There are many online market research methodologies. This chapter touches on three of the most popular and useful ones: surveys, online focus groups and social media monitoring.
Which methodology should you choose?
That all depends on a variety of factors, from your research question and purpose to your budget and time. Here are some general pointers:
- Surveys: Ideal for collecting large amounts of quantitative data, and some qualitative data. They are quick and easy to set up, and can run automatically.
- Online focus groups: Ideal for engaging consumers and collecting qualitative data such as opinions, ideas and feelings about the brand. They require a larger time investment and a willing group of participants.
- Online monitoring: Ideal for collecting qualitative data on brand sentiment, and can also provide some quantitative data around volume of interest in the brand. This data can be collected passively, and there are several tools that can automate this.
Surveys are questionnaires that contain a series of questions around a specific topic. Their purpose is to gather large volumes of quantitative data easily, though they can also collect some qualitative data.
Conducting surveys online allows for data to be captured immediately, and data analysis can be performed easily and quickly. By using email or the Internet for conducting surveys, geographical limitations for collecting data can be overcome cost effectively.
Technology allows you to compile sophisticated and user-friendly surveys. For example, as opposed to indicating impressions on a sliding scale, respondents can indicate emotional response. Or the survey can be tailored depending on previous answers, such as questions being skipped if they are not relevant to the respondent.
You can run ongoing online surveys at minimal cost. Simple polls can be used in forums and on blogs to generate regular feedback. Website satisfaction surveys are also an easy way to determine the effectiveness of a website or marketing campaign.
One application of surveys is allows for instant feedback on questions or ideas from an existing community, such as a trusted group of thought leaders, your brand’s social media fans, or a pre-created research community. Examples include Facebook polling apps and real-time mobile survey platforms.
How you design a survey and its questions will directly impact on your success. A survey can include any number and type of questions, and more complicated questions should appear only once users are comfortable with the survey. Be careful that you do not introduce bias when creating questions by asking leading questions.
Incorrect: We have recently introduced new features on the website to become a first class web destination. What are your thoughts on the new site?
Replace with: What are your thoughts on the changes to the website? In general, you will also find that you get more accurate answers when phrasing questions in the past tense than in the continuous tense.
Incorrect: How many times a week do you buy take-away food?
Replace with: In the past month, how many times did you buy take-away food? Questions in the survey should be brief, easy to understand, unambiguous and easy to answer.
Types of survey questions
Open-ended questions allow respondents to answer in their own words. This usually results in qualitative data.
What features would you like to see on the website for the digital marketing textbook (https://www.redandyellow.co.za/cours...tbook-digital/)?
Closed questions give respondents specific responses from which to choose. These are typically multiple-choice questions with either one or multiple possible answers. This results in quantitative data.
Do you use the digital marketing textbook website?
What features of the digital marketing textbook website do you use? Tick all that apply.
- Case studies
- Free downloads
- Additional resources
3. Ranked or ordinal
These questions ask respondents to rank items in order of preference or relevance. Respondents are given a numeric scale to indicate order. This results in quantitative data.
Rate the features of the digital marketing textbook website, where 1 is the most useful and 4 is the least useful.
- Case studies
- Free downloads
- Additional resources
4. Matrix and rating
These types of questions can be used to quantify qualitative data. Respondents are asked to rank behaviour or attitude.
Rate the features of the digital marketing textbook website according to the following scale:
1 = love it, 2 = like it, 3 = no opinion, 4 = dislike it.
- Case studies
- Free downloads
- Additional resources
Online focus groups involve respondents gathering online and reacting to a particular topic. Respondents can be sourced from all over the world and react in real time, arguably being freer with their responses since they can be anonymous in an electronic environment.
Online focus groups are ideal for having frank, detailed conversations with people who have an interest in your brand. This means they result in primary, qualitative data. This information can then be used to create quantitative research questions.
Online focus groups can be conducted using a range of technologies. The simplest is to use a text-based messaging program or online forum and there are many options available. More sophisticated tools allow for voice or video conferencing, and can make it easier for the researcher to pick up clues from the respondent’s voice and facial expressions. Some tools allow the researcher to share their desktop screen with respondents in order to illustrate a concept or question.
