Customer Satisfaction Survey Options

[Faith]

Hi All,
Is there any useful link from which i can get some information to create a customer satisfaction survey for a web project?
I tried searching Google....but ddnt find what i was looking for...can someone pls help me.

Faith

[bwelford]

I tried the Google search for "customer satisfaction survey", in other words with quotes. There seem to be some forms there. Were none of them suitable, Faith? If not, perhaps you can tell us what they were lacking and we can try to find another answer.

[bragadocchio]

Hi Faith,

It might help if you map out some of your requirements for us here.

Should the survey run on a certain type of database, such as MySQL or SQL or another. Are you looking for an asp solution, or a php one, or something else?

Would you want it to appear as a separate html page, or as some type of popup, or is what you are looking for some type of email template after a person has completed some type of transaction?

Do you mean "customer" as in someone who visited the site and completed a transaction? Or just a visitor to the pages of a site.

The more information you can provide us about what type of requirements you might need to meet, the more helpful we can be.

[travis]

Faith,

It is probably worthwile learning some variable definition and capture techniques and learn how to program forms. You may also need some relational database design skills.

If you approach it from that perspective, then its a simple matter of a few pages designed with variable storing procedures and a final submit button.

Surveys can be devised simply in Flash, asp, .NET .php and other languages, but I suggeest that you will require some form of active server to record data.

If you can store variables between pages, then you can isolate each question to their own respective page, which is a nice effect, and then you can have back and forward buttons if people want to check their questionarre, form or survey.

We model surveys around multi-page forms.

Example in asp.NET

http://www.parallel-solutions.com.au/step1.aspx?jid=4

Example in Flash (currently under construction)

http://hutchens.opensearch.com/pps/employm...employment.html

Things you have to worry about are

(a) What if people type nothing or click nothing ?
( What if the same person comes back and fills the survey again ?
© What indication will you give them about the results and when ?
(d) Autoresponse email once the survey is completed.
(e) Database capture of information.
(f) Subsequent retrieval of data into a meaningful display.

Its as simple or sophisticated as you want to make it.

If you learn it from a programming perspective, then you can use your system for surveys, employment forms, online testing, contact systems, and a wide variety of other functionality.

[Faith]

before you all start groaning...let me apologise first. SORRY...i should have been more specific. so here goes.
Currently I am involved in one of the biggest projects in the industry here. our team deals with many departments within our client's oraganization. We need to find out whether the client is satisfied with the progress of the project, the technologies used, the concept, the QA process, etc.
When it comes to the survey I am more concerned abt the type of question to ask, how to pose them, basically in tht line....
Sorry if i mislead u guys...thousand apologies :-)

[Fabius]

The best and most detailed I found using the keywords "customer satisfaction survey form example" on a metasearch engine is located here.

Hope that helps.

[DCrx]

People lie on surveys and focus groups, often unwittingly explains Nielsens's first rule.

A company conducted focus groups for their Product X, which had as its main competitor Product Q. They asked people who were using Product Q, \"Why do you use Product Q instead of Product X?\" The respondents gave their reasons: \"Because Product Q has feature F,\" \"Because Product Q performs G faster,\" \"Because Product Q lets me do activity H.\" They added, \"If Product X did all that and was cheaper, we'd switch to it.\"

Armed with this valuable insight, the company expended time, effort, and money in adding feature F to Product X, making Product X do G faster, and adding the ability to do activity H. They lowered the price and sat back and waited for the customers to beat a path to their door.

But the customers didn't come.

Why not?

Because the customers were lying. In reality, they had no intention of switching from Product Q to Product X at all. They grew up with Product Q, they were used to the way Product Q worked, they simply liked Product Q. Product Q had what in the hot bubble-days was called \"mindshare\", but what in older days was called \"brand loyalty\" or just \"inertia\".  

Customer satisfaction only measures the difference between what customers expect to get minus what they perceive they get. For example, a study of satisfaction among grocery shoppers found that customers were most satisfied with canned food. Did they love canned food? No. But they got exactly what they expected to get.

Sacrifice is the difference between what a customer really wants and what they are forced to accept. Customers might compare the great service they got at one kind of business as the yardstick they measure unrelated business service by. A great working app of one type makes the user wonder "why can't all applications work this easily." Finally, satisfaction is based on past experience, not a novel new user experience which nobody has yet provided.

