Customer Analytics Case Study Report

DATA4700

Digital Marketing and

Competitive Advantage

Lesson 4

Customer Lifetime Value

1 Identify the components that make up customer

lifetime value (CLV)

2 Evaluate the impact of customer acquisition costs

(CAC) and customer retention costs (CRC) on CLV

3 Conduct RFM analysis to segment your customer

base

Lesson Learning Outcomes

Review of Lesson 3 1. Customer churn

– Decreases revenue

– Expensive to recruit new customers

– Existing customers spend more with you

2. Predicting customer churn with machine learning

– Decision tree to classify

– Logistic regression to assign a probability

3. Preventing customer churn

– Customers leave due to bad experiences, customer service, or product fit

– Customer loyalty keeps them engaged

4. Customer relationship management (CRM)

– “Holistic approach to managing customer relationships to create shareholder value”

– Workforce benefits for the company as well

Key framework #1: The 3 Cs – customers, company and competitors

Key framework #2: STP – segmentation, targeting, positioning

Key framework #3: The 4 Ps – product, promotion, place, price

Digital marketing portfolio: devices, platforms, media format and type

Customer lifetime value

• One of the key benefits of using a good CRM system

is that you can identify where the highest customer

lifetime value (CLV) might be coming from.

• Analysing CLV allows you to determine which of your

customers are the most profitable to you, over a long-

term horizon.

• This then permits you to prioritise these most valuable

customers in your digital marketing outreach efforts.

https://webengage.com/blog/how-to-increase-customer-lifetime-value/

Customer lifetime value

• To illustrate the underlying concept of CLV, here is a simple

example:

• What type of business could this be?

• What appear to be the main determinants of CLV?

https://www.tractionwise.com/en/magazine/customer-lifetime-value-calculation/

Customer lifetime value

• CLV is the net profit amount that we predict someone will

bring to your company while this person is your customer.

• If we are talking about net profit, we have to take costs into

consideration as well, since profit = revenue − costs.

• Therefore, you must also understand the customer

acquisition cost (CAC) of converting a prospect into a

paying customer.

Customer acquisition costs

Some examples of CAC include:

• cost of e-mail distribution lists and management

• cost of social media accounts and posts management

• cost of website management

• cost of online ads:

– pay per click (PPC) and search engine optimisation (SEO)

https://saltmarketing.ie/how-to-reduce-your-customer-acquisition-cost-and-optimise-your-marketing-strategy/

1. Besides the CAC examples listed on the previous slide,

what are some other types of CAC that a company incurs?

Once your company acquires a customer, there are then

costs associated with holding onto that customer and

preventing churn. These are called customer retention costs

(CRC).

2. What are some examples of CRC that you can think of?

Discuss in groups of two or three and be prepared to share

your insights with the rest of the class.

Breakout activity: Customer costs

Class discussion: Customer costs

Reminder: write in your worksheets!

1. Other types of CAC:

2. Examples of CRC:

9

Segmentation – a refresher

Segmentation is the act of using data to divide the market into

separate groups of people based on their differing characteristics.

These characteristics can be divided into four different groups:

• Demographic factors, such as age and gender

• Psychographic factors, such as attitude and lifestyle

• Behavioural factors, such as loyalty and price sensitivity

• Geographic factors, such as region and climate

RFM analysis

One method of segmenting your customer base according to

their CLV is to conduct an RFM analysis.

• RFM stands for:

– Recency

– Frequency

– Monetisation

https://techwave.net/blog/customer-lifetime-value/

RFM analysis

• Recency

– How much time has passed since a certain customer last

purchased from your company?

• Frequency

– How many times has this customer purchased from your

company within the past (x) weeks/months/years?

• Monetisation

– How much has this customer

spent with us?

– What are the transaction

amounts and the profit margin?

Segmenting using RFM analysis

• For each of the three categories, classify each customer as being

HIGH, MEDIUM or LOW for that category.

• For example, after looking at your customer database, you might

see that:

– 1/3 of your customers have purchased within the past month

– 1/3 of your customers last purchased 2–5 months ago

– 1/3 of your customers last purchased at least 6 months ago

• Therefore, each of the customers who purchased within the past

month would be classified as HIGH for recency.

• Customers in the middle group would be classified as MEDIUM,

and those who haven’t purchased in the past 6 months are LOW.

Segmenting using RFM analysis

• Take a similar approach with the other two categories.

• For example:

– 1/3 of your customers have purchased at least 3 times in the

past year

– 1/3 of your customers have purchased twice in the past year

– 1/3 of your customers have purchased at most once in the

past year

• Therefore, each of the customers who purchased at least 3 times

would be classified as HIGH for frequency.

• Customers in the middle group would be classified as MEDIUM,

and those who have purchased once or not at all in the past year

are LOW.

Segmenting using RFM analysis

• For example:

– 1/3 of your customers spent more than $500 with you in the

past year

– 1/3 of your customers spent $200 – $499 with you in the past

year

– 1/3 of your customers spent less than $200 with you in the past

year

• Therefore, each of the customers who spent at least $500 with you

in the past year would be classified as HIGH for monetisation.

• Customers in the middle group would be classified as MEDIUM,

and those who have spent less than $200 with you in the past year

are LOW.

