Introduction

NBY (Nike By You) is a inline and online service to allow Nike members customize their Nike products. The inline store services include Grab & Go and Essentials. For Grab & Go services, Nike members can select items among Laces, Pins, Aglets and create carvings as they want. NBY Essentials offers athlete-led, quick and easy personalization packages, across specific footwear and apparel blanks. For the online NBY services, Nike members can design their shoes based on colors, materials, logo, and so on, which provides an easy way to help Nike members personalize their shoes. The purpose of this study is to help teams explore the characteristics of the members who purchased NBY, and analyze how NBY influence customer behaviors.

This study focuses on three main aspects:

  1. Exploratory data analysis
  2. New NBY abtesting (before and after 12-10)
  3. Modeling

Data

1. Inline Transaction Data 1

ALL transactions occurred in China. Time period is from 01/01/2021 to 12/30/2021.

2. Online Transaction Data 2

ALL transactions occurred in China. Time period is from 01/01/2021 to 12/30/2021.

2. Member Data 3

Member data includes information about the members who made inline purchases in stores, including register_date, UPMID, Membership segments, and so on.

4. Scalper Data 4

Scalper data was based on Yunhai’s Scalper Detection Machine, including scalpers’ UPMID, and so on.

Exploratory Data Analysis

1. HOI NBY Exploratory Analysis

To explore the current characteristics of NBY members, we randomly selected non-NBY members and compare the differences of the two groups.

The distribution of categories of the products purchased

From the bar chart, the category purchased difference of NBY member and non-NBY member is not very significant. The lifestyle category for NBY member is slightly higher than that of non-NBY member, and the run_train category for NBY member is slightly lower than that of non-NBY member.

The new member percentage

The new registered member rate for NBY member is higher than the non NBY member, which means NBY could help to acquire new customers.

HOI NBY member and non-NBY member gender distribution comparison

From the bar chart above, there’s no significant difference on gender distribution of NBY members and non-NBY members for HOI.

HOI NBY member and non-NBY member age distribution comparison

From the HOI age distribution, NBY member shows the younger age trend comparing with non-NBY member.

Inline NBY member and non-NBY member segment distribution comparison

The NBY members show the much stronger interests in Lifestyle than the non-NBY members. Besides, NBY members are less likely to be classified as Sneakerhead.

The high value member percentage

From the bar chart, the high value member proportion for NBY member and non-NBY member is not very significant.

The percentage of members who like purchasing combo

From the bar chart, the difference of tendency to buy combo for NBY member and non-NBY member is not very significant.

2. Online NBY Exploratory Analysis

The distribution of categories of the products purchased

About 60% of the products consumed by online NBY members are lifestyle products, which is higher than non-NBY members. Besides, NBY members buy less Jordan brand products than non-NBY members.

The new registered member rate for NBY member is higher than the non-NBY member, which means NBY helps to acquire new members.

Online NBY member and non-NBY member gender distribution comparison

As for the online gender distribution comparison, the percentage of female NBY members is 62% larger than that of the non-NBY members, and the gap is much higher than that of the HOI gender distribution.

Online NBY member and non-NBY member age distribution comparison

Similarly, from the online age distribution, NBY members also show the younger age trend comparing with non-NBY members.

Online NBY member and non-NBY member segment distribution comparison

For the online NBY member segments, “Family” takes the largest proportion and is much higher than that of the non-NBY members. It is speculated that it’s because the family buyers are more likely to shop online and buy Nike products for their family and friends, and the customization characteristic of NBY is more likely to match their interests. Besides, The proportions of Performance and Sneakerhead of NBY members are much lower than that of non-NBY members.

The high value member percentage

The high value member proportion for non-NBY member is greatly higher than NBY member.

The percentage of members who like purchasing combo

The proportion of members who like buying combo for non-NBY member is higher than NBY member.

