Marketing and Retail Analytics
- Module 1
1
Marketing
• Activities undertaken by a company to promote the buying or selling of a product or
service.
• Includes advertising, selling, and delivering products to consumers or other
businesses.
Customer Purchase Funnel
4 Ps of Marketing
Management process through which goods and services move from concept to the
customer.
• identification, selection and development of a product
• selection of a distribution channel to reach the customer's place
• determination of its price
• development and implementation of a promotional strategy
B2B Marketing
• Involves targeting other businesses.
• Physical products that companies sell to other businesses are ‘industrial goods.’
• Management consultancy, training provision, IT services, provision of temporary
staff
• Ex., B2B Companies - Microsoft, IBM, Oracle, SAP, FedEx
B2C Marketing
• Targeting the individual people (Consumers) who purchase products and use the
services.
• Offerings - FMCGs such as food, beverages, or durable goods like cars, televisions,
refrigerators. etc.
• Ex., Netflix, Mc Donald, Apple, Coca Cola
B2B + B2C marketing
• integrated marketing approach
• The Coca-Cola company, for example, knows that its B2C marketing must
succeed. Put simply; it has to persuade supermarkets and smaller stores – B2B –
to provide shelf space.
Products vs Services
Products Services
physical and tangible. This implies that a product can be held, it intangible and can only be felt
can be seen, felt or smelled.
product can be returned to the seller for replacement or refund cannot be returned to the provider
in the event that it is wrong or damaged
can be owned by the purchaser since ownership is transferred cannot be separated from their providers
the moment a transaction has been performed
The billing process of a product is not continuous The billing process of a service is continuous
quality of products can be compared since these are physical Services vary according to who provide them, where, when and
features that can be held. how. The quality of the service depends on the service provider
Customer vs Consumer
• Customer is a person who buys the goods or commodity or services and pays the
price for it.
• Consumer is the one who consumes the goods, i.e. the user of the goods.
• Each and every marketing activity is directed towards influencing the behavior of
customers, i.e. to induce them in such a way that they take an action intended by
marketers. So, customers are regarded as the king of the business.
Examples of marketing campaign
• Vodafone zoo zoo/Pug
• Coca-Cola Small World Machines - Bringing India & Pakistan Together
• Maggi – 2 minutes noodles, nostalgia campaign
• Amul The Taste of India
Retailing
11
Retail
• Retail is the process of selling goods and services directly to customers
• A retailer is any individual who operates their business either through a brick and mortar store or over
an online ecommerce platform.
• A retail transaction is usually complete once the buyer pays for the goods or services received.
• Big retailers such as Amazon, Walmart, Tesco, and Sainsbury's display goods from manufacturers or
wholesalers.
• point of purchase could be a brick-and-mortar retail store, an Internet shopping website, a catalog, or
even a mobile phone.
Supply Chain
• A supply chain is a process the occurs between companies and suppliers in order
to distribute products to end users.
Image source: https://www.thebalancesmb.com/what-is-retail-2892238
Major Retailers
• Tata group
• Reliance
• Amazon
• Walmart
• Apple
• Mc Donald
E-tailing
• Transactions that are conducted over an electronic network, where the buyer and
merchant are not at the same physical location
• Ex., Amazon, Ebay, Flipkart
Marketing and Retail Analytics
16
Marketing Analytics - Objective
• Improve Marketing Effectiveness
• Understand what marketing channels perform better and optimize marketing ROI by allocating
available marketing spend across channels
• Predict future sales to minimize sales lost due to unavailability of products in the store
• Maximize customer value by enabling business to take right decision at each stage of customer life
cycle (Acquire -> Grow -> Retain -> Win-back/Re-Activate)
– Understanding Customer Preference
– Identify High valued customer
Marketing Analytics - Objective
• Identify opportunities for sales growth with in category
• Measure cannibalization impact of promoting one brand on the portfolio
• What promotion works better? What will be the ideal price gap with competition brand?
• Increase cross concept engagement of the loyal customers for multi format retailer
Customer Segmentation
• The fundamental question answered by Analytics
is “Who” are the right customers to target-to
acquire or increase average transaction value or
prevent attrition/churn or to bring back to store
• Building block of doing such analytics is
understanding of the customer by creating
subsets of Homogeneous groups by using several
segmentation techniques
What is Segmentation?
The process of dividing a market into distinct subsets (segments) of consumers with
common needs or characteristics. A market segment consists of a group of
customers who share a similar set of needs and wants.
Criteria for effective Segmentation
• Segment must have measurable size and purchasing power.
• Marketers must find a way to promote effectively to and serve the market segment.
• Segment must be sufficiently large to offer good profit potential.
• Firm must aim for segments that match its marketing capabilities.
Types of Segmentation
1. Objective Segmentation: Classify customers in to sub groups with an objective in mind.
Example, A retailer wants to understand the nature/characteristics of the customers
who shops only during Promotion/Sale period.
