On the Existence of eLoyalty Networks in Online Auctions and Their Structure

1
On the Existence of e-Loyalty Networks in Online
Auctions and Their Structure
Inbal Yahav, Wolfgang Jank R.H. Smith School of
Business University of Maryland
2
Data
300 Sellers 2000 Repeating Buyers
3
Motivation
Sellers
Bidders
NO
Actors

• High profit
• High conversion rate

• Get the product

• Get the product

• Get the product (quality)

Objective

• Low price?

• Auction design (e.g., open price, duration, etc.)

Means

• Trust

• Feedback score

Lit
We introduce the notion of e-loyalty
IS THAT ENOUGH??
4
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?
5
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?
6
E-loyalty Network

• Bipartite graph with
• First nodes set sellers (red)
• Second node set buyers (white)
• Arcs purchases, with the width corresponding to
  the number of interactions

7
Measuring E-loyalty
19
19
12
12
15
15
12
12
15
15

• Normalize arcs weight by buyers experience

9
9
10
10
10
12
10
12
17
17
12191515101712 10912131
30
30
20
Sellers
100
Buyers
8
Measuring E-loyalty
0.14
0.1
0.11
0.1
0.11

• Normalize arcs weight by buyers experience

0.05
0.08
0.08
0.13
0.1
12191515101712 10912131
30
0.38
30
0.38
20
0.24
Sellers
1
100
Buyers
9
Measuring E-loyalty
0.14
0.1
0.11
0.1
0.11

• Derive loyalty curves

0.05
0.08
0.08
0.13
0.1
0.38
0.38
0.24
Sellers
1
Buyers
I carry on with the illustration for sellers
perceived loyalty curves
10
Measuring E-loyalty
PCA
Input
m sellers
First PCA Scores (75 of the variation)
(discrete grid)
11
Initial Observations
Sellers perceived loyalty curves

• Most of the buyers are loyal

Buyers loyalty curves

• Few sellers dominate the market

• Low first PCA scores corresponds to high loyalty

12
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?
Buyer feedback
Seller feedback
Auction design
13
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?
Buyer feedback

Seller feedback
Auction design
14
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?
Buyer feedback
Seller feedback
Auction design
15
Where to Look?
Feedback?
Buyer
Gender?
Year of Birth?
FOCUS
Feedback?
Seller
Number of Auctions (Volume)?
Auction Design?
16
Methodology

• Cluster sellers by
• Feedback score
• Volume
• Auction design (average)
• Start price
• Shipment costs
• Price

17
Methodology
18
Research Questions

• 1. How to define and measure e-loyalty?

2. What factors drive loyalty in online
auctions?
3. How does loyalty impact auction outcome
(price, conversion)?

19
Methodology
Average Effect
Most non-loyal
Effect Variance
20
Conclusions

• Most of the bidders are loyal
• Few sellers dominate the market
• Loyalty is derived by a tradeoff between trust
  and price sensitivity
• Unique product sellers have highest conversion
  rate and price, but with the highest variance
• For all other sellers, higher loyalty results in
  higher price and conversion rate, with lower
  uncertainty

21
More Information?
Inbal Yahav iyahav_at_rhsmith.umd.edu Slides and
eBay Data Collector http//www.rhsmith.umd.edu/fa
culty/phd/inbal/