QBP Bike and Size Recommendation Engine for Shopify
Interactive recommendation experience that helps shoppers through the bicycle decision-making process
QBP Bike and Size Recommendation Engine for Shopify
First Glance
Quality Bike Parts, commonly known as QBP, supports the bicycle industry through a broad network of brands, products and retail partners. QBP features an extensive product catalog and a wide variety of bicycle styles, frame geometries and sizing considerations. With such a vast catalog, QBP recognized that customers needed a simpler way to find the right bike and determine the appropriate size.
Selecting a bicycle is a highly personal decision. A shopper must consider where they plan to ride, the type of experience they want, their physical measurements and the sizing rules of the specific bicycle model. Even experienced riders can struggle to compare options across categories. For newer riders, the number of choices can quickly become overwhelming.
QBP wanted to create an interactive recommendation experience that could guide shoppers through this decision-making process. The proposed solution needed to support two related customer journeys:
- A Product Finder that recommends one or more bicycles based on a shopper’s answers
- A Size Finder that recommends an appropriate frame size using rider measurements and model-specific sizing rules
The tool also needed to do more than provide a single recommendation. QBP wanted a reusable platform that could support multiple Shopify stores, different bicycle brands and additional recommendation experiences in the future.
Sunrise Integration worked with QBP through a detailed discovery process to define the customer experience, administrative tools, recommendation logic, Shopify integrations, data requirements and technical architecture for the new QBP Recommendation Engine.
The Challenge
QBP’s biggest challenge was to transform their complicated product selection and sizing rules into an experience that was easy for customers to use while remaining manageable for internal teams.
Reducing Product Selection Complexity
Bicycle shoppers may begin their search with only a general idea of what they need. They may know that they want to ride on pavement, trails or mixed terrain but may not understand which bicycle category or model best fits those goals.
A useful recommendation tool needed to ask relevant questions one step at a time and translate the customer’s answers into a meaningful result. The experience had to feel simple on the storefront even though the underlying recommendation logic could involve multiple answers, branching paths and product mappings.
The tool also needed to support more than one recommendation. Depending on the quiz and customer responses, the experience might return a primary bicycle, a group of bicycles or supporting content that helps the shopper continue their research.
Managing Model-Specific Sizing Logic
Bike sizing introduces another layer of complexity. A customer’s height and inseam may be used differently depending on the bicycle model, frame geometry and manufacturer sizing chart.
The system needed to collect measurements in a customer-friendly format while normalizing the data behind the scenes. For example, a customer might enter their height in feet and inches or centimeters. The application would need to convert that value into a consistent unit before applying lookup rules.
Some sizing experiences would also require one calculated result to be passed into a later step. A height measurement might produce an estimated inseam value, which could then be reviewed or adjusted by the customer before the final sizing lookup is completed.
Creating a Flexible Administrative Experience
QBP did not want to rely on developers every time a question, answer or recommendation rule changed. Internal users needed an administrative tool that would allow them to:
- Create multiple recommendation experiences
- Add and organize quiz slides
- Configure questions and answer choices
- Define branching paths
- Build sizing lookup tables
- Select predefined layouts
- Enable or disable quizzes without deleting their data
- Connect recommendation experiences to specific products or stores
The system needed to provide flexibility without becoming an unrestricted page builder. Too much design freedom could result in inconsistent layouts, broken experiences and difficult long-term maintenance.
Capturing Useful Data Without Creating Friction
QBP wanted the tool to support retention marketing, segmentation and personalization. However, the recommendation experience could not require customers to submit an email address before receiving a result.
The application needed to display recommendations immediately while still capturing useful interaction data such as:
- Quiz starts
- Quiz completions
- Selected answers
- Recommended bikes
- Recommended sizes
- Product context
- Store or brand context
- Session timestamps
The system also needed to distinguish between anonymous shoppers and known Shopify customers. When a customer was logged in, selected quiz outputs could be saved to Shopify customer metafields for future use.
The Solution
Sunrise Integration conducted a detailed discovery and translated QBP’s business goals into a structured product plan for a reusable Shopify Recommendation Engine.
The proposed solution combines a customer-facing quiz framework, a Shopify embedded administration application, a rule-based recommendation engine, a structured data model and a secure AWS deployment architecture.
A Unified Recommendation Framework
Rather than creating separate applications for every use case, Sunrise Integration designed a common framework in which each recommendation experience is built from a sequence of configurable slides.
A recommendation can be used for a Bike Finder, Size Finder, rack finder or another future guided-selling experience. Each recommendation contains its own slides, actions, logic and result presentation.
This architecture gives QBP a consistent foundation while allowing each quiz to have different content and rules.
Slide-Based Customer Journeys
Each recommendation is composed of individual slides or steps. QBP administrators can create slides using predefined layouts that support three primary content areas:
- Images
- Text
- Customer actions
Customer actions can include buttons, number inputs, height inputs, submit actions and get actions.
This slide-based structure allows QBP to create a guided experience that presents one decision at a time. It also makes complex recommendation flows easier to manage because each question, calculation or result is contained within a defined step.
Predefined Layouts for Consistency
Sunrise Integration recommended a controlled layout system instead of an unrestricted visual editor.
Administrators can select from approved slide layouts and add the appropriate images, text and actions. The application controls the overall HTML structure while the Shopify theme provides typography, colors and other visual treatments.
This approach helps QBP maintain consistent branding across stores while reducing the risk of administrators creating unsupported or inaccessible layouts.
