable. 

My Take -  

I couldn't agree more with ur answer...but one thing I want to point out is that
you are talking more from the perspective of the management rather than
that of the end-users. Extensibility would be a priority for the guy
who sanctions the project budget...i.e. if I wanna add x functionality
which was not there, I would like to shell out much less dollars than I
did for the system implementation...it should be incremental and not
break/recreate existing functionality....but that is something we devs
nowadays usually keep in mind from the very design phase....however
what we really lack is to know that "Am I making that guy happy to whom
the tool is being rolled out by the management?" Because at the fag end
of the day, that is success to me.....to enable the user smile a bit
more...as the system makes his life easier....I don't want to make him
grim with a 200 page user manual :)

Alex replied

Just a precision. My answer was exactly about concerns of the
end-users. Because they suffer more than managers if applications do
not make thier "life" easier.

I would recommend to watch "over the shoulder" how the end-users work
and visit them often. ( I think, this is also one of the principles of
Toyota production system.)

I replied

I agree...in fact if you have the book "In Search of Excellence" with you right now, please have a look at Page 28-29, Introduction....it depicts a 28-year old HP engineer being present at a shop floor and demo-ing the features to the customers with the same enthusiasm as that of a professional salesperson....I would consider myself a truly successful professional if I can reach anywhere near that efficiency in the quest to be user-centic.

All in all, a thrilling discussion, am overwhelmed :) 

LinkedIn rocks!