Enabling India’s leading multi-brand online beauty retailer to leverage technology to boost conversion & engagement rates

Since the requirement was to build a scalable solution that could seamlessly manage the rising user base, the entire application infrastructure was built on cloud. The team defined a strategic roadmap for development, ensuring cost optimization, business continuity, and minimum time to market in mind. Our team selected Amazon Web Services (AWS) as the most suitable cloud service provider. Some of the prime cloud services that were used during development include:

• Amazon Elastic Compute Cloud (EC2) for the deployment of the solution
• Amazon RDS for MYSQL with read replicas to manage the load on the database
• AWS Elastic Search to fetch the required search query in milliseconds

Moving the infrastructure on the cloud gives the benefits of scaling the resources on-the-fly, as the requirement strikes. This brought a striking difference in the CapEx and OpEx of setting up and maintaining the infrastructure of the application.

Daffodil software developed a Q&A platform that enabled the users to ask and answer beauty related questions. It allows customers to share their advice and unbiased point of view on a certain topic, and create beauty-centric conversations so that the people in the community can make informed product choices.

The Q&A platform was built using ReactJS and NodeJS as the core technologies. To make the solution SEO-friendly, server-side rendering of web components was done. This ensures that all HTML components are present in the source code so that the search engine can request, crawl, and index them immediately. 

Along with this, users can view live video sessions shared by various beauty experts. These live videos are fetched directly from the client’s YouTube channel and can be viewed live on the Q&A platform or on-demand as well. 

In order to improve the user experience, the content in the application was properly categorized into various components such as Q&A, posts, live sessions, etc. Since the way of presenting data on the platform was different and had varying components, a new database (with different schema) was developed. The data from the previous platform (users, questions, answers, comments, likes, etc.) was migrated to the new database. In this process, team Daffodil migrated data of over 1.5 million users that contained over 30,000 questions and 70,000 answers. A few highlights of the forum are:

• Users can ask a question with text, images/videos, links, and by tagging a product
• Users can create a post, format the text, add photos/videos, tag products, etc. 
• Users can like, comment, share posts, follow a question, etc. 
• Users can apply filters to find Q&A and posts according to their interests
• Live videos by the client’s team can be watched on the platform or can be accessed on-demand.

Daffodil also created a custom admin panel for the solution that enables the team at the client’s end to manage the content, moderate it, keep a track of business KPIs, and update relevant content on the platform to engage users. 

The dashboard of the panel gives a view of the total number of users on the platform, the number of new users, Q&A stats (number of questions, the average number of questions, number of answers, average number of answers), etc. Along with this, there are several modules that enable the team to:

• Schedule questions, answers, or posts: The client’s team can create new posts, questions, and answers and schedule them for publishing at a defined date and time.
• Manage Widgets on Home Page: Featured posts, tip of the day, announcements that are displayed on the homepage of the solution is managed through the admin panel.
• Product Suggestion during Live Sessions: The team at the backend can make product suggestions during live video sessions. The real-time conversation/chat during the live sessions is powered by PubNub that enables the team to suggest/tag products in the chat while the videos are live.
• Manage Users on Platform: Depending upon the participation of users in Q&A, the admin team can mark answers as “accepted”, bring that answer on the top of other answers, and can verify a user depending upon his continuous contribution on the platform.