Data Presentation & Method

Definition

Data sources are exploding, with real time access to data now a real possibility. There are also many new metrics to monitor adding greater complexity to the world of marketing metrics.

The job of a researcher is to: discover – find meaning in the data to answer our (and clients’) questions; and disseminate – communicate findings simply and efficiently; increasing the chance that research will be correctly understood and used

If someone has to work as hard as the researcher themselves to understand the findings, that researcher’s job has not been done well.

Key Findings

Descriptive models seek to uncover marketing phenomena and to represent them. This is the classical task of science.
Decision models (market mix models) that do not incorporate any descriptive knowledge seem to have no track record of established success.
In contrast, good descriptive models have long provided benchmarks and have helped us slowly to learn about causes and their effects.
The term statistically significant is widely misused and vastly overused: the ‘significant’ result is not necessarily interesting, important, large, meaningful, causal, or predictable in the future. It is NOT a badge of scientific rigour. It just says the observation “probably happened” in that population.
Significant does not mean important, nor large
Significant does not imply causation
Significant does not mean generalisable
Reporting significance often takes the place of reporting what the result actually means
Graphs can be extremely good at showing up a simple qualitative pattern, such as a curve or linear relationship. But they usually fail in communicating numbers or quantities.
A graph seldom conveys any tangible information unless it has an explicit storyline. Such a storyline will be much more memorable if it is also stated in words, eg saying ‘A is bigger thanB’,rather than just showing it graphically. Yet few graphical presentations do that.
A free choice pick-any approach is quicker to administer than rating scales or ranking measures and also provides a wider range of brand perceptions.
Always provide consumers with a list of brands. Failing to do so makes it difficult to detect perceptions for smaller share brands or non-users.
How recently and how much someone uses a brand affects their chance of giving brand perceptions.
Larger share brands get more perceptions for (pretty much) any attribute than do smaller share brands.
The relationship between the brand perceptions given by users and non-users is predictable.
Extremely positive overall evaluations are less likely to stimulate responses from non-users.
Attributes representing functional qualities are more likely to stimulate responses from non-users.
Negative attributes are equally likely to generate responses from users and non-users.
Big brands get higher scores for negative attributes than do smaller brands.
Responses to negative attributes come mainly from former users, this is why non-users (as a group) have a higher propensity to respond.
The Golden Rule of Forecasting applies to all forecasting problems. Firms that follow the Golden Rule of Forecasting gain a competitive advantage from more accurate forecasts, and can provide better service to their customers at lower cost.
The short form of the Golden Rule of Forecasting is: be conservative. Conservative forecasting requires consistency with cumulative knowledge about the situation, and with findings from the decades of research on forecasting methods.
The Golden Rule of Forecasting rejects methods that fail to make proper use of cumulative knowledge, such as the complex statistical techniques associated with the terms big data, analytics, data mining, stepwise regression, and neural networks.

Best Practice

Undertake our online courses into:
» Data Reduction and Panel Data
Analysts should use the checklist of Golden Rule guidelines to help them to improve their forecasting to identify dubious forecasts quickly and inexpensively, and to make better decisions, especially when the situation is uncertain and complex.
Be careful about mixing positive overall evaluations, attributes representing functional qualities, or negative attributes in multivariate analysis along with ‘normal’ attributes.
When interpreting a score for a negative attribute, one must take into account market share, rather than just interpret the raw percentage.
Use a pick-any approach to measuring brand perceptions
Prompt for brands, in any market
Check for differences/changes in usage weight and recency amongst users
With non-users, distinguish between lapsed customers and those who have never tried
Both positive and negative attributes need to be interpreted in the context of market share to determine if a score is good, bad or as expected for a brand
In attribute lists, identify and analyse separately extremely positive evaluations, representations of functional qualities and negative attributes.
Set realistic, not romantic, marketing and business goals: Growth is rare and the exception rather than universal; Differentiation is ephemeral rather than sustainable; Advertising that’s publicising rather than persuasive; Profits are satisfying rather than maximising; Knowledge that is accumulating rather than instantaneous and perfect.
In report writing, six rules can help:
1. Start at the End: Give the main results and conclusions first.
2. Revise: Your first drafts won’t be good enough.
3. Signpost: Tell readers what is coming, and help them to see where they’ve got to.
4. Cut Down on Long Words. And on long sentences.
5. Be Brief: Leave things out – sections, sentences, words.
6. Think of your Readers: Their problems are different from yours.
In making data user-friendly, turning data into information, five simple rules can help:
1. Order the rows and/or columns by some measure of size.
2. Round to two effective digits.
3. Give averages as a visual focus.
4. Use table layout to guide the eye.
5. Give a brief verbal summary.