Segmenting using RFM analysis

• This results in 27 (3 x 3 x 3) different possible ratings

combinations:

Segment Recency Frequency Monetisation

#1 HIGH HIGH HIGH

#2 HIGH HIGH MEDIUM

#3 HIGH HIGH LOW

#4 HIGH MEDIUM HIGH

#5 HIGH MEDIUM MEDIUM

#6 HIGH MEDIUM LOW

#7 HIGH LOW HIGH

… … … …

#27 LOW LOW LOW

Segmenting using RFM analysis

• Each customer can now be assigned to one of the 27

segments:

Customer ID Recency Frequency Monetisation Segment #

#5536641 HIGH MEDIUM LOW 6

#3149720 MEDIUM LOW MEDIUM 17

#8279026 LOW HIGH HIGH 19

#4387253 LOW MEDIUM HIGH 22

#9452337 HIGH LOW MEDIUM 8

#2910239 MEDIUM HIGH LOW 12

#1824912 LOW HIGH MEDIUM 20

#7695184 MEDIUM MEDIUM LOW 15

… … … … …

Segmenting using RFM analysis

• This allows you to look at all the customers within any one

segment and investigate whether there are any similarities

amongst most of the customers within that segment:

Customer

ID

R F M Segment # Age Postcode

#8279026 LOW HIGH HIGH 19 37 3000

#4183325 LOW HIGH HIGH 19 25 2471

#3694537 LOW HIGH HIGH 19 46 8125

#5305708 LOW HIGH HIGH 19 71 3141

#2816994 LOW HIGH HIGH 19 29 4003

#6427843 LOW HIGH HIGH 19 44 5045

#1938659 LOW HIGH HIGH 19 55 6121

… … … … … … …

Segmenting using RFM analysis

• Sometimes a helpful “nickname label” can be given to a segment.

• This segment consists of low recency, high frequency and high

monetisation customers.

• Collectively, they might be known as “lucrative churn risk”, for

example – customers who used to spend a lot, and quite often,

with your company, but who haven’t done so in a long while now.

Customer

ID

R F M Segment

#

Age Postcode

#8279026 LOW HIGH HIGH 19 37 3000

#4183325 LOW HIGH HIGH 19 25 2471

#3694537 LOW HIGH HIGH 19 46 8125

… … … … … … …

Breakout activity: Segmenting

using RFM analysis

1. Consider the low frequency, high recency, high monetisation segment

from the previous example. What types of digital marketing activities

would you do with them?

2. Choose one of the three segments below and create a “nickname”

label for that segment. What types of digital marketing activities would

you do with that segment?

Discuss in groups of two or three, and be prepared to

share your insights with the rest of the class.

Segment Recency Frequency Monetisation

#1 HIGH HIGH HIGH

#7 HIGH LOW HIGH

#15 MEDIUM MEDIUM LOW

Class discussion: Segmenting

using RFM analysis

Reminder: write in your worksheets!

1. Digital marketing activities to do with the “lucrative churn risk”

segment:

2. “Nickname label” and digital marketing activities for your chosen

segment:

Variations on RFM

There are a few variations of RFM that might be more appropriate to

certain business contexts, such as:

• RDM: Recency, Duration, Monetisation

– Duration measures how long a customer spends reading,

viewing or shopping on your website

• REM: Recency, Engagement, Monetisation

– Engagement measures how much interaction a customer has

with your website, such as videos watched, polls responded to

and number of different pages visited

• The process of dividing customers into HIGH, MEDIUM and LOW

categories works the same way for RDM and RFE as it does for

RFM

Is customer churn always bad? • Is it alright to lose certain customers?

• Professor Sunil Gupta of Harvard Business School

encourages digital marketing professionals to:

– focus attention on keeping the best customers

• How do you define “best customers”?

(You now have a framework to answer that)

– if you make a retention offer* to an existing customer:

• Will the customer respond positively to it?

• How much will this retention offer cause the customer

to spend with you?

• How much will the offer cost your company?

– If the retention offer isn’t worth it, you’re better off letting

the customer churn

*Examples of retention offers include: annual fee waivers for credit

cards, bonus points, airline miles or spending credit to renew a subscription

https://www.hbs.edu/faculty/Pages/profile.aspx?facId=261323

Dataset workshop: RFM and CLV

• Refer to the supplementary document “Week 4 data

workshop guide” in conjunction with these slides

Roadmap for this dataset workshop:

• Step 1:

Use one of the formulas below for calculating CAC and

CLV, depending on what data are available and the

appropriate context to your business problem:

CAC = (marketing+wages+infrastructure+other costs)/

(#customers acquired)

CLV = m/(1+i-r) or CLV = A*T*R*M

Dataset workshop: RFM and CLV

• Refer to the supplementary document “Week 4 data workshop

guide” in conjunction with these slides

• Step 2:

Install the software to be used (if you haven’t already done so)

• Step 3:

Work in groups of 3 or 4 students to use RFM analysis to

segment the customers in the dataset into 27 segments

Dataset workshop: RFM and CLV

• Refer to the supplementary document “Week 4 data

workshop guide” in conjunction with these slides

• Step 4:

Use the CAC and CLV formulas from Step 1 and the RFM

analysis in Step 3 to determine which segments are most

attractive to target

Dataset workshop: RFM and CLV

• Refer to the supplementary document “Week 4 data

workshop guide” in conjunction with these slides

• Step 5:

Think about what type of digital marketing activities you

would conduct towards the segment you would like to target

because of their high CLV

• Give your chosen segment a nickname label

• What types of segmentation factors (demographic,

psychographic, behavioural and geographic) have you used

in completing this exercise?

Summary of Lesson 4 1. Customer Lifetime Value (CLV)

– Net profit amount spent with your company while someone is your customer

– Must take into account customer acquisition cost (CAC) and customer retention

cost (CRC)

2. RFM (Recency, Frequency, Monetisation) analysis

– Segmentation approach to maximising CLV

– Churn is not always a bad thing for low CLV segments

28

Key framework #1: The 3 Cs – customers, company and competitors

Key framework #2: STP – segmentation, targeting, positioning

Key framework #3: The 4 Ps – product, promotion, place, price

Digital marketing portfolio: devices, platforms, media format and type

Customer churn and customer relationship management (CRM)