News
Tactics
Views
retailer that sells a wide variety of products. If it picture of future business trends. Having timely
runs a loyalty-card programme, it can capture and accurate forecasts can lead to better decisions
members’ transaction data and then use the based on sound predictions of the future.
data to model which consumers are likely to While spreadsheets provide a rudimentary
buy computers or home theatre equipment or understanding of these variables, they can’t
handheld devices. provide reliable answers from a multitude of
systems, let alone create predictive models and
2) the correct assessment of the importance deliver them to all pertinent users.
of each factor within those historical data. If managers can assemble accurate information
This involves using mathematical algorithms about the future, they can proactively structure
to assign weights to each factor that affects a workfl ows and allocate resources to maximise
particular outcome. The more variables that productivity and profi ts. The more precise the
infl uence an outcome, the more diffi cult it is projections, the greater the profi ts. Without
to decide which factors are more important predictive modelling, managers can only guess at
than others. Assess the relative weight of the what will happen based on past experience.
factors correctly, and you have a highly effective
modelling and scoring tool.
Dr Rado Kotorov is technical director
of strategic product management
3) the automation of data collection,
forecasting, and risk assessment. As you no
for Information Builders, a provider
doubt know, processing large amounts of data of business intelligence and web
and assessing the relative importance of each
reporting software.
By Dr Rado Kotorov
variable is not an easy task. It takes time and
requires special skills. Merchandisers and other
operational users don’t have the time, the
know-how, or the functional responsibility to
BI in brief
analyse data in such a manner. They need to
make decisions quickly to do their jobs. This is
where the latest software tools come into play.
Business intelligence solutions enable retailers to conduct the
following types of analysis:
Pulling it all together
customer profi tability—projecting the initial sales curve and lifetime value of
Because they are built upon highly specialised
a customer relationship
technologies, statistical software and predictive
modelling tools have historically been quite
channel usage and profi tability—assessing and predicting the most suitable
expensive. This is no longer the case. The rising
and effi cient channels for each contact activity and each customer
popularity of open-source software such as the
product preference and profi tability—assessing value and return on
R modelling language has created affordable
investment on a product basis across customer groups and channels
alternatives to commercial packages. A BI
environment that supports the R language lets bundling/cross-selling/upselling—identifying products that complement each
retailers create models and distribute them to other or will sell well together
end users as they would any other BI function—
customer loyalty and churn—monitoring which customers are loyal, which are
through a familiar interface that uses forms,
charts, maps, and dashboards.
likely to leave, when they are likely to leave, and what factors infl uence their
Businesses have invested in many types of
decisions
information systems, from supply chain and
demand forecasting—generating reliable estimates of short- , medium- , and
inventory management to transportation
long-term demand
management, planning, and merchandising.
But knowing how many transactions took place
market-basket analysis—assessing links and patterns in the choices
is not the same as understanding why they took customers make to maximise conversions on websites
place, what factors infl uenced the outcome, and
customer segmentation—dividing the customer base into groups that share
how to use that information in the future. It’s
no wonder that nearly 60 percent of retailers
common characteristics
surveyed by AMR Research last year said they
event-trigger analysis—discovering correlations between events, such as
wished to expand, improve, or replace their
demographic changes or holidays
existing BI capabilities to obtain centralised
information for timely decision-making.
marketing optimisation—incorporating information about customers, offers,
An integrated BI platform extracts
and channels to devise campaigns that pull well
information from multiple sources and
buyer preference analysis—conducting thorough analysis of market baskets,
transforms it into forward-looking insights
shopping patterns, and lifecycle purchase histories
that can increase revenue, profi tability, and
effectiveness. When used within a predictive
demand planning—using detailed planning and forecasting scenarios to
modelling environment, that information can anticipate demand across stores and channels for each item sold.—RK
help managers formulate a comprehensive
ceb175.indd 23 7/11/09 13:10:13
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45