I've never seen a well designed, or even use centered online survey. And the proportion of "how do I code a form" technical questions to "how to I learn anything useful" questions is inexplicable. (Even after the Kerry exit poll debacle). However a very good intro on question and survey design can be found here.

Fund a feature and bounty systems (for bugs) actually get at information.

Also check out this related thread Customer Satisfaction Doesn't Count.

[stever]

For some great pointers you can go to the Council of American Survey Research Organizations ( http://www.CASRO.org ) or the Marketing Research Association ( http://www.mra-net.org ) for info on survey design as well as industry guidelines for ethical internet market research.

Everything written above is basically correct, but there are a few caveats.

Most people are honest when completing surveys - but often they do not know their true motivations and often their stated preferences do not help you much.

If you ask people to rate the importance of different aspects of airplane travel, for example, they will tell you that seat size and food is not that important compared to safety. But seat size and food (or length of flight) are critical in their choices of what airline to choose - until one airline has a crash.

Similarly, if you ask owners of Mercedes and owners of Hundais to rate their cars based on their overall satisfaction, Mercedes comes out ahead. If you ask them about their expectations, then Hundai comes out ahead. Mercedes owners have higher expectations. Different markets.

It is all in how you ask the questions and how you analyze the data. For a website, you may want to ask a likelihood-to-recommend question and then ask about satisfaction with various aspects of the site. Then you can create a statistical model that will tell you which aspects of the site can be best improved for the best increase in customer recommendations.

I would recommend avoiding paying people to take surveys. Paying people to take surveys only means that they will answer quickly and often and less truthfully.

I would also recommend asking only as many questions as you absolutely need to ask.

I hope that this helps.

Steve Runfeldt
justASKthem.com

[cre8pc]

An excellent thread, that prompted me to contact one of my Mentors, Steve Runfeldt, of JustAskthem (his company hosts these forums) to check out.

Steve's experience with customer satisfaction surveys (and Marketing) is long and impeccable. In that history, he was Vice President of CustomerSat.com, before forming his own company, which develops web applications that analyze survey data for large corporations.

I'm always interested in this subject and how I can translate it into web site feedback forms and surveys. Like user testing, it requires skills and a deep understanding of the nature of surveys and how respondents will respond. And, how to analyze the data properly.

Hence, the knowledge in this thread is quite interesting!

Steve, you've tried to teach me the proper and least biased way to frame survey and feedback questions. Do you have any basic guidelines or suggestions for Faith, and possibly some quick info on how to set up ratings types questions?

[Faith]

Hi all,
Thanx for your feedback, really appreciate it. found some really interesting links, especially the "Master the concepts of questionnaires and survey design" tutorial.

Cre8pc, u took the words right out of my mouth "Do you have any basic guidelines or suggestions for Faith, and possibly some quick info on how to set up ratings types questions?", just what i want.
Hope Steve replies!..:-)

Thanx all.
Faith

[TheManBehindTheCurtain]

On the implementation side: In previous jobs I've scripted forms, used forms created by a consulting company, and used Survey Monkey (http://www.surveymonkey.com/).

Considering that you can subscribe to Survey Monkey for $20 U.S. a month and get up to 1,000 responses, that's the way I'll go in the future. All the technical implementation is done for you, including reporting. This freed enormous amounts of my time for more important tasks (like actually designing the survey questions) and was far more dependable. If you've got the budget for it, or can pass it along to a client, it sure beats reinventing the wheel.

Good luck!

[fisicx]

Had quite some experience myself in this field - often on the receiving end of useless surveys but many times as the originator of the survey.

Time and time again we discovered that people's perception of a project and hence any survey about the project was biased towards what they knew about the project.

As an example. We were publishing a UK wide aviation policy document. We wanted to get people's views on what type of policy they wanted (they were after all the end user). But asking a question like: what areas of the existing policy require improving?' will elicit such a wide range of responses that nothing meaningful could be determined. What we did instead was to hold a number of workshops. We explained the reason for the change (to make cross organization operations simpler), how we intended to implement the change (piecemeal) and then asked the question: which area of policy has the most impact on your work? We then allowed each organization to discuss and then report back. We received about a 65% response and there was a clear majority on what we needed to fix first.

The point of all this. A cold survey can be very subjective. Face to face discussions followed by a survey will be much more objective.

And although people do not lie, they don't like change - if the project is changing the way they do things they are more likely to express dissatisfaction. But talk to them first and they will be your friend.