Configurable Buttons and Branching
Button actions form the foundation of standard quiz navigation. Each button can be assigned a value and connected to another slide within the recommendation.
This allows QBP to create branching experiences in which the next question depends on a previous answer.
For example, a customer who selects trail riding could be directed to a different series of questions than a customer who selects urban commuting. Both customers remain within the same recommendation experience but follow paths that are more relevant to their needs.
Administrators can also define starting slides, terminal result slides and restart actions.
Structured Measurement Inputs
The Size Finder includes specialized input types for customer measurements.
The height input presents feet and inches by default while allowing the customer to switch to centimeters. Regardless of the format selected by the customer, the application normalizes the measurement into a standard unit before applying recommendation logic.
Standard number inputs can also be used for inseam or other measurements.
This creates a familiar customer experience while ensuring that all values entering the recommendation engine use a predictable format.
Submit Actions With Built-In Logic
Sunrise Integration designed submit actions to handle both navigation and data processing.
When a customer submits a slide, the application identifies the inputs on that slide and applies the selected logic configuration. The submit action can:
- Combine values
- Convert measurements
- Evaluate conditions
- Perform a lookup
- Produce an output
- Send the customer to the next slide
Each slide can contain only one submit action. A submit action becomes available only when the slide includes an input field. These rules reduce configuration errors and make the behavior of each slide easier to understand.
Table-Based Recommendation Logic
The recommendation engine uses table-style mappings that allow QBP administrators to define input-to-output rules.
A rule can map an exact value to an output or map a range of values to a result. For example:
- If the input equals a specific answer, return a selected product category
- If a height falls within a defined range, return a frame size
- If a calculated measurement falls within a range, return a sizing recommendation
This approach provides a transparent method for managing recommendations. QBP teams can review the tables and understand how each result is produced without reading application code.
Product-Level Configuration Through Metafields
The proposed Shopify integration avoids the need for separate product templates.
The application block or embed can be installed once within the shared product detail page template. Each bicycle product can then reference the appropriate Recommendation Engine configuration through a Shopify product metafield.
When a customer visits a product page, the application reads the product’s configuration reference and loads the correct Size Finder or recommendation experience.
This approach supports a large catalog while reducing theme maintenance.
Customer Metafields and Local Storage
Each slide includes an option that determines whether its output should be saved to a Shopify customer metafield.
Administrators can specify an existing metafield namespace and key. When the customer is logged in, the selected result can be written to the appropriate customer record.
The application can also store values in local browser storage. This supports session continuity and allows the experience to remember previous measurements or result context when technically appropriate.
Marketing and Klaviyo Readiness
The Recommendation Engine was designed to create structured fields that can support future retention marketing and segmentation.
Potential data points include:
- Bicycle interest
- Recommended model
- Recommended category
- Recommended size
- Store or brand
- Product viewed
- Quiz completion status
When Klaviyo is enabled and the appropriate custom fields are configured, customer metafield data can be made available for segmentation and campaign use.
This could allow QBP to develop targeted follow-up experiences based on a shopper’s interests or recommendation history.
Analytics and Reporting
Sunrise Integration defined a foundational reporting layer that allows QBP to evaluate how each recommendation experience performs.
The initial reporting requirements include:
- Quiz start count
- Quiz completion count
- Completion rate
- Most frequently selected answers
- Most frequently recommended bikes
- Most frequently recommended sizes
These metrics can help QBP identify where customers abandon the experience, which answers are most common and how recommendation patterns change over time.
The data can also support future quiz optimization. QBP can revise questions, adjust recommendation rules or improve result content based on actual customer behavior.
An Embedded Shopify Administration Application
The Recommendation Engine will be managed through an embedded Shopify application.
From the application, QBP administrators will be able to view existing recommendations, create new recommendations, manage statuses and build the slides that form each experience.
The administration system will support multiple QBP stores. A recommendation can be connected to one or more stores depending on the intended use.
Shopify session records will securely store the information required for authenticated Shopify API communication.
The Outcome
The discovery process gave QBP a clear plan for transforming complex bicycle selection and sizing rules into a reusable Shopify recommendation platform. Sunrise Integration defined a unified framework that can support Product Finders, Size Finders and future guided-selling experiences.
The completed tool provides QBP with:
- A defined customer journey for bike and size recommendations
- A slide-based administrative experience
- A structured branching and lookup system
- A scalable Shopify product integration strategy
- A tag-based method for managing product results
- A plan for storing anonymous and known customer sessions
- Customer metafield support
- Klaviyo-ready data requirements
The solution design also reduces long-term dependence on developers. Once the application is implemented, QBP administrators will be able to create new recommendation experiences, change content, update logic and manage product mappings through the application and Shopify.
For customers, the proposed Recommendation Engine creates a faster and more confident shopping journey. Shoppers will be able to answer a focused set of questions, receive relevant bicycle recommendations and determine an appropriate frame size without navigating complicated product charts on their own.
For QBP, the platform creates a new source of structured customer insight. Recommendation data can help reveal which riding styles, bicycle categories and sizes are most frequently requested. Known customer data can also support future personalization, segmentation and retention campaigns.
Most importantly, the discovery established a flexible foundation rather than a single-purpose tool. QBP can begin with a focused first version and expand the platform across additional products, brands, stores and recommendation use cases as its needs evolve. Sunrise Integration provided QBP with a practical roadmap for creating a more guided and measurable bicycle shopping experience.
