One of the most exhilarating competitions took place late last year. And no, we’re not talking about the PopTart Bowl. Rather, over 550 participants entered the Datapalooza hackathon that lasted for six weeks, concluding December 2023.
Datapalooza encouraged developers to ignite the partnership between Twilio Segment, Amazon Web Services (AWS), and Data Bricks by building projects that highlighted the value of integrating these tools.
Submissions could include machine learning models, mobile or web apps, dashboards, integrations, widgets, extensions, or other tools.
Without further ado, what you've all been waiting for, let’s have a little drum roll to announce the winners! 🥁
First place 🥇
In first place, we have Tiny Segment by K Singh!
The data generated from customer data platforms like Twilio Segment is collected and processed to reveal customer insights. Based on these insights, customers can take actions like asking the sales team to reach particular users, improving the pricing, or updating our website pages. However, many times, these lead to delayed actions and missed opportunities.
What would it look like to instantly update relevant parts of website pages or app screens based on data about a repeat user?
Enter Tiny Segment!
TinySegment helps write rules and update content based on how customers interact with websites/apps.
One example would be if a user has visited a pricing page 5+ times, spent on average 1 minute on the pricing page, and left without clicking the pay button, customers could update the pricing table for that particular user (mapped via Segment's anonymous ID) and offer a heavy discount.
TinySegment consists of a web app and an open-source javascript library to be used on top of the Twilio Segment platform and AWS S3 buckets. Web-App helps view AWS S3 logs, create code components to replace particular parts of the website, and manage configurations and APIs.
TinySegment.JS library helps integrate and auto-update a website's parts, based on who is visiting the page, and their behavior/history as per Twilio Segment track/identify logs.
Second Place 🥈
In second place was Data Tailor AI created by Ashwin Kumar and Lucas Lu!
Data Tailor AI uses Segment’s tracked data to help Large Language Models (LLMs) write hyper-personalized marketing emails for customers. With Data Tailor, marketers no longer have to define their own rules for audience creation. End-customers benefit from receiving outreach that is greatly curated for what they want to purchase.
Here’s how it works:
Customer behaviors and actions are tracked via Segment, then piped through to Databricks where it is cleaned, transformed, and loaded into SQL tables.
Within Databricks, Data Tailor AI ingests data from those SQL tables and engineers a detailed prompt to send to the LLM (OpenAI’s GPT-4). It receives a customized email for that customer and stores it in a separate Databricks table.
Periodically, these generated emails are sent from Databricks to SendGrid (email automation) through Segment’s Reverse ETL feature. Marketers can use SendGrid to send customized emails out as a one-time send or part of a recurring campaign.
This entire process is automated so marketers can set up a workflow to create nurture campaigns and more, powered by LLMs and AI. And though we cover marketing emails in this hackathon, Data Tailor can easily be expanded for other use cases like sending customer support messages.
Third place 🥉
And in third place, we have EasyClick, created by Revanth N D, Aravind Swaminathan, Mukesh Kumar, and Rohith ND.
Effectively managing customer data is crucial for businesses in today's dynamic environment, but manual processes in product subscription management can lead to inefficiencies and errors. The challenge lies in effectively predicting customer loyalty to reduce churn rates and enhance retention.
With technology advancing at a breakneck pace, leveraging automated systems for customer data analysis and targeted marketing is becoming increasingly crucial for businesses to stay competitive and relevant.
EasyClick is a subscription management platform that uses predictive analytics for customer loyalty ratings, enabling businesses to target key customer segments. It not only simplifies subscription management but also transforms it into a strategic asset for businesses, driving growth and fostering long-lasting customer relationships.
For example, EasyClick can streamline the billing process, making it hassle-free for both businesses and customers. It can also identify key customer segments, allowing businesses to focus their marketing efforts more effectively.
Honorable mentions
With so many creative entries, we couldn’t highlight just three. Below are some of the honorable mentions:
Mind Link
Mind Link offers a personalized mental health experience with data from user interactions, mood tracking, and preferences to curate a customized mental wellness plan. Through targeted recommendations and timely interventions, Mind Link ensures users' mental health journey is uniquely theirs.
