Designing web-surveys

As more people have access to the Internet, it is becoming easier to use the Internet to conduct market research via web based surveys.

The basics are similar to most market research surveys. Decide who you want to talk to. Find out how to get hold of them online and then design mechanisms to find the information you want. The major difference with web- surveys is that you are able to design much more sophisticated surveys than could be carried out either on paper or via the telephone. In addition, you need to have surveys that are designed to look good and engage the respondent.

SurveyGarden is our dedicated on-line research facility for hosting our own and third party questionnaires.

Recruiting and online samples - hints and tips

On-line surveys are the main means of collecting quantitative data on the Internet. The main forms of Internet research comes from surveys placed on a web site (hosted surveys) either for access via an email address or via a pop-up survey on the site itself.

Of the two, pop-up surveys are very good for direct research about the web site being visited as people are recruited on the spot, but pop-up blockers that come with many browsers mean that many potential respondents will not see the invitation. Increasingly Javascript is being used to overlay the survey on top of the page to get round these pop-up blockers. These normally redirect people to a conventional web-survey

Because of the problems of controlling the sample, the main method of Internet research is therefore to send emails direct to individuals with links to the on-line surveys URL. This can be direct from the research agency or using a panel specially recruited to take part in research. Depending on the survey itself the URL and questionnaire may include aspects such as unique IDs or password access to prevent public access to the survey and ensure only valid respondents take part, and to reduce problems of one person completing several surveys (known as 'professional' respondents). Companies such as Toluna, GMI, Sample Surveys, Research Now provide access to online panels for consumer and specialist audiences such as medical or pharmaceutical research.

Where a panel or a direct email list is not available, an open survey method might be used where people are contacted (by phone or email) and asked to take part or to spread a link in a 'snowball' sample where friends are encouraged to ask other friends to join in. This can be particularly useful for special interest groups like sports associations, or using social networking sites.

Smarter questionnaires

One of the major strengths of using the Internet for collecting information means that you can control the content more tightly. For instance ensuring that routing is correct, validating answers or using previous answers to generate new questions for tasks such as pricing research, managing multiple languages and currencies in one survey for international market research, offering techniques such as predictive text and more interactive approaches to filling in a questionnaire.

In addition Internet research opens up new possibilities for research methods because of the availability of a computer to drive the questionnaire. We have designed numerous 'internet-only' research techniques such as anchored scale or repertoire-based conjoint analysis for collecting information about how people make decisions. In addition, with larger sample sizes available at lower cost, you can radically transform the way you think of a questionnaire. Our non-linear research technique exploits the fact that research questions are rarely asked in the order that is of most interest to respondents.

Despite the newer forms of questionnaire, it's still the case that an online survey needs to be sharp and to the point. Grid-type questions tend to grate on respondents if used excessively (try sliders for a more interactive approach, but still limit their use). The only area this is not the case is in text-based questions, where typically respondents give much longer answers than in any other form of market research. Respondents also expect to be able to say what they think, so always give some space at the end of the questionnaire for extra comments.

DIY online surveys

If you just want regular bog-standard market research, there are plenty of off-the-shelf packages which are readily available for producing web-surveys and there are sites such as Zoomerang or Surveymonkey, which allow you to field simple questionnaires quickly and cheaply, sometimes even providing you with sample too.

However, for us it is the fact that web-surveys open up a range of different questioning and investigation techniques that makes Internet research such a rich area. For instance rather than using scales or measures of likelihood of use where answers are notoriously difficult to interpret into the real world, it is much more realistic to ask questions involving real choices. We produce our surveys in-house allowing us to build in elements such as trade-off exercises, BPTO or even to use uniquely internet techniques such as non-linear questionnaires.

For questionnaires that are longer than a 10 minutes, it is advisable to use incentives and reminders to encourage people to take part. The overall design of the questionnaire is also important as on-line surveys do suffer drop-out rates. This is not just about the design of the questions, but also making sure that the look of the survey is also properly designed and will work on the wide range of browsers that are in-use.

Once data are collected, producing analysis and reports are carried out in a similar fashion to other market research. However, the use of computers means that results are typically available faster than in non-Internet research. With many of the new techniques we are developing, there is increasingly the possibility of creating dynamic market models which help marketers look at the value of trade-offs in the markets they serve

However, easy though some internet research seems to be, controls are needed to check interview quality, particular for incentivised surveys. How do you ensure one person one survey? Do you have check questions for consistency and some form of quality measurement? Are there external pieces of information that can be cross-checked with answers to know you've interviewed the right person?

On the qualitative side of internet research, there are a number of research possibilities. Email or forum-based groups are lead by a moderator who introduces everyone and starts with some simple open-ended questions. In the subsequent discussions via email respondents here from each other and from the moderator who probes and tests the needs and wants of those in the group. Emails can be passed to the group directly, or after checking via the moderator (for instance to avoid flaming or bad language), or in summary form from the moderator (to reduce the volume of email that has to be read).

Such discussions have a formal discussion guide so the moderator can ensure all topics are investigated and typically last one to two weeks.

A common alternative is chat groups. These require several people to simultaneously log in to a chat area where a discussion is held (everyone typing frantically). These are most analogous to traditional qualitative groups, but responses tend to be shorter and more abrupt because of the typing involved.

Naturally, as there is no visual or face-to-face element to the qualitative research, it can be difficult for the interviewer and other respondents to understand things like tone or body language.

More advanced qualitative techniques include things like delphi research where experts are invited first to complete a survey and then after everyone has completed the survey, they are shown the results and invited to change their response or justify why their answers might be different from the consensus.

Community research forums are dedicated groups set up as an interactive panel for feedback and qualitative discussion in a social networking type space with commenting, forum and messaging between users plus polls and information updating to test reactions.

Blog research and text-analysis

There's a lot of hype now around blog, news and social networking research - listening in to the worldwide conversations that people are having about products and services and using text analysis to drill down and capture insights. Listening to online chatter has been around since the mid-1990s, but the volume now available means that automated tools are now available to be able to capture, track and analyse this information. Whilst it can be interesting, we have found there is a danger that the loudest or most eloquent opinions are heard often drowning out the views of genuine customers and so potentially steering the business in the wrong direction. Note that this type of research should also be very cheap, since in reality any one can do it

For help and advice on carrying out on-line market research contact info@dobney.com or take a look at SurveyGarden.