Wiki Categories

Model Evaluation

Consumer Purchase Data

What is consumer purchase data?

Consumer purchase data is the information about consumer purchases, including the product or service, time of purchase, and the amount spent. This data indicates purchase history, customer buying patterns, and other relevant details, such as stock availability and product appearance. Often augmented with the customer identity data, it covers purchases made from all the devices linked to the shoppers' identity.

Consumer purchase data and the derived insights enable companies to predict the next purchase, make recommendations, and deliver personalized experiences.

Where does the data come from?

The main data sources are either first-party, third-party, or both.

First-party consumer information comes from the transaction records and receipts in the enterprise systems, reports from the sales and support staff, and website engagement data. The website engagement data uses registration details, visitor tracking with cookies and IP addresses, and participation in loyalty programs.

Third-party data comes from vendor partners who track the company websites as well as competitor websites. It often contains insights from online activities, including marketing data from consumer marketing, advertising campaigns, and content distribution.

What types of attributes should I expect when working with consumer purchase data?

The typical attributes of this data are:

  • Purchase amount
  • Products or services purchased
  • Time of purchase
  • Mode of purchase: in-store, website, mobile app
  • Method of purchase i.e. types of credit cards
  • Location of purchase
  • Duration of the purchase visit
  • Reason for purchase: ads or personal recommendations
  • Frequency of visits
  • Participation in loyalty programs
  • Use of reward points or coupons
  • Frequency of returns and refunds
  • Total online orders made over a period of time
  • Amount spent over a time period
  • Number of online orders by cost category
  • The average dollar amount spent per order
  • Average days between online orders
  • Amount spent on: furniture, gifts, children products, home care, computing home office, garden, travel, apparel, beauty, pets, automotive, sports, and leisure items
  • Time since last online order
  • Most frequent retail purchase category
  • Most recent retail purchase date
  • Demographics - name, address, phone number, email

First-party in-store purchase data can be anonymous or known if the customer has filled any form. Website or mobile purchase data has a record of consumer demographics. Third-party data tracks the information online and provides detailed demographics.

How should I test the quality of the data?

Consumer purchase data is a fast-moving category, and the frequency of updates is a major factor in its quality. Collecting a set of historical data and testing it against the latest data helps ascertain its quality. Testing for data consistency is essential to ensure that the derived insights are accurate.

To test the quality of the data:

  • Validate that the data is frequently updated.
  • Ensure that the data matches the requirements for the intended use.
  • Test the data for accuracy and consistency.
  • Verify that the data is complete and does not have gaps or missing data points.
  • Ensure that the data is privacy compliant, as it contains personal data or personally identifiable information (PII).

Who uses consumer purchase data?

Marketers use consumer the data for market research, to understand what the their target audiences are interested in, to improve customer relationships, and optimize customer experience. Businesses use this data to power marketing analysis, plan promotional campaigns, and drive business growth. This data also improves the online recommendation engine.

As this data comprises customer purchase patterns, it can be used for fraud identification and prevention. If a new purchase is outside the pattern of products purchased and the amount spent or done from a different location, it can be identified as someone else fraudulently using the card.

What are the common challenges when buying the data?

Since it is a fast-moving data category, the data quickly loses its value if not updated frequently. Testing that the data brokers deliver recent data is critical in ensuring accurate insights. Besides data timeliness, data accuracy and consistency are also common challenges. 

  • Data timeliness: The data changes quickly. To get the correct intelligence on purchases, the vendor must provide the most recently updated datasets. If the data is obsolete, it will not deliver accurate insights and can affect strategic decisions.  Data timeliness is a major challenge for this data, like any consumer data category, and testing for it is essential to ensure its relevance to the use cases.
  • Data consistency: Consumer purchase data comes from multiple sources, meaning there may be inconsistencies across them. First-party data is often consistent, though the third-party data collected from online sources can have multiple errors. Data consistency is a key challenge in getting accurate insights. 
  • Data accuracy: While the first-party data comes from a company's own online and offline purchase records, the third-party data is collected from several sources. These sources may not provide accurate data and can lead to inaccurate insights.  The variety of attributes, volume of data, and speed of data updates make it challenging to manage data accuracy.
  • Privacy compliance: Consumer purchase data commonly includes sensitive and personal information. It can also include personally identifiable information (PII), and protecting it is a challenge. The vendor must provide privacy regulation compliance certification applicable to the region and industry.

