Michelson & Associates, Inc.

Internet Surveys (web based customer surveys)

Micheslon & Associates Inc. Internet Surveys help customers gain valuable web based information, measuring performance and customer satisfaction.

With the explosion of the Internet today, Internet Surveys are a highly effective way to reach consumers on line.

Internet Surveys are useful for:

•
Developing questions

•
Understanding key metrics & statistics

•
Performing practical analysis

•
Increasing response rates

There are several benefits of performing research on line:

•
Less expensive

•
Broader reach

•
Speed

•
Flexibility

•
Automated reporting

•
Wider range of options for collecting data

There are also limitations of on-line research:

•
Difficult to target customers

•
Recruiting

•
Validating

•
Logic controls

•
No true face to face interaction

Internet Surveys can be used to track both Satisfaction & Performance Research.

Tracking Satisfaction:

•
Customers & Employees

•
Identify and measure key experience factors

•
Measure factors that can be changed

•
Relative to expectations

Tracking Performance:

•
Operational in nature

•
Employee performance

•
Systems performance

There are seven main types of Internet Surveys:

1.
Email (text)

2.
Bulletin Boards

3.
Web HTML – Flat Form

4.
Web Fixed – Form Interactive

5.
Web Customized Interactive

6.
Downloadable Surveys

7.
Web-Moderated Interviewing

The pros and cons of using different Internet Survey Methods—Which works best for you?

1. Email (text)

-
The first internet research method

-
It’s easy, similar to paper and pencil surveys

-
No logic testing (Screening only, no if-them logic, limited auto entry)

-
Best for internal surveys

-
20 questions limit

-
Low rate of future interest (boring format)

2. Bulletin Boards

-
Modified Delphi Method (collected over time, allows for build on topic)

-
Function like a slow motion focus group

-
Thread is visible to all participants

-
No automated analysis – all open ended

-
Best for customer panels

3. Web HTML Surveys

-
Flat HTML form is most common

-
Participants click buttons and boxes, fill in text, then submit form

-
Graphics, audio and video clips, animation can be used

-
Analysis requires CGI scripting (new software will automate analysis)

-
No interactive logic controls

4. Web Fixed-Form Interactive

-
Based on 3rd party survey authoring tools (CATI, DBM – allows for logic controls)

-
Plays questions on the web the same way as an interviewer

-
Limits the range of options in which a survey can be displayed

-
Few tools allow for self hosting (those that do are expensive)

5. Web Customized Interactive

-
Most powerful and flexible option

-
Involves highly skilled programmers

-
Provides all modern technical controls (screening, skip patterns, logic, error checking)

-
Flexibility in layout is the main benefit (aesthetics fits within frames , dynamic and descriptive navigation pop-up messages)

-
Almost always done on in-house server

6. Downloadable Surveys

-
Requires previously installed software

-
Survey form is downloaded to respondent

-
Allows for survey to be completed off-line

-
Best for sophisticated respondents (Panels or pre-recruits

-
Expansive and time-intensive

7. Web- Moderated Interviewing

-
Good for chat interviews and other discussion formats (like a focus group)

-
Logic and control provided by skilled moderator

-
Best for bringing together geographically displaced participants

-
Participants must be qualified, pre-recruited and motivated

How To Develop Internet Questions:

Study your objectives
What are you trying to find out/accomplish?

Which type of Q&A will work best for you?

Binary responses (yes/no)

-
These are easy to analyze.

-
Make sure you ask only one question

-
Question should be neutral, not leading

-
Offer neither N/A or undecided

-
Best when followed by "Why?"

Likert scales (from 1-5)

-
Best for measuring satisfaction

-
Allows determination of how hot or how cold

-
Scales could force polarity or allow for neutral

-
Easy to tabulate by average /mean distribution

Multiple response

-
Which areas would you most like to see improved?

-
More difficult to tabulate than single response questions

-
Percent base will be higher than sample

-
Cannot be represented in a pie chart

Open-ended

-
Allows respondents to mention anything not covered in survey

-
How else may we better serve you in the future?

-
Also allows for "other" specifics in multiple response

-
Most difficult to analyze (code according to most frequent responses

Make Sure You Understand Key Metrics & Statistics:

-
Metrics are numbers used to indicate satisfaction/performance

-
Baseline metrics define current status

-
Trend metrics define changes over time

-
Key metrics need to be easy to grasp

-
Overall satisfaction = 82% very satisfied

-
Indexes are commonly used to compare multi-location performance

-
Random sampling is the key (every 7th respondent)

-
Basic rules of sample size confidence

• 100 = 90% +/- 5
• 400 = 95% +/- 5
• 1,024 +/- = 98% +/- 3

-
Awareness of segment samples

The Keys To Practical Analysis:

-
Keep things simple

-
Try to automate analysis and reporting

-
Decide on a metric and stick to it

-
Test metrics and wording before implementation

-
Changing wording of questions changes results – stay with original wording for trend analysis

-
Develop a baseline metric be conducting the survey with a strong sample the first time (over sample if needed)

-
Focus on simple statistics (ave. frequency, percentages

-
Data mining can be done later for more specific analysis

-
Report on total sample (segment at own risk)

How To Increase Response Rates:

-
What’s in it for respondents?

-
Is your topic interesting?

-
Make it easy for participation

-
Use monetary incentives (drawings, cash)

-
Information/access incentives

-
Multi-media recruiting

-
Email, banner, pop-up invitations

-
Fax, voicemail, letters, invoice memos

Some Final Words of Wisdom On Internet Surveys:

-
Make sure you understand your main objectives

-
Know your audience

-
Make your survey fun/interesting

-
Keep questions relevant

-
Keep it simple

-
Outsource data collection/reporting

Effective On-line Mystery Shopping:

-
Measure service performance

-
Pose as a real customer

-
Make inquiries or place orders

-
Capture customer experience based on pre-determined criteria

-
Let employees know it will occur

-
Let employees know what’s being measured

The cost of a typical NetSurvey project include:

•
Consultation

•
Questionnaire design

•
Participants Incentives

•
Sample construction

•
Data tabulation

•
Graphic presentation of all data

•
Analysis

•
Written and verbal presentation of results with specific marketing recommendations

For a Specific Quote On a NetSurvey For Your Company, Please Click Here

Michelson & Associates, Inc.
SiteSurvey Services
For a quote on a customized InternetSurvey study,
call Mark Michelson @ 516-576-1188
or e-mail: solutions@serviceevaluation.com.

Michelson & Associates, Inc.
C/O Service Evaluation Concepts, 210 Crossways Park Drive
Woodbury, NY 11797
V: 516-576-1188
F: 516-450-3466
solutions@serviceevaluation.com