Consumer analytics is
the analysis of continuous individual/household level (customer level)
behavioural data to address business issues.
The terms aggregate
and disaggregate data provide for a broad categorization of the nature
of data. The data that Tesco was accustomed to seeing before the launch of
Clubcard was aggregate in nature. They knew how much of each item they sold
each week or each day. This is the data that is captured every time a bar code
is scanned at the checkout; what it does not reveal is who bought the item.
Disaggregate consumer data on the other hand is
consumer or household level data. It is captured for instance when the consumer’s
loyalty card (e.g., Clubcard) is also scanned at the checkout. The data is continuous
in nature — we have data for the same households/individuals
continuously over time.
The term consumer analytics has been
defined in a number of different ways. I would like to use it specifically in
the context of consumer level disaggregate data. Consumer panels, loyalty
panels, consumer/customer transaction, e-commerce sites, social networking sites, search engines, websites in general — all these sources yield disaggregate,
continuous data on the behaviour of individuals, customers or households. It is
the analysis of this type of consumer level disaggregate data that I am
classifying as consumer analytics.
In the past, the data pertained mainly to
consumers’ buying and consumption habits, and their tastes and preferences. Now
it increasingly also includes their browsing or interaction behaviour on the
net.
Interactions include the clicks, navigation paths
and browsing activities on websites. The field of web analytics, or the
analysis of behaviour of web users, falls largely within the scope of consumer
analytics. In addition to refining the elements of the marketing mix, the focus
of web analytics lies in improving the effectiveness of the website, in terms
of conversion rates and other performance parameters. This subject is covered
in brief in Chapter Digital Marketing.
While behaviour is the key characteristic of
consumer analytics data, it often is enriched by demographic, geographic,
psychographic and socialgraphic information.
The methods and techniques covered in Chapter Consumer Analytics and Consumer Panels, fall within the field of consumer analytics. The focus in this
chapter lies mainly on data management tools and technologies, machine learning
techniques, data mining, crowd sourcing and co-creation, optimization
techniques and visualization techniques. Big data and cognitive systems are
also covered, and so too some of the application areas.
Consumer analytics is not as recent a phenomenon
as it is popularly thought to be. Some companies at the forefront of consumer
analytics were founded in the 1980s and 1990s. The biggest change over the
years is not the science, but rather the technology, and the advent of big data.
Back in the 1990s, to run a resource intense
consumer panel analysis over about 35,000 homes in urban India, I would leave
one of my PCs on overnight, and it may still be running the next morning. Today
a similar analysis would take a second or two to run on my laptop.