Indian Institute of Management Bangalore
Analytical Marketing
Case Analysis
ABB Electric Segmentation
Submitted To:
Professor Srinivas Prakhya
18th July 2024
Submitted By: Group No. 4
Aman Jaswani 2317017
Asmita Johri 2317021
Sarthak Mehta 2317019
Savaibhai G. Thakkar 2311119
Shaily Mehta 2311093
Shashank Baheti 2311136
Vishal Mohata 2311268
Context:
ABB Electric, founded in 1970 with funding from ASEA-ABB Sweden and RTE Corporation, aimed
to produce medium-sized power transformers for North America. By 1974, nearing breakeven,
it faced severe challenges: a 50% drop in industry sales, reducing its market share to 6%.
Competing against giants like General Electric and Westinghouse, ABB Electric's unfocused
salesforce and pressure to reduce costs and standardize products hindered its progress.
ABB Electric's sales were predominantly to IOUs, whose substantial inventories led to projected
sales declines of up to 80% annually for the next few years. This left the company with minimal
penetration among RECs, municipalities, and industrial firms, markets dominated by its major
competitors. President and CEO Daniel Elwing recognized that increasing market share was
crucial for growth.
ABB Electric implemented a marketing information system (MKIS) supported by a customer
survey to identify key product attributes and segment the market. The survey, targeting 7,000
decision-makers, achieved a 40% response rate, providing critical data for the MKIS. This
strategy aimed to understand customer needs better and position ABB Electric as a customer-
driven company.
Questions:
1. At present you have information on the Descriptor Data tab of the ABB Electric Data
(Customer Choice).xls spreadsheet about the location of customers (districts 1, 2 and 3)
and the sales potential of each account or prospect. Based on this information alone, to
what companies would you direct the new direct marketing program? Specify the
accounts and customer or prospect types.
Current Customers: ABB should focus on its current low-spending customers (less than
$500k), specifically the bottom seven customers (highlighted in Table 1) from its existing
customer base of 18.
1
Non-Customers: ABB should target customers in Districts 1 and 3. As per its current
strategy, ABB will focus on the Competitive segment and Switchable segment. In these
districts, ABB should try to persuade customers away from the firms which have market
share just above ABB. Though the annual purchase volume is highest in District 2, there are
already 2 firms competing for the top spot for the market share. ABB would have to
differentiate itself from both the firms in order to gain a significant market share in District
2.
Ann. Purchase Volume ($
Customer K) District Firm Chosen
44 10,997 2 A
11 1,722 3 A
37 767 2 A
42 752 1 A
21 749 3 A
57 736 1 A
58 700 3 A
54 660 1 A
3 643 2 A
50 584 3 A
69 528 1 A
13 466 2 A
61 462 3 A
41 444 1 A
45 415 1 A
87 395 2 A
24 322 1 A
38 182 2 A
Table 1: Low spending customer for Firm A
Total No. of Total Ann. Purchase Volume Average Purchase Volume
District Customers ($) ($)
1 31 23,359 754
2 32 53,639 1,676
3 25 25,761 1,030
Total 88 1,02,759 1,168
Table 2: Distribution across districts
2
Total Ann. Purchase
District A B C D Volume ($)
1 17% 27% 40% 16% 23,359
2 25% 7% 34% 34% 53,639
3 16% 56% 10% 18% 25,761
Table 3: Market share across districts
2. Use the choice modeling approach based on the responses provided by 88 firms from your
region. The data consists of the evaluation of ABB Electric and the three main competitors
on eight variables: (1) Price, (2) Energy losses, (3) Maintenance requirements, (4)
Warranty,
(5) Availability of spare parts, (6) Ease of installation, (7) Salesperson problem solving
support and (8) Perceived product quality. Perform a customer-loyalty–based
segmentation for your customers and prospects.
a. Which variables are the key drivers of choice in this market?
b. Based on your analyses, on which firms would you focus your efforts? Why?
Part (a):
To understand which variables are the key drivers of choice in this market, we have run
conditional logit model regression analysis keeping Edison as the baseline.
Regression equation with all variables:
logit(P) = 2.1807*Price + 2.6557*Energy Loss + 0.5935*Maintenance + 1.1407*Warranty -
0.1326*Spare Parts + 0.5200*Ease of Install + 2.0323*Prob Solver + 2.6392*Quality –
0.1238*Alternatives_ABB – 0.6712*Alternatives_GE - 0.6873*Alternatives_Westinghouse
3
The "brand equity" variables Intercept_ABB, Intercept_GE, and Intercept_Westinghouse
are based on the baseline brand, McGraw Edison, which has a base coefficient value of 0.
They stand for each brand's inherent worth (brand image coefficients).
When a brand has a significant and large positive coefficient for an Intercept term, it
indicates that, in the absence of any other factors, its choice probability will be higher than
that of the base brand, which is McGraw Edison. When the intercepts are not significant, it
indicates that there is no "brand-specific" additional effect, such as because of the brand
equity associated with that brand, and that the other factors in the model (price, energy
loss, etc.) already adequately explain the decision probabilities.
