There are at least seven different research methodologies that seek to identify what the customer really values. These range from large sample quantitative techniques to in-depth qualitative techniques, or a combination of the two.
One thing is clear. While customers are good at saying what they like or dislike, they are much less accurate at identifying the issues that drive their purchasing decisions. They can tell you their "expectations" - what they rationally view as being important - but these are often quite different from their main buying motives, which tend to be more emotional than rational.
- Qualitative techniques
Qualitative techniques are classically used at the start of a program to understand what the potential buying motives might be. These issues can then be quantified during a second stage of the research.
The most common approaches are focus groups, where 8-10 people get together to discuss the buying environment in depth, or one-to-one/one-to-two interviews ("depths"/"paired depths"). Creative workshops can also be used.
Qualitative research often uses a class of techniques known as "projective techniques". These are designed to get around respondents' rational responses, and to explore their emotional ones. Common projective techniques include:
- collage, where people tear out pictures from magazines to describe a specific situation or brand
- photographs, where they are shown photographs of situations, events or people to help explore the issue
- card sorts, where respondents are given cards to sort into specific categories to identify groupings
- speech bubbles, where figures are shown with speech bubbles coming out of their mouths - what are they saying?
- complete the sentence, where the interviewer starts a sentence about a topic or brand, and asks respondents to finish it
- consequences, where respondents are asked to tell a story around the table
- word association, where respondents say the first thing that comes into their heads when they are given a specific impulse
- the brand as a person, where people are asked to describe or act out the brand as if it were a person
Some brand managers find that they have all the information they need at the qualitative stage. U.S. companies, however, invariably require a quantitative analysis. Even U.S. focus groups tend to be more quantitative than qualitative.
Most qualitative marketing research agencies can offer you these techniques.
- Core/key values - attribution theory
Core/key values works from the premise that, if left to their own devices, customers will talk most about the things that are most important to them.
This technique, therefore, works off as blank a sheet as possible, and the interviewer should only say enough to encourage the respondent to keep talking for 30-45 minutes in order to capture sufficient "cause-effect" statements.
A "cause-effect" statement is where a respondent says in one part of the interview "I love diamonds", and in another part "De Beers make the best diamonds".
Core/key values research relies upon rigorous content analysis of the interviews which are always recorded. Having "parsed" the interviews into "cause-effect" statements, the analysis identifies the topics wthat were discussed most often and what people said about those topics.
The final results will then show drivers of acceptance/rejection and choice in a quantitative way, although the sample sizes tend to be more qualitative than quantitative in nature. It will also show your and competitive brand performance against each driver.
Core/key values is offered by ESP research.
- Laddering
Laddering is a technique based on Personal Construct Theory that argues that everyone walks around with their own picture of the world, and thus their own understanding of how it works. This picture is made up of "constructs", or opinions. Some of these opinions are critical to the way we individually live our lives, others are more peripheral - we don't really care whether people agree with us or not.
Laddering works on the basis that people are more likely to talk about their peripheral opinions, than about their core ones (the ones they will really fight for). However, it is believed that there is a connection between the two types of opinion, and so a researcher can get the respondent to "ladder" from one to the other.
A typical laddering process will start with "triads" - what is similar between two of the brands that is different from the third? Having got a number of "base attributes", the interviewer will proceed to ask a series of "why?" questions:
Q. What do Brand A and Brand B have in common vs. Brand C
A. Brand A and Brand B are better quality
Q. Why is this important to you?
A. It eliminates re-dos
Q. Why is this important to you?
A. It makes me look good in front of my boss
Q. Why is this important to you?
A. I get promoted
Q. Why is this important to you?
A. I can afford more things - a better car, a better house, longer holidays
Q. Why is this important to you?
A. My family value me more
So, while this respondent may be talking about product quality, the underlying motivations are material wealth and status within the family.
The full laddering analysis will map the minds of customers today and show the most important drivers of customer behaviour. It will also map your and competitive brand performance in the perceptions of your customers.
The main exponent of laddering is Withlin Research.
- Conjoint analysis
Conjoint analysis is a quantitative technique that seeks to identify statistically the "utility value" of each element of the customer decision making process.
The way this works is to offer customers choices between two or more offerings, and to see which offerings they choose. Each offering will be broken down into several attributes:
Q. When travelling on business, which of these would you be more likely to choose?
Offer A
- Suite with bed a 2 sofas
- Sea view
- Triple bed
- Bath & shower
- $250 a night
Offer B
- Double-room
- Sea view
- Double bed
- Bath & shower
- $100 a night
By varying the attributes across the different offerings, the analysis can identify which attributes are driving the purchase decision.
A new development of this process is to use computer software to allow people to add or extract attributes from an offering, usually where they can see the price implications of their decisions.
Conjoint analysis is the preferred technique for those with the "hard science" approach to market research.
Conjoint analysis is now widely available through quantitative marketing research agencies.
- Decision tree analysis
Decision tree analyses work with standard quantitative research to identify which attributes coincide with the outcomes you require - typically purchase/retention or loyalty.
The techniques underlying this analysis vary (CHAID and Partial Least Squares are popular ones), but the basic concept is the same - which areas do you need to improve to optimise the outcomes you require?
Decision tree analysis is now widely available through quantitative marketing research agencies.
- Correlation-based expectancy-value models
The "expectancy-value model" is a technical term for a process that identifies the importance of an issue and then asks customers to score each brand against each issue. By weighting the brand performance scores by the importance of the issue and adding up the results, these types of model are often highly predictive of eventual market performance.
The importance in these models is derived globally (i.e. not by individual) by correlating attribute scores for a brand with an overall assessment of the brand. This approach recognises that the decision making process may be different for each brand.
Almost any marketing research agency can offer correlation-based value analysis models.
- Stated importance
The simplest way of trying to find out how important an issue is to customers is to ask them directly. Unfortunately, this is not a very reliable method. Customers tend to over-emphasise rational reasons and to under-estimate emotional ones, whereas emotional drivers are often more influential in the actual decision making process than rational ones, even in a business-to-business environment.
However, in key account work where there can only be a small sample size, stated importance may be the only quantitative option.
Knowing what customers think is important to them is still useful, because you can play back to them what they want to hear, even though you know you have to address other issues to gain/retain their business.
The following member of Mud Valley would like to bring their services to your attention:
We can help you in two ways - we have a mass of smart strategic brand marketing tools, processes and workshop techniques for you to use, and a mass of smart brand marketing agencies as members across the world with niche knowledge and experience to support you thereafter.