As to the actual survey - make it as simple as possible. But do not ask people to grade things from 1 to 5 unless you are absolutely clear what each grade means. Comes back to the satisfaction thing. Ask: were you satisfied with the delivery of the product? If you think 5 is excellent then what you think is excellent may for me be only very good so I will give it a 4. But what you really meant was: did the goods arrive within the expected time frame, was the courier courteous, was the packaging undamaged, etc. The answer is yes or no - with room to explain why if I want to.

Not an easy thing to do a customer survey, I wish you luck.

[AbleReach]

Thanks to all for this thread!


With another "wow" for cre8asite,

Elizabeth

[fisicx]

What I should have added is that peoples answers to a survey are often influenced by recent events.

Ask anyone what they think of politicians after one of them has been convicted of fraud and you will get a very different answer from that given after a politician donates a weeks wages to a charity.

And another little story.

We were having a huge problem implementing a new system of ordering stationary. The admin office was finding all sorts of problems with the new system and it was only after talking to the head of the department we discovered that she used to be in charge of the stationary cupboard. It was her little empire and we had taken it away from her. Even though her job was now a lot easier she was upset that her responsibility had been eroded. Result, one dissatisfied customer - but dissatisfied for the wrong reasons.

Peoples perceptions are clouded but their own prejudices get the people on your side, involve them in the process and you do not even need to do a survey.

Probably a bit late for you now, Faith, but I would recommend binning the idea of a survey and going to talk to people instead - it makes people feel wanted and involved.

[bwelford]

Two very good points there, fisicx.

... peoples answers to a survey are often influenced by recent events.

That includes even the questions they've already answered in the survey. Depending what you've asked, you may have got them in to a certain mindset so that they are predisposed to anwer a later question in a certain way. It's more a problem in political or social surveys, but it's always something to think about.

... I would recommend binning the idea of a survey and going to talk to people instead ...

Well even if you're not 'binning the survey', it's always a good idea to talk to some people in a completely open-ended way to find out what are the important issues for them on the topic. Then you can make sure your survey covers those aspects. What you need is an undirected or unfocused discussion. The experts in consumer research who charge big bucks for doing this call them 'Focus Groups'. I've never understood that term. :?

[DCrx]

Focus groups are not quite the answer either.

It isn’t enough to hear what people say they want. You have to find a way for people to show what they want, and that won’t happen in the stilted environment in which focus groups occur.

Because innovation is more important in today’s economy than ever before, the guidance from focus groups can be downright dangerous. Most focus groups simply confirm to companies that what they are doing is right—give or take a tweak—and discourage them from striking off in a new direction. For fundamentally, people answer the questions asked of them. All too often, the answers provided are as much a function of the thinking from which the questions are generated as they are an indication of true desires.

....With focus groups, the medium is the mistake. Having participants talk about what they want inherently limits the prospects of uncovering new opportunities. It is the using of a good or service that defines the customer experience, so you have to observe customers in action to understand their unarticulated needs.
-- Focus Pocus

I agree with a previous comment -- users and customers don't lie (putting it that way makes for an attention-grabbing headline). But many of these techniques assume a degree of self-insight that isn't realistic. Put the same soda into cups labeled "1" and "2" and cup one gains a statistic edge.

Product testers create a more random looking stream of alphanumeric text to identify products and mask out this effect.

To sum up, the very attraction of focus groups and surveys is they seem like an easy fix. Just put up a web survey ....what could go wrong?

Too often the surveys and focus groups are used to support a decision which has already been made, not to inform one. Secondly, these technique seem like they would take less time, less money (and require less careful thinking) than a proper user test methodology. This mindset going in sabotages the process of discovery before it gets started.

The Japanese have a term "three actuals. Actual user. Actual place. Actual situation." This is the way to get context into the mental model people bring to things like product use. By watching for unexpected behavior, one can gain insight into questions they wouldn't have thought to ask. And there is less chance to focus exclusively on the product's good points.

[AbleReach]

Users who see the survey and fill it out before they've used the site will offer irrelevant answers. Users who see the survey after they've used the site will most likely leave without answering the questions.

From DCrx's what could go wrong

They suggest formal testing using a captive audience, with a survey at the end. Both what users say and what they do are taken into consideration. To continue:

With techniques like paper prototyping, you can test designs and question users without implementing a thing.

Sounds elaborate but much less expensive and much less confusing than trying to do it all online. You'd also be testing more than what is already in use, and so getting to the problems and potentials before confusing the public with fluxuating designs.

Fascinating topic. Good thought and resources. Thanks!


Elizabeth