Good options for conducting online focus groups include:
Google Hangouts: www.google.com/+/learnmore/hangouts
Focus groups are less formal than surveys meaning the researcher will have specific questions to ask, but the conversation usually grows and develops organically as participants discuss their impressions. Usually running for between one and two hours, focus groups are used to get consumer views on:
- New products or marketing campaigns
- Existing products and campaigns, and how they can be improved
- Sentiment around the brand
- Views on a brand’s new direction or visual style
- Ideas for how the brand could improve its position or branding.
Online focus groups are excellent for collecting a lot of qualitative data quickly. When setting up the group, try to include enough participants to keep the conversation alive, but not too many so that some get drowned out by others, eight to ten is a good range. Also consider that you may run into technical troubles if people are connecting from different locations and Internet connections so be prepared to do some basic troubleshooting if this happens.
There are a number of different ways that you can recruit participants for an online focus group. This could include inviting people from your existing customer database, going through a traditional market research recruiting agent, or putting a call out on your website or social media communities. It is common practice to offer a small incentive to people who participate in a focus group, as it is a fairly time-intensive activity.
Finding out if people are talking about you is quite difficult in the offline world, but almost effortless online. Rather than having to conduct real-world surveys and interviews, in the digital world you can simply ‘listen’ to the conversation happening about you.
Keywords – the foundation to categorising and indexing the web – make it simple to track conversations taking place online. Customers don’t always use channels designated by a company to talk about that organisation, but the good news is that the Internet makes it easy for a company to identify and use the channels that customers have selected.
Online tools allow a company to track mentions of itself, its staff, its products, its industry and its competitors or anything else that is relevant. This is called online monitoring, online listening, or data sentiment analysis. It involves using digital tools to find and tap in to existing conversations. The tool then gathers and collates all the mentions it finds, so that you can analyse the data for insights.
Typically, searches include the following main focus areas:
- Brand name
- Key products
- Key personnel (names, job titles, etc.)
- Key campaigns and activities
- Brand names
- Product launches
- Website updates
- Job vacancies
- Key people.
There are four different types of searches you can perform to track relevant brand keywords. Each modifies the specific type of data collected and aims to improve the quality and depth of the data you gather.
The four operators are:
- Broad match – for example, Apple Computers. This is when any of or all words must be found in the mention.
- Direct match – for example, “Apple Computers”. This is denoted by quotation marks and dictates that the tool should find mentions only where the phrase appears complete and in order in the content.
- Inclusive match – for example, Apple +computers. This is denoted by a plus sign directly before a word or phrase. This will direct the tool to search for any mention that contains both Apple AND computers, although not necessarily in that order.
- Exclusive match – for example, Apple –fruit. This is denoted by a minus sign directly before a word or phrase. This will instruct the tool to include only mentions that contain the first word or phrase but not when the second word is also in the same mention.
Combinations of these four types of searches (operators) can be used to improve accuracy.
“Apple Computers” +”steve jobs” –fruit.
Applying this theory to the groupings above, some keywords used for Apple might be:
- “Apple computers”
- Apple +Macbook, “iPod Nano”, “Macbook Air”, “iTunes” +music –radio
- “Steve Jobs”
- “Consumer Electronics Show” +“Las Vegas”
It is also important to track common misspellings and typos, all related companies and all related websites.
Tracking the names of people key to a company can highlight potential brand attacks, or can demonstrate new areas of outreach for a company.
Brand names, employee names, product names and even competitor names are not unique. To save yourself from monitoring too much, identify keywords that will indicate that a post has nothing to do with your company, and exclude those in your searches.
For example, “apple” could refer to a consumer electronics company, or it could appear in a post about the health benefits of fruit. Finding keywords that will indicate context can help to save time. So, you could exclusive-match words such as “fruit”, “tasty” and “granny smith”.
Tools for data sentiment analysis
Thankfully, online listening does not entail hourly searches on your favourite search engine to see what conversations are taking place online. There are many different tools that monitor the web, and supply the results via email alerts or a web dashboard.