Reports (15)

Reports

Showing 15 results

Many consumers give a different customer satisfaction score when re-interviewed.

Report 109
May 2022

We find that many consumers give a different satisfaction score when re-interviewed, even though the overall satisfaction score for the brand stays the same. And, this effect is seen among people who did not make a purchase from the service provider between surveys, so it is not due to an intervening service experience changing their satisfaction level. The overall proportions of people in our study who said they were satisfied, neither satisfied nor dissatisfied, or dissatisfied stayed the same between two surveys, so the overall brand score was stable. But the proportion of people within each group that gave the same score in both surveys was not 100%: 83%, 56%, and 43% for satisfied, neither satisfied nor dissatisfied, or dissatisfied respectively. There is more stability for the more popular response of ‘satisfied’, and less stability for the less popular initial response of ‘dissatisfied’. We conclude satisfaction scores behave quite like brand attributes: stable in aggregate but unstable at an individual level. Moreover, there is more over-time stability for satisfaction survey responses that received higher initial agreement to begin with, just like brand attributes.

This report gives three marketing implications that arise from the findings: one on the evaluation of intervention efforts, the second relating to customer word of mouth, and thirdly, a more general out-take on marketing metrics.

REPORT 104
August 2021

Net Promoter Scores bounce up and down from survey to survey. This spurious variation is likely to force managers to come up with spurious explanations.
We advise against NPS tracking, but if you must do it, use the mean average willingness to recommend scores instead of ‘Net’ scores.

Report 87
May 2018

A long standing empirical generalisation is that brand users are 2-3 times more likely to notice advertising for the brand they use than are non-users of the brand, but who are category users. The generalisation was observed in advertising awareness measures where the brand name was either part of the cue or retrieval process. We wondered if the presence of the brand name in the question was creating the user bias artefact. This had us worried that the empirical finding might be an artefact of the measurement approach and using another advertising awareness measure would give us different results. So we extended our testing to other types of advertising awareness approaches where the brand name is not involved, such as execution-prompted or media-prompted. We find the empirical generalisation still holds, confirming that user bias is a real phenomenon and not an artefact of the measure used.

REPORT 69
May 2015

This report presents the 'Golden Rule' of forecasting: be conservative. Firms that follow the 'Golden Rule' gain a competitive advantage from more accurate forecasts, and can provide a better service to their customers at a lower cost.

REPORT 53
July 2010

The follow up to Report 51: who do you really compete with? All marketers can use DoP analysis to profile the opportunities/threats in their market.

REPORT 41
May 2007

This 3 page report is a useful reference for evaluating and designing your brand tracking and for interpreting the results.

REPORT 27
March 2002

This report outlines Data Reduction as a simple alternative approach to Correspondence Analysis for analysing brand image data.

REPORT 25
September 2001

Which research for which problem? This report compares the merits of two main types of marketing models.

REPORT 24
May 2001

This report explicates five ‘romantic’ marketing goals – and their realistic alternatives.

REPORT 23
August 1999

Statistically signiﬁcant is often misused overused: a 'significant' result is not necessarily interesting, important, large, meaningful, causal, or predictable.