Attribute centric product search
RS Components – QuickFinder
RS Components, the number one high-service distributor of electronics and maintenance products across Europe and Asia Pacific.
Customer Service teams currently use a category based search tool that follows the companies category taxonomy. However, users often search for products by describing their attributes, rather than category. This is especially true when Customer Service teams deal with customer enquiries.
This results in search results being too broad and often returning many irrelevant products. This increases the resolution time during enquiries, as agents are forced to refine search results manually. This is in part due to the current infrastructure relying on traditional search techniques and relational databases.
The time to resolution is further increased, as Customer Service teams face a high degree of tool fragmentation. This requires agents to navigate multiple tools to resolve each customer enquiry. Examples of these tasks include technical datasheet lookup or accessing information on discontinued products.
Hack and Craft created an improved end to end search capability. This comprised a dynamic web frontend, a new product data structure and a layer of search APIs.
We created a system based on attribute centric search logic, which more closely aligns with real user enquiries. We then standardised the product data into the eCl@ss ADVANCED format and served it from a new graph database. The new infrastructure allows for greater scalability and speed, while returning faster and more accurate search results.
The tool also incorporates new functionality and data such as:
- Cross referencing RS Part Numbers and Manufacturer Part Numbers
- Searching for product equivalence and compatibility
- Access to data on discontinued products, accessories and alternatives
- Access to the latest technical data sheets for each product
How We Did It
Hack and Craft used our unique four phase agile methodology (Discovery, Development, Roll Out and Continuous Delivery), allowing incremental refinements based on stakeholder and real-world feedback.
At the start of the project we formed a cross departmental Innovation team comprising key stakeholders, including Customer Service and web search experts who understood the types of queries end-users make. The Innovation team took part through each stage of the development process, playing a key role in steering the solution to best market fit.
The Discovery phase included an initial workshop to identify the core user stories, pain points, integrations and data sources that we required.
The Development phase comprised a series of iterative sprints, where we rebuilt the product data in a new graph database and provided a layer of search APIs on top. At the end of each sprint Hack and Craft deployed a working version of the tool. This allowed the Innovation team to test, experiment and provide feedback.
During the Roll Out phase we refined the tool to best market fit by conducting Alpha testing and feedback sessions with end users.
In Continuous Delivery Hack and Craft provides a managed service and supports the tool long-term. We also engage with RS Components product owners in monthly steering groups.
The dynamic web frontend enables the display, sorting and filtering of product search results. Hack and Craft standardised the product data into eCl@ss ADVANCED compatible formats for transfer and use across multiple systems. We stored this data in a modern graph database, which facilitates user searches based on attribute rich search strings.
The rest of the backend infrastructure and APIs are based on an extensible micro services architecture. This flexibility allows the onboarding of additional services or re-use of services in other RS projects.
Integrations with RS Components’ APIs enable the display of real-time stock and price information. Further integrations include access to the latest technical datasheets for each product.
QuickFinder has consolidated the functions of multiple tools, reducing fragmentation. The underlying infrastructure has also increased search speed and accuracy. This has streamlined the workflow of end-users, resulting in the faster resolution of customer enquiries.
The infrastructure developed for this project will also help facilitate future projects. As other business units and projects can re-use and leverage the innovations already gained.