In the case of transformers, it is not unexpected that these Intercepts are not important
because customers' decisions are probably influenced by functional product features.
Choice probabilities are not much influenced by perceptions about maintenance and spare
parts.
Because the question is phrased, "rate this firm's price on a 1–10 scale, where 1= poor and
10=good," a high score indicates a low price, which explains why the price coefficient in this
case is positive.
It is true that there is a greater chance of choosing a brand with a higher price perception.
4
The following, in order of ranking, are the primary choice-drivers in this market, according
to the preceding table:
1. Energy Loss
2. Quality
3. Price
4. Problem Solver
5. Warranty
6. Ease of Install
Despite having a positive value estimate, "Maintenance" is not statistically significant.
Part(b):
Based on elasticity calculations, we should focus on Westinghouse for the following reasons:
High Sensitivity to Price and Warranty Changes:
The firm identified by the elasticity calculation shows the highest sensitivity to
changes in price and warranty. This means that small adjustments in these variables
can significantly impact customer choice, making it a strategic point of focus.
Maximizing Impact:
By concentrating efforts on the firm with the highest elasticity, the company can
achieve the greatest return on investment. This is because changes in price and
warranty for this firm will likely lead to a more substantial shift in customer
preferences compared to other firms.
Resource Allocation:
Focusing on the firm with the highest elasticity allows for more efficient allocation
of resources. Instead of spreading efforts thin across multiple firms, concentrating
on the most responsive firm ensures that marketing and operational changes are
more effective.
5
Competitive Advantage:
By understanding and leveraging the elasticity of the most responsive firm, the
company can gain a competitive edge. This strategic focus can help in capturing a
larger market share and improving overall profitability.
Data-Driven Decision Making:
The use of elasticity calculations provides a data-driven approach to decision-
making. This reduces the risk of subjective biases and ensures that strategies are
based on empirical evidence, leading to more reliable outcomes.
Enhanced Customer Satisfaction:
By focusing on the firm that customers are most responsive to, the company can
better meet customer needs and preferences. This can lead to higher customer
satisfaction and loyalty, further strengthening the firm's market position.
6
3. Assume that marketing efforts targeted at companies in the Loyal and Lost categories
result in no incremental gain. On the other hand, suppose that you could retain or win
half the companies in the Switchable and Competitive segments with this program. How
much improvement in sales productivity can you realize by applying this choice model to
the allocation of your efforts?
We used two ways to identify the groups that were switchable and competitive. Examining
the Expected and Observed options on the Estimation tab was the initial approach.
Customers who were anticipated to be ABB but selected a competitor, as well as customers
who were not predicted to be ABB yet selected ABB, were included to this customer focus
pool. The second approach involved focusing just on the probabilities. The ABB column
was sorted so that only probability over 30% and below 70% were displayed. We selected
this
7
because, in our opinion, consumers with probabilities above 70% were loyal to ABB, while
those with probabilities below 30% were lost.
We combined all the clients from the two distinct approaches, and then we examined the
yearly purchase value on the Descriptor Data Tab (refer to the table below). 15% of the
entire market share was made up of customers. Since the question assumed we could
attract or retain half of the customers, we summed up all the customers' yearly purchase
prices and took 50% of that worth. We can increase our sales productivity by $50,000 by
using this strategy.
Annual
Customer
Purchase
4 $ 562
7 $ 664
11 $ 1,722
13 $ 466
16 $ 894
26 $ 899
36 $ 511
37 $ 767
41 $ 444
50 $ 584
61 $ 462
84 $ 1,404
86 $ 480
Total $ 100,126
50% Total $ 50,063
4. What other recommendations would you offer to ABB Electric to improve its
segmentation marketing program?
Preventing the poaching of its "Loyal" sector ought to be ABB Electric's top priority. The
Estimation Sample Details page will retrieve pertinent information to describe the clients in
8
this market. Once more, ABB Electric would benefit from using the probability to separate
these consumers and concentrate on influencing their decisions to repurchase. This would
guarantee that the "loyal" customers eventually become lifelong supporters of ABB Group.
Customers in the "Competitive" category, who are somewhat more likely to purchase from
ABB than the next competitor, are another market worth examining. Once these clients
have moved on to the "Switchable" or "Lost" categories, it will be challenging to win them
back. Appropriate measures to enhance retention in this area would correspondingly raise
profitability.
5. Comment on the uses and limitations and possible extensions of this modelling approach.
When ABB is making decisions about sales and marketing, it should consider keeping track
of client information and then analyzing past trends. In terms of applications and
restrictions, it would be possible to lessen restrictions by including fields that show client
loyalty. Data on how businesses make decisions might be provided to remove another
restriction. Creating relationships with the departments that will be making these choices
will aid in increasing sales. Additionally, ABB's already-established competitive edge would
be strengthened and maintained.
According to the case, the sales force was not very concentrated and just completed the
tasks it believed were necessary at the time. The sales force requires an organized working
method so they can decide which aspects of the market they want to compete on, such as
quality or pricing, and then concentrate on improving the product on those fronts.
Furthermore, the business could look at focusing its marketing efforts on a particular
product instead.