2. Subjective Segmentation: Classifying a group of heterogeneous customers in to
subgroups of homogeneous customers. And then profile/analyze (profile) the subsets
to understand the characteristics of the customers in each subgroup. Example –
Behavioral segmentation: Segment customers based on their transactions over a
period.
Types of Segmentation
Dimensions of Segmentation
– Geographic Segmentation
– Demographic Segmentation
– Psychographic Segmentation
– Behavioral Segmentation
Geographic Segmentation
• Division of an overall market into homogenous groups based on their locations.
• People living in the same area have similar needs and wants that differ from those living in other areas
• Climate: Choice of clothing in Bangalore (Pleasant & Low humidity) vs Chennai (Hot & Humid)
• Using Geographic Segmentation – Marketers focus on core regions, those from which they draw 40
to 80 percent of sales.
Demographic Segmentation
• Division of an overall market into homogenous groups based on variables such as gender,
age, marital status, income, occupation, education, household size, and stage in the
family life cycle; also called socioeconomic segmentation.
• Ex., Working women who regularly use the Internet make most of the decisions about
retail items, healthcare goods and services, and fitness products.
Psychographic Segmentation:
• Division of a population into groups that have similar psychological characteristics, values
and Lifestyles.
• Psychographics clearly provides a differentiated description of the customers compared to
demographics and behavioral traits.
• Can help marketers more effectively create goods and services for a target market.
• Generally acts as a good supplement to geographic and demographic segmenting.
Behavioral Segmentation:
• Segmenting population into groups based on transaction behavior, attitudes
toward or reactions to a product and to its promotional appeals.
• Behavioral segmentation is about understanding customers not just by who they
are, but by what they do, using insights derived from customers’ actions.
• Advantages of Behavioral Segmentation
– Personalization. Understand how different groups of customers should be targeted with different
offers, at the most appropriate times through their preferred channels, to effectively help them advance
towards successful outcomes in their journeys.
– Predictive. Use historical behavioral patterns to predict and influence future customer behaviors and
outcomes.
– Prioritization. Make smarter decisions on how to best allocate time, budget and resources by identifying
high-value customer segments and initiatives with the greatest potential business impact.
Expected Output
• Develop Meaningful Segments
• Segment Description / Segment Profiling
• Provide actionable insights
Example Clustering
• Customized Recommendation in Amazon/Flipkart/Netflix
• Loan offers by Banks
• Ola offers for first time users
• Diet Coke
RFM
30
RFM
• RFM uses transaction/purchase behavior data to segment a pool of customers. The
resulting customer segments can be ordered from most valuable to least valuable.
• Fundamental thoughts/assumptions behind RFM approach
– Customers who have purchased from you recently are more likely to buy from you again
than customers who you haven’t seen for a while.
– Customers who buy from you more often are more likely to buy again than customers
who buy infrequently.
– Customers who spend more are more likely to buy again than customers who spend less.
RFM
• RFM analysis is based on the marketing axiom that "80% of your business comes from 20%
of your customers."
Steps
1. The first step in building an RFM model is to assign Recency, Frequency and Monetary
values from 1(least) to 5 (highest) to each customer.
– Recency is simply the amount of time since the customer’s most recent transaction
(most businesses use days, though for others it might make sense to use months, weeks
or even hours instead).
– Frequency is the total number of transactions made by the customer (during a defined
period).
– Monetary is the total amount that the customer has spent across all transactions
(during a defined period).
Steps
2. Aggregate the 3 values for each customer
3. Rank the customers based on the final RFM score, and group them
4. Name the groups
5. Target the required customers
6. Tailored recommendations for each group
K-Means
35
Centroid based: K-Means Clustering
• K-Means is the most used clustering technique
• The ‘means’ in the K-means refers to averaging of the data; that is, finding the centroid.
• Based on Within cluster variance i.e Squared distance from the centre(centroid) of the
cluster
• Aims to partition the n observations into k clusters so as to minimize the within-cluster sum
of squares (i.e. variance).
• Have to pre-define K, the no of clusters
• K-Means uses Eucledian distances
• Each point is associated to a centroid and then centroids are re-calculated to minimize MSE.
3
6
K Means Algorithm
K Means Algorithm
Steps in K means clustering
1. The data set is a collection of features for each data point.
2. Data points will be clustered based on feature similarity.
3. K (no. of centroids/clusters) is given first (Elbow method).
4. Random centroid (randomly generated or randomly selected from the data set.) considered based
on the value of K
5. The centroids should be carefully placed, as far as possible from each other.
6. The algorithm works iteratively to assign each data point to one of K groups based on the features
that are provided.
Steps in K means clustering
7. 2 points are similar if they are close to the centroid. Euclidean distance measures the distance
between the data point and the centroid
8. Total distance computation is given by Kn
9. If data is close to the centroid, group them
10. Once all points are mapped to closest centroid, recompute centroid of the group. Sometimes, points
in 1 cluster may shift to another cluster
11. Repeat until there is no further movement of data points, or threshold for no. of iterations has
reached.
Elbow Method
• Choosing the Optimal K
• Graph between wcss and # of clusters
4
1
Thank you
42
