Fimbre wants to develop a community-based recommendation app
Develop a robust product recommendation app. Acquire, engage and retain users at the lowest possible cost.
The idea behind Fimbre was to develop a community-based recommendation app. The requirement was to create a mobile platform where people could share genuine product reviews. User analytics was the core pillar of Fimbre’s strategy as the success of the application was enormously based on user engagement. The client wanted to leverage our data-driven development approach to draw valuable insights, understand the user behavior, and build the application accordingly.
Fimbre chose Unthinkable to bring this vision to life and launch the Fimbre app promptly as our data-driven development approach gives us an edge over other custom development companies and helps us deliver an end-to-end solution in the true sense. Fimbre wanted to draw user insights that can help them build long-term capabilities and design to improve customer experience and Unthinkable was able to provide that.
An innovative approach to software engineering by simultaneously using user analytics to improve the product through data-driven product development, user engagement, and acquisition
The Fimbre app works as a community-based product recommendation app,allowing users to give genuine ratings to their most preferred products which help other users make informed buying decisions. The app also has a provision to redirect to e-commerce platformsthat are integrated within the app for users to purchase the products. As a part of the app’sreward system, users get 10% of the earnings whenever someone shops from their recommendation
The project took place in three phases – Data-driven software development, data-driven user
engagement, and data-driven user acquisition. This strategy enabled us to take user actions andfeedback into account as user analytics is what differentiates conventional software engineering from our data-driven software engineering.
The Unthinkable team developed the solution architecture, defined the data structure, designedthe user interface, and delivered an MVP in close collaboration with the product owner and the key developers of Fimbre. The team started with an in-depth analysis of the business idea to help Fimbre define a clear product vision and prioritize the scope.
The process of data-driven software engineering works on a consumer-first approach that involves tracking the consumer behavior analytics and making design iterations based on the user behaviorinsights making it progressively easier for the client to achieve the intended goal of the
The MVP of the app was built using NodeJS and MongoDB for the backend, and ReactNative for thefrontend. The agenda was to place a major focus on analytics to thoroughly track the user
journey and draw actionable insights to make the app better.
Once the MVP was developed and launched, the product design team used Mixpanel, a tool used to analyze key user actions to help improve conversion rates, identify high and low-performing user segments and locate the steps that cause friction. It also helps break down conversion data by any user attribute or behavior to understand which users convert best. Mixpanel provided data to the product design team to understand the exact events where maximum user drop-offs took place.
CleverTap is a customer engagement and retention platform that provides the functionality to integrate app analytics and marketing to draw user insights.
they had the screens where the drop-offs were taking place they could go back and make designiterations to make them more user-friendly. In this process, the team identified the UI/UX issues that led to the user drop-offs and fixed them simultaneously which in turn led to a major decrease in the drop-off rate.
This data was used to understand the user behavior and accordingly build a market-ready version of the Fimbre app.
The next stage of the development process was data-driven user engagement. The idea was to meaningfully connect the users to the client’s application. Firstly, our user engagement team segmented the clients into various categories based on their behavior such as active users, engaged users, dormant users, purchasers, about-to-uninstall users, and many more. The behavioral insights were drawn from the data they received through CleverTap, which is a customer engagement and retention platform that provides the functionality to integrate app analytics and marketing to draw user insights. Once the segmentation was done, workflows were created for each segment to get them to re-engage with the app. An example of this would be a strategy to send push notifications to non-active app users in order to get them to re-engage with the app. Through CleverTap, the user engagement team could draw out the details about non-active users along with their demographic details and send push notifications recommending a question or product based on their interests or requirements.
This practice helped Fimbre grow its userbase manifold within a short period of time as multiple-touchpoint workflows were formulated aimed at individual users to ensure long-term user engagement is maintained. However, the strategy was not just for dormant users to re-engage but also for active users to increase their number of events (clickables or screens) while on the app. Whenever a moderate user gets a notification recommending its product of interest or an alert about winning rewards for their recommendation, they would seemingly engage more with the app.
After testing out multiple workflows, custom funnels were made for each user segment as per the user responsiveness for all the workflows. For instance, when a user abandons a recommendation before submitting it, CleverTap creates a separate segment of such users, and a custom workflow is created which for instance sends a notification alerting the user to take further action on the abandonment issue and complete the recommendation process. These user funnels were regularly monitored and modified as per the user behavior and consisted of different channels of communication for each segment.
Concrete workflows were created to identify a journey that a user should take. All the users were segmented on the basis of the ones that are passing through and the ones that are dropping off along the way and the reason behind these behavior patterns were identified and fixed.
The next step was acquiring new users and it was a two-step process. The first step was increasing the organic reach and the second was performance marketing, i.e. paid ad campaigns. The ASO (App Store Optimization) team helped to get the app to rank on the AppStore. Sensor Tower, an enterprise-grade market intelligence and performance metrics tool was used by the ASO team to search for competitive keywords and optimize the app for the same
The app screens and description were uploaded as per the suggestions given in Sensor Tower. Along with that, multiple experiments were run such as A/B testing for different versions of the app listing to figure out the best approach. Another instance for the experiments would be testing user response for app categories where the team realized that listing the app in the lifestyle section of the app store had better results than listing it as a social app.
Once the app was optimized in the AppStore, the user acquisition team started running organic and paid campaigns on various channels to get new users onboard. Paid ads were run on Facebook with deep linking to ensure that custom ads were shown to the audience as per its demographic and behavioral data to increase user acquisition. For instance, if the Facebook data shows that a person has recently become a mother, ads about newborn care product recommendations would be shown to her, and if she clicks the CTA, she’d be redirected to the landing page of the product recommendation she was interested in.