Market Research and Consumer Intelligence are both very technical fields, with high requirements for specificity. As such, they can be minefields of unfamiliar vocabulary and acronyms. We’ve created a list of commonly-used terms in these areas, and provided you with simple definitions of each to allow you to negotiate this complex field with ease.
A method of comparing two versions of a newsletter, or webpage, or advert in order to see which performs better. All variables apart from the one you’re testing must be kept the same.
A prejudice towards a certain outcome that will unfairly influence the results of a survey.
The value that your brand adds to your company by being well-known and well-liked.
A ranked list of the top-performing brands in any given industry based on consumers’ unprompted recall of the brand; how much they like the brand; and how strongly they would recommend the brand.
A data-point that incorporates how likely a consumer would be to purchase that brand, and how likely they would be to recommend it.
A tool allowing companies to monitor the strength and perception of their brand over time.
Confidence Interval and Margin of ErrorConfidence Interval is the level of certainty you have in a result. For instance, if you have a confidence level of 95%, and your survey tells you 90% of people would choose option A over option B, you can be confident that in the wider population 90% of people would choose A, 95% of the time. I.e. if you sent a survey 100 times, 90% of people would choose A in 95 of the 100 surveys. Confidence Interval is always used in combination with Margin of Error.
Margin of Error is the likelihood that this result is accurate. For instance, if you have a margin of error of 3, and your survey tells you 90% of people would choose option A over option B, you can be confident that in the wider population 87-93% of people would choose A.
Margin of Error and Confidence Interval both need to be included in a sample size calculation as both indicate how applicable your results are to the wider population. In our example, if you have a confidence interval of 95% and a margin of error of 3%, and your survey tells you 90% of people would choose option A over option B, then if you sent the survey 100 times, 95 of those times the results would come in that 87-93% of people choose option A.
Information about trends in human behaviour and consumer habits. It can be within your industry as a whole, or can be more specifically about how consumers engage with your business.
The practice of doing tests on different audience groups. So if a respondent participates in test 1, they are not allowed to participate in test 2.
Free Text Response
When the respondent can answer the question with their own words. No restrictions are placed on what they can write. These type of answers are great for allowing respondents to expand on their views (providing the ‘why’ and additional context). They’re also useful for open ended suggestions and brainstorming with smaller sample sizes.
The level of detail that you go down to in a set of data. For example: granularity in regions would be whether you can separate data out into country-level, or county-level, or city-level, or postcode level. The more granular research can be, the more targeted your research becomes.
Grid Questions (also called Matrix Questions)
Questions where you can ask about 2 variables at once. Options will be displayed horizontally and vertically, and respondents check all options that apply to them. For example: ‘Which social media channels do you use and what do you use them for?’ (Facebook, Snapchat, Twitter, Instagram would be listed across the top, with ‘Talking to friends’, ‘Sharing photos’, ‘News updates’, and ‘Following influencers’ down the side.
Incidence Rate (IR)
The percentage of respondents qualified to complete a particular survey, versus the total population. For example: if a survey is about the taste of different coffee brands in the UK, the Incidence Rate will be the percentage of Britons who drink coffee.
A question that is framed in such a way that it encourages respondents to give a particular answer. For example: ‘How uncomfortable do you find outdoor furniture?’ would be better worded as ‘How would you describe your comfort level when using outdoor furniture?’ since the first question presupposes that respondents find it uncomfortable in the first place. Leading questions should be avoided in surveys to avoid unfairly influencing the results.
Length of Interview (LOI)
The estimated time it will take a respondent to complete a survey. It can be useful to give respondents the LOI before they begin, so that they do not drop out mid-way through.
A gradual scale measuring the strength of a consumer’s attitudes to a statement. For example: Always; Frequently; Sometimes; Occasionally; Never.
The percentage of the market’s total revenue being earned by one particular company.
The process of gathering information about consumer behaviour (their opinions, reasons, motivations and preferences). It is often carried out to assess the viability of a new product, or to track how a current product is perceived and used by its consumers.
A message that appears between questions in a survey to tell the respondent something they will need to know in order to be able to answer the next question, or set up the context. For example: ‘Imagine you’ve just arrived at a major supermarket.’
When respondents are asked about one specific product or concept so that it can be evaluated on its own. For example: a coffee company would show a sample of respondents one potential new package idea for their coffee. Respondents would then discuss the merits and drawbacks of this one idea, removed from any other products or concepts.
Multiple Choice Answer
The respondent is given a list of possible options to answer from, and they may choose several. For example: How many coffee brands have you heard of? Starbucks, Nero, Costa, Puccino’s.
