Problem Statement

Predicting Coupon Redemption

XYZ Credit Card company regularly helps its merchants understand their data better and take key business decisions accurately by providing machine learning and analytics consulting. ABC is an established Brick & Mortar retailer that frequently conducts marketing campaigns for its diverse product range. As a merchant of XYZ, they have sought XYZ to assist them in their discount marketing process using the power of machine learning.

Discount marketing and coupon usage are very widely used promotional techniques to attract new customers and to retain & reinforce loyalty of existing customers. The measurement of a consumer’s propensity towards coupon usage and the prediction of the redemption behaviour are crucial parameters in assessing the effectiveness of a marketing campaign.

ABC promotions are shared across various channels including email, notifications, etc. A number of these campaigns include coupon discounts that are offered for a specific product/range of products. The retailer would like the ability to predict whether customers redeem the coupons received across channels, which will enable the retailer’s marketing team to accurately design coupon construct, and develop more precise and targeted marketing strategies.

The data available in this problem contains the following information, including the details of a sample of campaigns and coupons used in previous campaigns -

User Demographic Details

Campaign and coupon Details

Product details

Previous transactions

Based on previous transaction & performance data from the last 18 campaigns, predict the probability for the next 10 campaigns in the test set for each coupon and customer combination, whether the customer will redeem the coupon or not?

Dataset Description

Here is the schema for the different data tables available. The detailed data dictionary is provided next.

You are provided with the following files:

train.csv: Train data containing the coupons offered to the given customers under the 18 campaigns

Variable	Definition
id	Unique id for coupon customer impression
campaign_id	Unique id for a discount campaign
coupon_id	Unique id for a discount coupon
customer_id	Unique id for a customer
redemption_status	(target) (0 - Coupon not redeemed, 1 - Coupon redeemed)

campaign_data.csv: Campaign information for each of the 28 campaigns

Variable	Definition
campaign_id	Unique id for a discount campaign
campaign_type	Anonymised Campaign Type (X/Y)
start_date	Campaign Start Date
end_date	Campaign End Date

coupon_item_mapping.csv: Mapping of coupon and items valid for discount under that coupon

Variable	Definition
coupon_id	Unique id for a discount coupon (no order)
item_id	Unique id for items for which given coupon is valid (no order)

customer_demographics.csv: Customer demographic information for some customers

Variable	Definition
customer_id	Unique id for a customer
age_range	Age range of customer family in years
marital_status	Married/Single
rented	0 - not rented accommodation, 1 - rented accommodation
family_size	Number of family members
no_of_children	Number of children in the family
income_bracket	Label Encoded Income Bracket (Higher income corresponds to higher number)

customer_transaction_data.csv: Transaction data for all customers for duration of campaigns in the train data

Variable	Definition
date	Date of Transaction
customer_id	Unique id for a customer
item_id	Unique id for item
quantity	quantity of item bought
selling_price	Sales value of the transaction
other_discount	Discount from other sources such as manufacturer coupon/loyalty card
coupon_discount	Discount availed from retailer coupon

item_data.csv: Item information for each item sold by the retailer

Variable	Definition
item_id	Unique id for itemv
brand	Unique id for item brand
brand_type	Brand Type (local/Established)
category	Item Category

test.csv: Contains the coupon customer combination for which redemption status is to be predicted

Variable	Definition
id	Unique id for coupon customer impression
campaign_id	Unique id for a discount campaign
coupon_id	Unique id for a discount coupon
customer_id	Unique id for a customer

To summarise the entire process:

• Customers receive coupons under various campaigns and may choose to redeem it.
• They can redeem the given coupon for any valid product for that coupon as per coupon item mapping within the duration between campaign start date and end date
• Next, the customer will redeem the coupon for an item at the retailer store and that will reflect in the transaction table in the column coupon_discount.

Public and Private Split

• Test data is further randomly divided into Public (40%) and Private data (60%)
• Your initial responses will be checked and scored on the Public data.
• The final rankings would be based on your private score which will be published once the competition is over.

Note

• AV_amex_lgb_folds_v28.csv Private Score of 92.50 (Submitted)
• AV_amex_stack2_folds_v28.csv Private Score 92.811 (Best out of all - mean of CB and LGBM)
• Stacking always works, dont ignore whatever Public LB says
• Kaggle Link Best Kernel -**v31**

Success.ai’s Startup Data with Contact Data for Startup Founders Worldwide provides businesses with unparalleled access to key entrepreneurs and decision-makers shaping the global startup landscape. With data sourced from over 170 million verified professional profiles, this dataset offers essential contact details, including work emails and direct phone numbers, for founders in various industries and regions.

Whether you’re targeting tech innovators in Silicon Valley, fintech entrepreneurs in Europe, or e-commerce trailblazers in Asia, Success.ai ensures that your outreach efforts reach the right individuals at the right time.

Why Choose Success.ai’s Startup Founders Data?

1. Comprehensive Contact Information
2. Access verified work emails, phone numbers, and LinkedIn profiles for founders and key startup executives worldwide.

3. AI-driven validation ensures 99% accuracy, providing reliable data for effective outreach.

4. Global Reach Across Startup Ecosystems

5. Includes profiles of startup founders from tech, healthcare, fintech, sustainability, and other emerging sectors.

6. Covers North America, Europe, Asia-Pacific, South America, and the Middle East, helping you connect with founders on a global scale.

