ViewPrintto cypress business, is the design and marketing partner for millions of small businesses around the world. For more than two decades, VistaPrint has enabled small businesses to quickly and effectively create marketing products (from promotional materials and signage to print advertising and more) to get the job done, whether they operate in-store or online.
To help small businesses in their brand-building process, VistaPrint offers customers personalized product recommendations, both in real-time and online. vistaprint.com and through marketing emails. These product recommendations improve your customers' experience by making it more efficient to find the products they need, while increasing VistaPrint conversion rates. Since implementing Amazon Personalize, VistaPrint has increased its conversion rate by 10 percent and reduced its total cost of ownership by 30 percent.
In this post, we show you how VistaPrint uses a combination of Amazon Personalize, Twilio Segmentand ancillary AWS services and partner solutions to better understand your customers' needs and provide personalized product recommendations.
Previous solution and challenges
Prior to its current solution, VistaPrint had a product recommendation system developed in-house and hosted on-premises. The first challenge with their previous solution was that the solution could not automatically scale when demand increased. The second challenge was that changes to the internally developed system were time-consuming, because a high degree of machine learning and specialization in the e-commerce domain was required to make modifications.
These challenges led to the decision to create a new cloud-native system that can scale with increased demand and consists of serverless and software-as-a-service (SaaS) components that outsource much of the domain-specific functionality to enable operations. easier and faster times. -to the market for changes.
VistaPrint's New Personalized Product Recommendation System
Figure 1
As seen in Figure 1, the steps for VistaPrint to provide personalized product recommendations with its new cloud-native architecture are:
- Add historical data to a data warehouse. Data from upstream systems, including customer data platforms (CDPs) such as Twilio Segment, order management, product catalog, and user management systems, is collected in a data warehouse, which in the VistaPrint case is Snowflake.
- Transform data to create Amazon Personalize training data. Amazon Personalize uses data about users, items, and interactions, and this data is obtained from Amazon Simple Storage Service (Amazon S3) in CSV format. In the case of VistaPrint, they use Data bricks to perform the necessary data transformations before sending the data to Amazon S3.
- Import massive historical data to train Amazon Personalize models. After incorporating massive historical data into an Amazon Personalize data set, one or more solutions are trained using this data. In the case of VistaPrint, they use the user customization model recipes and similar elements.
- With user personalization, Amazon Personalize predicts which items a user will interact with based on previous interactions between all users.
- With Similar Items, Amazon Personalize generates recommendations for items similar to an item you specify.
To keep the personalization models relevant, steps 2 and 3 are repeated periodically to keep the training data up to date.
- Stream eCommerce website events to a CDP. A CDP is used to capture events from an e-commerce website, for example, when a user views a product or adds a product to their shopping cart. A CDP can also perform identity resolution, which helps identify the user regardless of whether they access a platform from a mobile or from a web client. VistaPrint uses Twilio Segment as its CDP.
- Generate product recommendations in real time as a customer navigates the eCommerce website.. When a customer browses an e-commerce website and a CDP captures these events, they are also forwarded to Amazon Personalize. Amazon Personalize, in turn, generates recommendations for additional products that a customer may be interested in. These recommendations are put back into the eCommerce website experience in real time.
- AWS Lambda is used to send Segment data to Amazon Personalize using Segment's Amazon Lambda Destination. VistaPrint uses Segment Amazon Lambda Destination to perform additional data transformations and gain flexibility to integrate with additional observability tools that are not shown, but other AWS customers may consider Segment destination. Amazon Custom Destination which is suitable for simpler integrations.
- VistaPrint created a personalization service that sits opposite Amazon Personalize. This service provides additional functionality on top of the Amazon Personalize APIs, including the ability to cache recent recommendations in Amazon DynamoDB and integration with VistaPrint authentication and authorization systems.
- VistaPrint created a Placement and Bidding Engine (POE) that allows data scientists and marketers to collaborate. Placement templates are used to create custom placements by allowing a marketer to select an Amazon Personalize model, the visual style of the placement, and additional features, such as displaying a customer's logo as it would appear on the final manufactured product. Figure 2 shows an example of one of these locations, called More with your designas seen in vistaprint.com.
- Generate product recommendations as part of email marketing campaigns.. In addition to offering real-time product recommendations on its website, VistaPrint uses personalized product recommendations in email marketing campaigns. The same POE system is used to design and place product recommendations in email templates.
Figure 2
Business impact
Since implementing its new personalized product recommendation system, VistaPrint has seen a 10 percent increase in conversions from personalized recommendations. Amazon Personalize also reduced VistaPrint's total cost of ownership by 30 percent compared to the previous on-premise solution.
Conclusion
VistaPrint's cloud-native personalized product recommendation system helps the company deliver a more efficient and useful experience to its customers, while increasing the company's conversion rates.
Amazon Personalize is at the center of VistaPrint's personalized product recommendation system and provides a fully managed, machine learning-based solution.
A customer data platform like Twilio Segment allows companies like VistaPrint to create a 360-degree connected view of their customers by aggregating data from all touchpoints with their customers across multiple business domains. This consistent view of the customer leads to more accurate and personalized product recommendations when combined with Amazon Personalize.
Next steps
VistaPrint's personalized product recommendation system is one product within a larger product data network. Read more about Vista's data mesh strategy in this previous post How Vista created a data mesh enabled by solutions available on AWS Marketplace
Also read more about the other topics in this post:
About the authors
Ethan Fahy is a Senior Enterprise Solutions Architect at AWS based in Boston, MA. Ethan has a background in geophysics and enjoys building large-scale cloud-native architectures to support scientific workloads.
Mouloud Lounaci leads the engineering team for marketing optimization at Vista. He is a machine learning enthusiast with around 10 years of experience building ai-based software products to solve complex customer problems. Every chance he gets, Mouloud hops on a plane to discover cultures, foods, and landscapes from around the world.
Emeline Escolivet is the Engineering Manager of the Recommendations team at Vista. With over 10 years of experience as a software engineer, she enjoys turning complex business problems into reliable software solutions. In her free time she likes to define herself as a hiker, dancer and lover of gastronomy.
Vibhusheet Tripathi is a senior data engineer on the Vista recommendations team. When he is not experimenting with machine learning systems, Vibhu enjoys reading, playing sports, and listening to music.