The ideal gas law is easy to remember and apply in solving problems, as long as you get the proper values a
Google has several bespoke search services, and periodically adds more to the list.
- Google Alerts: www.google.com/alerts. Google Alerts will send an email when the keyword is used in either a credible news item or a blog post.
- Google News: news.google.com. Google News searches all news items for mentions of a keyword.
- Google Patent Search: https://www.google.com/advanced_patent_search. Google Patent Search allows you to keep track of all filings related to an industry, and searches can be done to see if there are patent filings which might infringe on other patents.
- Google Video Search: https://www.google.com/videohp?hl=En. Video Search relies on the data that have been added to describe a video, and will return results based on keyword matches.
In addition to these mostly free tools, there are also a number of premium paid tools available to make the process easier and more robust. See section 4.7 on Tools of the trade for more suggestions.
Other avenues for online research
There are various tools available to the online researcher for conducting personal interviews, such as private chat rooms or video calling. The Internet can connect a researcher with many people around the world and make it possible to conduct interviews with more anonymity, should respondents require it.
Taking its cue from offline ethnography, online ethnography requires researchers to immerse themselves in a particular environment. In this way insights can be gathered that might not have been attainable from a direct interview. However, they do depend more heavily on the ethnographer’s interpretation, and are therefore subjective.
Online research communities
Although online communities are a valuable resource for secondary research, communities can also provide primary data. BeautyTalk is an example of an online research community that helps gather research data. The community platform can be used as a means to elicit feedback about products and can generate ideas for new products. This is qualitative data that can aid the company in exploring their research problem further. In many cases, social media can be used to gather insight about a brand or customer experience. It is important to remember, however, that a representative sample is necessary for making solid conclusions.
When developing websites and online applications, usability testing is a vital process that will ensure the website or application is able to meet consumers’ needs. Listening labs involve setting up a testing environment where a consumer is observed using a website or application.
Conversion optimisation aims to determine the factors of an advert, website or web page that can be improved in order to convert customers more effectively. From search adverts to email subject lines and shopping cart design, tests can be set up to determine what variables are affecting the conversion rate.
The Conversion optimisation chapter covers tools for running tests, such as A/B split testing and multivariate testing.
How to get responses: Incentives and assurances
As the researcher, you know what’s in it for you when sending out a survey. You will receive valuable insights that will aid in making business decisions. But what is in it for the respondents?
Response rates can be improved by offering respondents incentives for participating in the research, such as a chance to win a grand prize, a discount or special offer for every respondent, or even the knowledge that they are improving a product or service that they care about.
Some researchers feel that monetary incentives are not always a good thing. Some respondents may feel that they need to give ‘good’ or ‘correct’ answers that may bias results. Alternatively, you may attract respondents who are in it just for the reward. One approach could be to run the survey with no incentive, with the option of offering one if responses are limited.
Designing the survey to assure respondents that a minimal time commitment is required and their privacy is assured can also help to increase responses.
Room for error
With all research there is a given amount of error to deal with. Bias may arise during surveys and focus groups, for example, interviewers leading the respondents. Or bias may be present in the design and wording of the questions themselves. There could be sample errors or respondent errors. Using the Internet to administer surveys removes the bias that may arise from an interviewer. However, with no interviewer to explain questions, there is potential for greater respondent error. This is why survey design is so important, and why it is crucial to test and run pilots of surveys before going live.
Respondent errors also arise when respondents become too familiar with the survey process. The general industry standard is to limit respondents to being interviewed once every six months.
Sample error is a fact of market research. Some people are just not interested, nor will they ever be interested, in taking part in research. Are these people fundamentally different from those who do? Is there a way of finding out? To some extent, web analytics, which track the behaviour of all visitors to your website, can be useful in determining this.
When conducting online research, it is crucial to understand who is in the target market, and what the best way to reach that target market is. Web surveys can exclude groups of people due to access or ability. It is vital to determine if is this is acceptable to the survey, and to use other means of capturing data if not.