New NBY abtesting (before and after 12-10)


Post NBY Purchase AUR = Average Unit Revenue AFTER Their NBY Experience
Post NBY Purchase UPT = Average Unit Per Transaction AFTER Their NBY Experience.
Post NBY Purchase ADPT = Average Demand Per Transaction AFTER Their NBY Experience.Percent of transactions containing a mixture of FTW/APP/EQ AFTER NBY experiences. This helps us to see if NBY member purchase more head-to-toe combination than regular consumer.
Post NBY Purchase Product Design – Lifestyle = Percent of Lifestyle items purchased AFTER NBY experience
Repurchase within 30 days = Percent of NBY member that came back to HOI and made a purchase within 30 days AFTER the NBY experience.

• NBY Member’s AUR is much higher than control group, no matter the recent change of NBY. They usually buy more expensive items, and the New NBY does not influence much of this behavior.
• NBY Member’s AUR is much higher than control group, no matter the recent change of NBY. They usually buy more expensive items, and the New NBY does not influence much of this behavior.
• Because NBY users purchase less items, but more expensive items, the ADPT of the NBY member vs control group is usually lower. However, after the new NBY, we see the gap between the NBY member and control group is gone.
• Before the new NBY, there is no difference between NBY member and control group. However, after the new NBY, we see that NBY member combo purchase rate is significantly higher than control group.
• NBY Member’s tendency to purchase Lifestyle product is much higher than control group, even the control group has the same percentage of Lifestyle loving member. no matter the recent change of NBY. They usually buy more lifestyle items, and the New NBY does not influence much of this behavior.
• Before the new NBY, NBY member’s tendency to repurchase after the NBY experience is no difference comparing to control group. However, after the new NBY, we see that NBY member’s tendency to repurchase is much higher than the control group.

Modeling

1. HOI NBY Modeling

From the EDA part, we have conclusions as shown below:
1. NBY members show the younger age trend comparing with non-NBY members. NBY members have much higher percentage (8%) on age group, ‘0-17’, which is four times higher than that of non-NBY members.
2. The NBY members show much stronger interests in Lifestyle and less interests in sneakerhead than the non-NBY members.
Let’s verify these findings from the modeling.

Variables to be used in training:

• Gender
• Age Group
• Member Segment - Lifestyle
• Member Segment - SNKRHEAD
• Member Segment - Performance
• Member Segment - Family
• High Value Member - Member who purchased $375 or more in a year
• If_colorful - if a member puchased colorful Products (except for black, grey, and white for the primary color) in a year
• If_combo - Tendency to Purchase Combo Products. Member who purchased products with 2 or more divisions in a year
• Lifestyle - The order quantities of lifestyle products
• JORDAN_BRAND - The order quantities of JORDAN_BRAND products
• sports - The order quantities of sports products
• run_train - The order quantities of run_train products
• Dependent Variable: Buy NBY or Not (Binary Variable) - processed from HOI purchase history
Algorithm used: Logistic Regression Classification

\(Logit(P(Y=NBY(1)))=log(p/(1−p))=\beta_0 + \beta_1 Gender + \beta_2 AgeGroup + \beta_3 Lifestyle + \beta_4 Performance + \beta_5 Sneakerhead⋯+\beta_n HighValue*\)

Logistic Regression is commonly used for binary classification problems. one advantage of this algorithm is that it gives β for each term, and the signs can explain the effect of each term on dependent variable.


Model result
Variable Log_Odds
segment_Lifestyle 1.38
Male 1.11
age_0_17 1.05
Female 0.82
if_combo 0.54
if_colorful 0.17
sports 0.05
lifestyle 0.04
JORDAN_BRAND -0.05
run_train -0.16
age_31_35 -0.3
segment_SNKRHEAD -0.63
age_46_60 -0.68


The model result confirms our understanding from the EDA part. The member with younger age and ‘lifestyle’ segment is more likely to purchase NBY. Female is slightly more likely to purchase NBY than male, but the difference is not very significant. The member who like buying combo and with high values is more likely to purchase NBY. The Sneakerhead and the elder members show less interests in NBY comparing with others.

Model performance

ROC
Testing set (25% of the sample data) is used for measuring model’s prediction performance, and a ROC is created.