National Representation (NatRep)
A sample that proportionally replicates the demographics of a certain country in terms of gender and geographical distribution. For example, in the UK, the population is made up of 51% women and 49% men.
Net Promoter Score (NPS)
A measure of how likely a consumer is to recommend your brand. Respondents are asked to rate how likely they would be to recommend a brand from 0 to 10. The percentage difference is calculated between the negative scores (any responses of 0-6) and the positive scores (responses of 9 or 10) to give the overall NPS.
A group of people selected to take part in market research over a period of time. Panels are usually decided based on demographics or specific criteria, such as attitudes and behaviours.
The size of a group of people who qualify into a survey. For example in a survey about which way people are likely to vote in an upcoming UK general election, the population size is the number of people above voting age in the UK.
The tendency projects have to increase in scope as they proceed. For example: an initial market research project may illuminate trends which the company had not previously thought of, and subsequently want to expand their market research to explore further. Scope creep can be avoided by properly defining the terms of what you are investigating.
Prompted Brand Recall
When consumers can identify a brand if given a prompt: For example: if consumers can correctly say a brand’s name when shown the brand’s colours or logo.
Purchase Intent Score (PIS)
This measures how likely a respondent is to purchase from a certain brand on a gradual scale ranging from ‘Very Likely’ to ‘Very Unlikely’. It is calculated by dividing the number of ‘Very Likely’ responses by the total number of respondents.
A question designed to qualify respondents in or out of a survey based on a certain behaviour, rather than a demographic. For example: a company interested in market research about coffee, may ask ‘How regularly do you drink coffee?’ and then only include ‘Daily’ respondents in the rest of the survey.
Data presented in words (a respondent’s opinions, reasons and motivations; unprompted recall of brands etc.)
Data presented in numerical terms (integers, percentages, times, ratings etc.)
Proportions of the total sample size to be met by individual demographics. E.g. if you set a quota of 20% for people living in London, then only 20% of total respondents who take the survey will live in London.
The respondent is given a list of options, and they must drag and drop them into a ranked order. For example: ‘Rank these coffee brands in order from your favourite (at the top) to your least favourite (at the bottom)’.
The ability to ask follow-up questions to a research group after the research is finished.
An individual answering a survey.
Also known as Completion Rate or Return Rate, this is the number of people who completed the survey divided by the total number of people who the survey was given to.
Routing / Piping
This is when a certain response will lead a respondent down an alternative path within the survey.
A group of people chosen to answer a survey based on their demographic. For example: 250 women aged between 35 and 40.
The number of respondents answering any given survey. The technical calculation takes into consideration confidence level, margin of error, population size, and incidence rate.
The process of eliminating unsuitable respondents from taking your survey. For example: if you are asking about voting habits, you might ask ‘Have you voted in a general elections?’ to screen out any respondents who will not have the relevant information to proceed in the survey.
Single Choice Answer
The respondent is given a list of possible options to answer from, and they may only choose one. For example: ‘Are you a coffee-drinker?’ Yes or No.
A result is statistically significant if it is not attributable to chance. For instance, if a survey indicated that the population prefer coffee to tea and the result was deemed to be statistically significant then you could be confident that this wasn't the result of chance, but a true representation of the population.
A set of questions given to a sample of respondents in order to gain data about consumer behaviour.
Research conducted (and funded) by an independent company, rather than by an individual brand. The results are often presented as reports or presentations, and made available for public use.
Unprompted Brand Recall
The capacity a brand has to be spontaneously remembered by a respondent when asked to name one brand in a given industry. For example: ‘Thinking of coffee, which brand comes to mind first?’ Unprompted Brand Recall is calculated by the number of times your brand is mentioned divided by the total number of respondents asked. Unprompted Brand Recall is sometimes referred to as ‘share of mind’ as it asks how much of your consumers’ minds your brand is occupying.
‘7 minute magic’
The amount of time before respondents engagements diminishes. Surveys should take no longer than 7 minutes to complete in order to ensure high quality results. This typically equates to 22 questions, the limit we set on our surveys to ensure high engagement, completion rates and quality.
We hope you found this useful! If you’d like to start the journey to becoming more informed about your own consumer base, get in touch!
We can connect you directly to an global pool of respondents to answer all your burning questions about your brand to help you maximise your potential. We can create completely customised brand trackers to help you understand exactly how customers are using your products, and what they think of every part of their experience with your brand. Alternatively we can help you go further still, with our Brand Intelligence products that offers greater levels of understanding around market dynamics, consumer sentiment, competitive intelligence and more.