In this revision presentation, some core concepts of marketing data analysis are introduced. The presentation introduces topics such as test marketing, calculation of moving averages, extrapolation, correlation and qualitative methods of marketing forecasts (hunch & delphi).
2. What marketing managers want to know
• How big is the market? (measured by sales, volume etc)
• How fast is the market growing and what is the growth
potential?
• The key social, economic, political/legal and technological
factors that drive change in the market
• Who are the existing competitors and their market shares?
• Extent of branding and customer loyalty in the market
• How the market is segmented to meet different customer needs
• Customer preferences in terms of when and where they buy,
what prices they pay and which methods of promotion are
effective
• The potential for developing a competitive position in a market
– either through a USP or through effective price competition
3. Examples of when marketing analysis is really needed
• Forecasting sales for new products or
investments into new markets
• Gathering evidence to support a finance-
raising exercise
• To support a new marketing strategy or
significant changes to the marketing
objectives
• To help make decisions in relation to
significant organisational or operational
change
4. Test marketing
Test marketing involves launching
Test marketing involves launching
the product in small part (usually
the product in small part (usually
geographic) part of the target
geographic) part of the target
market in order to gauge the
market in order to gauge the
viability of a product or service in
viability of a product or service in
the target market prior to a main
the target market prior to a main
roll-out or launch.
roll-out or launch.
5. Aim of test marketing
To gather as
To gather as Product
Product
much
much
information as
information as Price
Price
possible about
possible about Promotion
Promotion
the optimum
the optimum
marketing mix
marketing mix Place
Place
6. Benefits and drawbacks of test marketing
Advantages Disadvantages
Data provided is from actual customer Danger of the competition learning
spending about the product and coming up
with a response before the full launch
Reduces the risk of a full-scale launch Test market may not be
– if the product fails a test then representative of the full target
significant costs may be saved market, leading to inappropriate
decisions
Provides a way to tweak the Delays in full launch may limit the
marketing mix before full launch revenue opportunity in markets
subject to rapid change
Can create a promotional “buzz” Costly and time-consuming to
which supports the main launch administer
8. A Moving Average
A moving average takes a data
A moving average takes a data
series and “smoothes” the
series and “smoothes” the
fluctuations in data to show an
fluctuations in data to show an
average
average
The aim is to take out the extremes
The aim is to take out the extremes
of data from period to period
of data from period to period
9. Moving average illustrated
The red line shows
the quarterly
moving average.
This is calculated
by adding the
latest four
quarters of sales
(e.g. Q1 + Q2 + Q3
+ Q4) and then
dividing by four.
The blue line shows the actual quarterly sales
figure which varies quarter by quarter
10. Moving average to extrapolation
The moving average helps shows the growth trend (expressed as aapercentage
The moving average helps shows the growth trend (expressed as percentage
growth rate), and extrapolation can use this to predict the path of future sales.
growth rate), and extrapolation can use this to predict the path of future sales.
This could be done mathematically using aaspreadsheet. Alternatively, an
This could be done mathematically using spreadsheet. Alternatively, an
extrapolated trend can simply be drawn on the chart as aarough estimate, as
extrapolated trend can simply be drawn on the chart as rough estimate, as
shown below:
shown below:
11. Benefits / drawbacks of using extrapolation
Advantages Disadvantages
A simple method of Unreliable if there are
forecasting significant fluctuations in
historical data
Not much data required Assumes past trend will
continue into the future –
unlikely in many competitive
business environments
Quick and cheap Ignores qualitative factors (e.g.
changes in tastes & fashions)
13. Correlation variables
Independent
Independent Dependent
Dependent
Variable
Variable Variable
Variable
The factor that
The factor that The variable that
The variable that
causes the
causes the is influenced by
is influenced by
dependent
dependent the independent
the independent
variable to change
variable to change variable
variable
14. Plotting correlation - example
(number per week)
Customer Enquiries
Advertising per week
(£’000)
15. Explaining the scatter chart (1)
Correlation is
Correlation is
usually measured
usually measured
by using a scatter
by using a scatter
diagram, on which
diagram, on which
data points are
data points are
plotted.
plotted.
The dependent variable is normally
The dependent variable is normally
plotted on the y-axis: the independent
plotted on the y-axis: the independent
variable on the x-axis
variable on the x-axis
16. Explaining the scatter chart (2)
A “line of best fit”
A “line of best fit”
(the regression
(the regression
line) attempts to
line) attempts to
plot the
plot the
mathematical
mathematical
relationship
relationship
between the
between the
variables based
variables based
on the data
on the data
points.
points.
17. Types of correlation
Positive A positive relationship exists where as
correlation the independent variable increases in
value, so does the dependent variable
Negative A negative relationship exists where as
correlation the independent variable increases in
value, the dependent variable falls in
value
No correlation There is no discernible relationship
between the independent and
dependent variable
18. Positive correlation
the UK to Florida
Holidays taken from
Pound / $ Dollar
Exchange Rate
20. No correlation
and other savoury pastries
Demand for sausage rolls
Number of weddings per
year in the UK
21. Strong or weak correlation?
• The line of best fit indicates the strength of
the correlation
• Strong correlation means that there is little
room between the data points and the line
• Weak correlation means that the data
points are spread quite wide and far away
from the line of best fit
• If the data suggests strong correlation, then
the relationship might be used to make
marketing predictions
22. Qualitative forecasting – two approaches
Delphi
Hunch
Method
An educated
An educated Opinion from
Opinion from
guess
guess the experts
the experts
23. Hunch
• A forecast based on a hunch is likely to be
influenced by the experience of the
forecaster, perhaps supported by market
research or from discussions with others in
the market
• An experienced manager will have strong
insights into the sales prospects for
individual products, business units
• The starting point for a hunch forecast is
often the previous years’ or period data
24. Delphi
• Involves getting a group of market experts
to provide an opinion on the forecasting
task – e.g. to estimate future sales growth
in a market
• Experts first give a confidential individual
opinion on the task
• Their forecasts then revised based on the
submissions of each expert to the group
• Ultimately the aim of the Delphi method is
to reach a “consensus” forecast
25. When to use qualitative
forecasting
• When there is little accurate or predictable
historical data available (e.g. in the case of
a new product launched into a new market)
• Where a market is subject to wide
fluctuations in demand (e.g. unexpected
surges or shocks)
• Where a market experiences significant
change (e.g. shortened product life cycles,
rapid changes in technology)
26. Using IT to analyse markets
Almost every major
market is analysed
using IT
27. The analysts
• Businesses themselves
• Competitors
• Suppliers
• Trade associations
• Government
• Industry regulators
• Industry analysts
28. Example - retailing
• Individual retailers analyse sales using their
EPOS terminals and other systems
• The Office of National Statistics produces
regular data on total retail sales
• Specialist market researchers like TNS track
retail sales in great detail at the checkout
• The British Retail Consortium produces weekly
and monthly data for its members (the BRC is
the trade association for retailers)
29. Example – the media market
• Thousands of households track radio
usage for RAJAR which is used to
measure demand and market share
• Industry regulator OFCOM produces
highly detailed market research on
sales, cost and other market
information for all consumer media
markets