The ROC Plot:
The model achieved an AUC of 0.7176.

2. Online NBY Modeling

From the online NBY EDA part, we have conclusions as shown below:
1. Female shows more interests in NBY than male. And similar to HOI, NBY members with the younger age are more likely to buy NBY, but the trend is not strong as HOI.
2. Most of the online NBY members are classifid as Family segment. And similar to HOI, the NBY members show less interests in sneakerhead than the non-NBY members.

Similarly, we use Logistic Regression Classification to analyze the variables which could affect or predict if a member buys NBY or not for online platforms. Except for the variables we used in HOI NBY modeling, we also include buyer_type in the online NBY model to better capture the member characteristics.

Variables to be used in training:
• Gender
• Age Group
• Member Segment - Lifestyle
• Member Segment - SNKRHEAD
• Member Segment - Performance
• Member Segment - Family
• buyer_type - LastYear
• buyer_type - Lapse
• buyer_type - ActiveNonBuyer
• High Value Member - Member who purchased $375 or more in a year
• If_colorful - if a member puchased colorful Products (except for black, grey, and white for the primary color) in a year
• If_combo - Tendency to Purchase Combo Products. Member who purchased products with 2 or more divisions in a year
• Lifestyle - The order quantities of lifestyle products
• JORDAN_BRAND - The order quantities of JORDAN_BRAND products
• sports - The order quantities of sports products
• run_train - The order quantities of run_train products
• Dependent Variable: Buy NBY or Not (Binary Variable) - processed from HOI purchase history

Algorithm used: Logistic Regression Classification

\(Logit(P(Y=NBY(1)))=log(p/(1−p))=\beta_0 + \beta_1 Gender + \beta_2 AgeGroup + \beta_3 Lifestyle + \beta_4 Performance + \beta_5 Sneakerhead⋯+\beta_n HighValue*\)

To check the relationships among all the variables, we create the predictive power score plot.



We can see that the features, segment, if_colorful, and preferred gender have strong predictive power score for if_NBY_member, which means they might be quite significant in the model to predict if_NBY_member.

Model result
Variable Log_Odds
if_colorful 6.44
segment_Family 3.1
segment_Lifestyle 0.61
Female 0.43
lifestyle 0.42
if_high_value -0.04
JORDAN_BRAND -0.15
sports -0.2
if_combo -0.47
age_41_45 -0.72
age_36_40 -0.75
age_46_60 -1.25
segment_SNKRHEAD -1.41
buyer_type_Lapsed -5.02


Similar to the results of HOI modeling, the member with ‘lifestyle’ and ‘family’ segment is more likely to purchase NBY and the Sneakerhead doesn’t show much interests in NBY. The members who like buying colorful and the high value members are more likely to purchase NBY online. Besides, the members with elder ages are less likely to purchase NBY, which is align with the conclusion we got from HOI modeling. The buyer type, ‘Lapsed’ is less likely to purchase NBY.

What’s different from HOI is, the female has more interests in NBY through online shopping than male, while the gender difference for HOI is not that significant. Besides, the member with ‘family’ segment is very significant to the prediction if a member purchase NBY or not.

Model performance

ROC
Testing set (25% of the sample data) is used for measuring model’s prediction performance, and a ROC is created.

ROC Plot:
The model achieved an AUC of 0.9424. The model can be used for both explanatory and predictive purposes.

Conclusion

• For both HOI and online platforms, the members with younger ages and ‘lifestyle’ segments are more likely to purchase NBY and Sneakerhead show less interests in NBY.
• FOr online platforms, female shows stronger interests than male. The ‘family’ segment and if a member like colorful products or not greatly influence if a member would buy NBY.

Acknowledgment

Thank Yunhai Zhang for the supervision and guidance.

  1. Source: Nike Online Order Line↩︎

  2. Source: Nike Online Order Line↩︎

  3. Source: Nike Membership Data↩︎

  4. Based on the scalper detection machine built by Yunhai Zhang ()↩︎