7. Continuously Updated Datasets

8. Real-time updates mean you always have the latest contact information, ensuring your outreach is timely and relevant.

9. Ethical and Compliant

10. Adheres to GDPR, CCPA, and global data privacy regulations, ensuring ethical and compliant use of data.

Data Highlights

• 170M+ Verified Professional Profiles: Includes startup founders and their teams across a range of industries.
• 50M Work Emails: AI-validated for precision in communication.
• 30M Company Profiles: Gain insights into the startups’ size, location, and industry focus.
• 700M Global Professional Profiles: Enriched data for comprehensive outreach and analysis.

Key Features of the Dataset:

1. Founder Decision-Maker Profiles
2. Identify and connect with founders and key executives who drive strategy, funding, and product decisions.

3. Engage with individuals who can approve partnerships, investments, and collaborations.

4. Advanced Filters for Precision Targeting

5. Filter by industry, funding stage, region, or startup size to narrow down your outreach efforts.

6. Ensure your campaigns target the most relevant contacts for your products, services, or investment opportunities.

7. AI-Driven Enrichment

8. Profiles are enriched with actionable data, offering insights that help tailor your messaging and improve response rates.

Strategic Use Cases:

1. Investor Relations and Funding Opportunities

2. Connect with founders seeking investment, pitch your venture capital or angel investment services, and establish long-term partnerships.

3. Business Development and Partnerships

4. Offer collaboration opportunities, strategic alliances, and joint ventures to startups in need of new market entries or product expansions.

5. Marketing and Sales Campaigns

6. Launch targeted email and phone outreach to founders who match your ideal customer profile, driving product adoption and long-term client relationships.

7. Recruitment and Talent Acquisition

8. Reach founders who may be open to recruitment partnerships or HR solutions, helping them build strong teams and scale effectively.

Why Choose Success.ai?

1. Best Price Guarantee

2. Enjoy top-quality, verified startup founder data at competitive prices, ensuring maximum return on investment.

3. Seamless Integration

4. Easily integrate verified contact data into your CRM or marketing platforms via APIs or customizable downloads.

5. Data Accuracy with AI Validation

6. With 99% data accuracy, you can trust the information to guide meaningful and productive outreach campaigns.

7. Customizable and Scalable Solutions

8. Tailor the dataset to your needs, focusing on specific industries, regions, or funding stages, and easily scale as your business grows.

APIs for Enhanced Functionality:

1. Data Enrichment API

2. Enrich your existing CRM records with verified founder contact data, adding valuable insights for targeted engagements.

3. Lead Generation API

4. Automate lead generation and streamline your campaigns, ensuring efficient and scalable outreach to startup founders worldwide.

Leverage Success.ai’s B2B Contact Data for Startup Founders Worldwide to connect with the entrepreneurs driving innovation across global markets. With verified work emails, phone numbers, and continuously updated profiles, your outreach efforts become more impactful, timely, and effective.

Experience AI-validated accuracy and our Best Price Guarantee. Contact Success.ai today to learn how our B2B contact data solutions can help you engage with the startup founders who matter most.

No one beats us on price. Period.

Media habits. Civic engagement. Topics: 1. Media habits: access news online in the last two days; kind of access: website of the news source, news aggregator app or website, articles or posts in personal online social networks or that were shared by friends, messaging app or alerts, search engine, email newsletters or notifications, email from relatives, friends, or acquaintances; reaction to the news: discuss it with relatives, friends, or colleagues, send an article or video about that news to someone, comment on or share the news on online social networks, share the news on a messaging app, leave a comment on a news website, search for more information about the topic, save the article for later, research online to try to learn more; use of selected online social networks in the last week. 2. Civic engagement: feeling of being informed by civil society organisations (CSOs) about issues that matter to the respondent personally; engagement with CSOs in the own country in the following ways: participation in demonstrations or similar activities organized by a CSO, regular volunteering in various activities for CSOs, engagement with CSOs mainly online or on social networks, actively encouraging other people to engage with a CSO, donate money to CSOs, engagement in another way, no engagement with CSOs; engagement of respondent’s CSO in European issues; factors that increase personal engagement in CSOs: conviction that personal engagement will have a real impact, knowledge how personal financial engagement will be used by the CSO, participation in concrete activities organized by CSOs, regular information on ongoing activities and projects, own opinion and input are taken into account, receive feedback on what has been achieved, choose a flexible form of engagement, other; most important topics to be treated by CSOs in the own country; impact of campaigns seen in the last two years on personal behaviour: share a video or an image from this campaign, discuss the campaign topic with relatives, friends, colleagues, decide to vote in elections, take concrete actions, take part in public discussions on the campaign topic, take part in an online consultation on the campaign topic, do some research online to find out more about the campaign topic, decide to donate money to a CSO, other; participation in public consultations in the own city in last twelve months; usefulness of the initiative; reasons for considering the initiative useful; reasons for considering the initiative not useful; recent reception of media reports on activities of the European Parliament; general direction things are going in the EU. Demography: frequency of internet use; image of the EU; assessment of the own country’s membership in the EU as a good thing; age; sex; nationality; age at end of education; occupation; professional position; region; type of community; own a mobile phone and fixed (landline) phone; household composition and household size; nation group; weighting factor. Additionally coded was: respondent ID; country; type of phone line.