MindLink built a modern data foundation with connections from Twilio Segment. Data flows seamlessly into the lakehouse, a central repository for all user information. This unified view empowers advanced analytics and personalizations, unlike siloed data systems.
MindLink also leverages Databricks and AWS to build a robust data management platform. This "composable" approach allows for flexible integration of various data sources and tools.
It enables advanced use cases like:
-
Multi-touch attribution: Understand the impact of different marketing channels on user behavior.
-
Customer segmentation: Group users with similar needs and preferences for targeted interventions.
Facets
Facets offers a much-needed improvement to dating apps. It aims to help users gain better self-awareness and realize room for improvements in their profiles. Facets encourages users to create a dating profile, which is sent to friends, family, and potential dates to a public link. In return, they receive information about the personality that the profile portrays.
This angle helps users understand the importance of what the profile communicates about them, and gives them data in a central location as well as additional feedback. When the user gets feedback, the Twilio Segment Engage Journey feature can trigger email notifications that let the user know they got new feedback.
Facets used react-native to build iOS and Android apps for the user to build their profile. Facets’ backend is based on nodejs and the database is using postgres. The builders used next js to build the feedback web app that is used to give feedback on the profile created in the react-native app.
HealthHub
HealthHub empowers users to seamlessly connect with healthcare professionals by providing location-based recommendations and appointment booking. Users can input their symptoms and the app intelligently suggests suitable doctors, streamlining the appointment process for both patients and healthcare providers.
It leverages a robust tech stack to deliver a secure and efficient user experience. The backend is developed using Node.js, ensuring scalable and responsive server-side operations. React is employed for the frontend, providing a dynamic and user-friendly interface.
Data modeling and analytics are powered by Databricks, enabling intelligent insights into health trends and optimizing the recommendation algorithm. MySQL serves as the relational database management system, ensuring data integrity and reliability.
Communication services, including email notifications, are facilitated through Customer.io, enhancing user engagement and appointment reminders.
AWS is utilized for cloud infrastructure, ensuring the scalability, security, and reliability of the HealthHub platform.
Superlatives
And finally, we landed on some pretty cool superlatives we just had to share.
Most unexpected marketing use case
DATArt takes business data and generates a single graphic that summarizes the distribution of the desired data filters based on volume in a visually interesting and intuitive way.
Figuring there was more to visualization than bar charts, the team craved a more fun and functional way to share insights.
Most wholesome
With a name like Pawsitive Care, you know you’re in for a treat.
Inspired by a passion for improving the well-being of pets and making pet care services more personalized and effective, the team set out to create a platform that not only enhances the experience for pet owners but also helps pet care service providers better target their audience.
Pawsitive Care uses cutting-edge AI and segmentation techniques to offer a comprehensive pet caring service. It combines predictive analytics with recommendations to warn users about potential health issues in pets at an early stage, saving both pet owners and service providers time and money. The platform focuses on pet boarding and veterinary services, aiming to match the right services with the right user based on their unique needs and characteristics.
The team built Pawsitive Care on a robust architecture that leverages various technologies. The system collects demographic and behavioral data through the Segment platform and supplements it with geographic and psychographic data from external sources (mock data). All this information is stored in Snowflake, ensuring a secure and scalable data storage solution. The AutoML feature of Databricks is employed to train and identify the best machine learning model using historical and external data (mock data). After using ML to predict user needs and behavior, we write results back to Snowflake.
Best use of open data
The final category goes to Student Loan Debt Visualization.
And as student loan debt in the US has reached staggering levels, students are promised debt cancellation to some degree; however, other motives seem to take priority.
By creating a visualization, the goal is to remind users of the crisis that individuals and families face with the financial burden taken on. Users can explore the total student loan debt balance of each US state. The visualization is built with d3.js. AWS S3 is used to deploy websites, and Twilio Segment to track user behavior and interaction.
Segment tracks user interaction of these information cards. And for this simple prototype, the builder chose to analyze this data as the more traction an idea gets, the more interested that user is. Users can get an immediate first-impression of ideas without relying on user surveys!
Conclusion
With over 550 participants, we saw many inspiring projects. This post shows only a small sampling of the magic that can be done with Twilio Segment, Databricks, and AWS. A big congratulations to all of our winners, and thank you to all of our participants.
As we enter 2024, and always, we can’t wait to see what you build!