What are similar data types?

Consumer purchase data is similar to audience data, social media data, demographic data, eCommerce data, retail data, and other related data categories used in marketing and sales.

You can find a variety of examples of B2B and company data in the Explorium Data Gallery.

Sign up for Explorium’s 14-day free trial to access the data available on the platform.                         

Access Consumer Purchase Data wit Explorium

What are the most common use cases of consumer purchase data?

The most common use cases for consumer purchase data are purchase intelligence, consumer intelligence, online recommendation engine, and fraud detection. This data category is also used for consumer data enrichment, personalizing experiences, targeted advertising, and building targeted account lists for account-based marketing.

  • Purchase Intelligence: Comprehensive analysis of where, when, what, and how customers buy in-store or online helps marketers plan their strategies to maximize their marketing investment. Purchase intelligence provides insights about purchase patterns, seasonal demands, customer loyalty, market share, and growth opportunities. Typically, purchase intelligence requires machine learning algorithms capable of handling large volumes of data and learning from them. 
  • Consumer Intelligence: This analysis delivers a deeper understanding of consumers with their purchase choices and preferences. Companies leverage these insights to design their promotional campaigns and ensure higher returns on advertising spend. Consumer intelligence also helps companies to understand the market requirements and accordingly align their strategies to drive growth. Business leaders and investors leverage the insights into consumer behavior to launch new ventures. 
  • Online Recommendation Engine: Good recommendations ensure that customers continue using the brand, store, or online platform. Companies have been using such engines to promote products or services for existing customers coming back to them. An online recommendation engine analyzes customer data to generate suggestions for products or services that the customers may find interesting and helpful. These engines learn quickly from real-time and historical data and respond to the changing preferences of customers. They generate accurate recommendations dynamically and leverage the customer interest for quick conversion. Consumer purchase data provides the history of purchases made by a customer and helps the engines recommend aligned products and services.
  • Fraud Detection: The data indicates purchase patterns of a customer, including the average amount spent per order, the most frequent purchase categories, frequency of purchases, and locations of the purchase. If a new purchase shows a deviation from this pattern, such as a new location, it can be identified as possible fraud.  Machine learning tools learn from the purchase patterns to detect and prevent such potential frauds.

Which industries commonly use this type of data?

The consumer products industry is the biggest user of this data. Retail, CPG, sports, entertainment, tourism, travel, hospitality, leisure, healthcare, financial service providers, insurance providers, and banking industries use this data for driving their marketing and sales strategies.    

How can you judge the quality of your vendors for consumer purchase data?

The critical factors to consider when selecting a vendor are domain experience and methods of data collection. Vendor websites can provide the basic information, case studies can indicate the level of vendor experience, and demos can help understand the range of datasets and ease of integration. Talking to a vendor rep is a good way of resolving your queries and judging the suitability of the vendor for your projects. 

  • Customer reviews and testimonials: Most vendors provide customer reviews and ratings on their websites. The ratings indicate the overall quality, and the reviews present more details about the datasets, their quality, timely delivery, and vendor engagement. Customer testimonials highlight the vendor's strengths and can also indicate the fit for your projects.  
  • Case studies: As the next level of deeper vendor analysis, case studies explain how the vendor helped customers solve their problems. The knowledgeability, commitment, and timely actions can be good pointers for shortlisting the vendors. Case studies also signal the vendor's capability of delivering a wide range of datasets and custom datasets. Analyzing case studies can be a valuable method to assess vendor quality.
  • Demo: Watching vendor datasets used in a specific project is a quick way to assess vendor quality. While some vendors provide recorded demos on the websites, some can make them available on request. Some vendors arrange a live demo, which can be a good opportunity to discuss your project requirements. If sample datasets are available, you can also use them to get first-hand experience. 
  • Interacting with vendor reps: The fastest and the most direct method for judging vendor quality is interacting with vendor reps, preferably in person. It can be a great opportunity to discuss your requirements, understand the integration process, and check the availability of custom datasets.  Such interactions also help build a long and fruitful vendor relationship that can benefit all your future projects.

Additional Resources:

Explorium delivers the end-game of every data science process - from raw, disconnected data to game-changing insights, features, and predictive models. Better than any human can.
Request a demo
We're Hiring! Join our global family of passionate and talented professionals as we define the future of data science. Learn More