an empirical analysis of algorithmic pricing on amazon
play

An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace - PowerPoint PPT Presentation

An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace Le Chen, Alan Mislove, Christo Wilson Northeastern University The rise of e-commerce E-commerce Sales over time (US Commerce Department) 400B 300B Sales in Dollar 200B 100B


  1. Key questions • Can we develop a methodology to detect sellers who are adopting algorithmic pricing? - Limited information, such as, price, rank, etc. 17

  2. Key questions • Can we develop a methodology to detect sellers who are adopting algorithmic pricing? - Limited information, such as, price, rank, etc. - Non-trivial because there is no ground truth. Therefore, we locate sellers behaving like “bots” 17

  3. Key questions • Can we develop a methodology to detect sellers who are adopting algorithmic pricing? - Limited information, such as, price, rank, etc. - Non-trivial because there is no ground truth. Therefore, we locate sellers behaving like “bots” • What are the different behaviors between the algo sellers and non-algo sellers? 17

  4. Algorithmic pricing detection 18

  5. Algorithmic pricing detection • How can algo sellers stay competitive? 18

  6. Algorithmic pricing detection • How can algo sellers stay competitive? - Matching to a “competitive” price 18

  7. Algorithmic pricing detection • How can algo sellers stay competitive? - Matching to a “competitive” price • Target price time series 18

  8. Algorithmic pricing detection • How can algo sellers stay competitive? - Matching to a “competitive” price • Target price time series - Lowest price 18

  9. Algorithmic pricing detection • How can algo sellers stay competitive? - Matching to a “competitive” price • Target price time series - Lowest price - Second lowest price 18

  10. Algorithmic pricing detection • How can algo sellers stay competitive? - Matching to a “competitive” price • Target price time series - Lowest price - Second lowest price - Amazon price 18

  11. Algorithmic pricing detection $1.4 Lowest Price Algo Seller Price Non-algo Seller Price $1.3 Product Price $1.2 $1.1 $1 t1 t2 t3 t4 t5 Timeline 19

  12. Algorithmic pricing detection $1.4 Lowest Price Algo Seller Price Non-algo Seller Price $1.3 Product Price $1.2 $1.1 $1 t1 t2 t3 t4 t5 Timeline 19

  13. Algorithmic pricing detection $1.4 Lowest Price Algo Seller Price Non-algo Seller Price $1.3 Product Price $1.2 $1.1 $1 t1 t2 t3 t4 t5 Timeline 19

  14. Algorithmic pricing detection $1.4 Lowest Price Algo Seller Price Non-algo Seller Price $1.3 Product Price $1.2 $1.1 $1 t1 t2 t3 t4 t5 Timeline 19

  15. Algorithmic pricing detection 20

  16. Algorithmic pricing detection • Detection criteria - Spearman’s rank correlation between seller and target price ‣ Product must have more than 1 seller - Correlation coefficient >= 0.7 - p-value <= 0.05 - Number of price changes >= 20 20

  17. Algorithmic pricing detection • Detection criteria - Spearman’s rank correlation between seller and target price ‣ Product must have more than 1 seller - Correlation coefficient >= 0.7 - p-value <= 0.05 - Number of price changes >= 20 • Discovered 543 sellers covering 513 products - 2.4% of total sellers - 31% of the bestselling products 20

  18. Algorithmic pricing detection 21

  19. Algorithmic pricing detection $20 $18 Price $16 $14 $12 Algo Seller Seller1 Seller 2 $10 11/22 11/24 11/26 11/28 11/30 12/02 12/04 12/06 12/08 12/10 Timeline (Year 2014) 21

  20. Algorithmic pricing detection 22

  21. Algorithmic pricing detection $10 Algo Amazon Seller 2 Seller 1 Seller 3 $9 Price $8 $7 10/30 10/31 11/01 11/02 11/03 11/04 11/05 11/06 11/07 11/08 11/09 Timeline (Year 2014) 22

  22. Behaviors of algo-sellers 23

  23. Behaviors of algo-sellers • Number of products sold by algo/non-algo sellers 23

  24. Behaviors of algo-sellers • Number of products sold by algo/non-algo sellers 100 Algo sellers Non-algo sellers 80 60 CDF 40 20 0 0 100 1000 10000 Number of products 23

  25. Behaviors of algo-sellers • Number of products sold by algo/non-algo sellers 100 Algo sellers Non-algo sellers 80 60 CDF 40 20 0 0 100 1000 10000 Number of products 23

  26. Behaviors of algo-sellers 24

  27. Behaviors of algo-sellers • Feedbacks received for algo/non-algo sellers 24

  28. Behaviors of algo-sellers • Feedbacks received for algo/non-algo sellers 100 Algo seller Non-algo seller 80 60 CDF 40 20 0 1 100 10000 1000000 Number of feedbacks 24

  29. Behaviors of algo-sellers • Feedbacks received for algo/non-algo sellers 100 Algo seller Non-algo seller 80 60 CDF 40 20 0 1 100 10000 1000000 Number of feedbacks 24

  30. Feedback loops 25

  31. Feedback loops • Algo sellers sell less types of products but receive much more feedbacks than non-algo sellers 25

  32. Feedback loops • Algo sellers sell less types of products but receive much more feedbacks than non-algo sellers • Recall the feature weights 0.4 Weight 0.2 0 P P F # A I s e r r v F i i F e c c g e B d e e e r b A a D R d a t ? b a i i c n f f a t k g i o c k Feature 25

  33. Feedback loops • Algo sellers sell less types of products but receive much more feedbacks than non-algo sellers • Recall the feature weights 0.4 Weight 0.2 0 P P F # A I s e r r v F i i F e c c g e B d e e e r b A a D R d a t ? b a i i c n f f a t k g i o c k Feature 25

Recommend


More recommend