SWS - AI-Powered Admissions Assistant
Study Work & Settle needed a more efficient way to handle prospective student enquiries and collect the information required before an admissions consultation. Instead of relying on staff to manually ask the same qualification questions through WhatsApp, the SWS Admissions Assistant turns the initial enquiry into a structured, automated intake experience. Students can begin from a simple web entry point, continue naturally through WhatsApp, complete their admissions profile, review their information, and submit it for the SWS team to assess. The result is a business workflow designed to reduce repetitive administrative work, improve lead consistency, and give advisors better information before speaking with prospective students.
How We Started
Study Work & Settle helps prospective international students navigate study opportunities, particularly across Germany and Italy. One of the challenges in this process happens before an advisor can provide meaningful guidance: the business first needs to understand who the student is, what they want to study, where they want to go, their educational background, their budget, and when they hope to begin. Gathering this information manually through messaging creates repetitive work for advisors and can lead to inconsistent or incomplete student profiles. The project began by asking a practical business question: how can SWS automate the repetitive part of admissions intake while keeping the experience familiar and accessible for students? The solution was designed around WhatsApp, allowing students to interact through a communication channel they already understand while the system handles the structured data collection behind the scenes.
Project Process
Understanding the Admissions Workflow
The first step was mapping what happens before an SWS advisor can meaningfully assess a prospective student. The intake process requires several pieces of information, including the student's name, intended study level, preferred country, tuition budget, programme interests, educational background, and desired intake date. Rather than simply building a chatbot, the workflow was designed around the actual information the business needs to move a prospective student toward a useful consultation.
Meeting Students Where They Already Communicate
Instead of requiring prospective students to create an account, learn a new portal, or complete a traditional multi-page form, the experience was built around WhatsApp. A lightweight web landing page introduces the service and directs the student into a WhatsApp conversation with a pre-filled START message. This reduces friction at the beginning of the journey while allowing SWS to keep its admissions process structured behind the scenes.
Automating Student Qualification
Once a student begins the assessment, the backend guides them through a structured sequence of admissions questions. Each response determines the next stage of the conversation, allowing information to be collected consistently without requiring an advisor to manually repeat the same introductory questions for every new enquiry. The workflow turns an informal WhatsApp conversation into structured admissions data that the business can use.
Turning Conversations Into Persistent Business Data
A key requirement was ensuring that the admissions process did not exist only inside a chat conversation. Student responses are progressively stored in PostgreSQL through Supabase, together with the student's current intake stage and application status. This means an assessment can survive server restarts, be resumed later, and eventually feed directly into internal advisor tools rather than forcing staff to manually copy information from WhatsApp.
Giving Students Control Before Submission
Before an assessment is marked as complete, the student receives a structured summary of the information they provided. They can review their study level, preferred destination, budget, programme interests, education and intended start date before confirming the submission. This creates a clearer handoff to the admissions team and reduces the chance of advisors beginning their review with information the student entered incorrectly.
Connecting WhatsApp to the Business Backend
The system integrates directly with the WhatsApp Business Platform through Meta's Cloud API. Incoming student messages are delivered to an ASP.NET Core webhook, interpreted according to the student's current intake stage, stored in the database, and answered through the WhatsApp messaging API. Delivery and status events are handled separately from student messages, allowing the conversational interface and the underlying business workflow to remain cleanly separated.
Designing for Continuity & Reliability
The initial conversation prototype stored student progress temporarily in application memory, which meant an interrupted server could also interrupt the admissions journey. The system was subsequently redesigned around persistent database-backed applications. Each response is saved as the student progresses, allowing the backend to recover the student's current stage and continue the conversation even after the application has restarted.
Preparing the Workflow for AI-Assisted Admissions
The structured intake system also creates the foundation for the next stage of the product: AI-assisted admissions analysis. Once a student confirms their profile, the application can be evaluated using the information already collected rather than asking advisors to reconstruct the student's situation manually. Ongoing development is focused on turning completed profiles into structured advisor assessments, highlighting relevant considerations and eventually supporting programme and university recommendations for destinations such as Germany and Italy.
Building Toward an Advisor Workspace
The long-term value of the system extends beyond the student-facing conversation. Completed applications are being structured so SWS advisors can eventually review student profiles, AI-generated assessments, application status and recommendations from an internal workspace. This turns the WhatsApp assistant from an isolated chatbot into the entry point for a broader admissions operations system.
Key Features
WhatsApp-Based Student Intake
Prospective students can complete their initial admissions assessment through WhatsApp instead of navigating a lengthy form or creating an account before speaking with SWS.
Automated Admissions Questions
The assistant guides students through the core information SWS needs before an admissions consultation, creating a consistent qualification process for every incoming enquiry.
Structured Student Profiles
Answers such as study level, destination, budget, programme interests, educational background and intended start date are converted from chat messages into structured application data.
Persistent Conversation Progress
Student progress is stored as the assessment continues, allowing an intake to survive backend restarts and providing the foundation for students to continue an interrupted assessment.
Review Before Submission
Students receive a summary of their information before the application is completed, allowing them to confirm that their admissions profile accurately represents their plans.
Application Status Tracking
Each student application maintains its current intake stage, completion status and relevant timestamps, giving the business structured visibility into where a prospective student sits in the intake process.
Direct WhatsApp Cloud API Integration
The system communicates directly with the WhatsApp Business Platform, allowing incoming messages and automated responses to become part of the SWS backend workflow rather than relying on manual staff intervention.
Low-Friction Web Entry Point
A responsive landing experience explains the assessment and directs prospective students into WhatsApp with the conversation ready to begin.
AI-Ready Admissions Data
Because student information is collected and stored in a structured format, completed profiles can be passed into an AI assessment layer without requiring staff to manually prepare the information first.
Advisor-Focused Architecture
The system is being developed with an internal advisor workflow in mind, where completed applications, assessments and future recommendations can be reviewed by the SWS team before guidance is provided to students.
Project Outcomes & Results
Transformed the initial admissions enquiry from a repetitive manual WhatsApp conversation into a structured automated intake workflow.
Created a low-friction student journey that begins on the web and continues through WhatsApp without requiring students to learn a new platform or create an account.
Converted conversational responses into persistent, structured student application records that can be used by the wider admissions workflow.
Reduced dependence on advisors manually asking the same preliminary qualification questions before every student consultation.
Created a consistent intake process so the business can collect the same core admissions information across prospective students.
Introduced progressive database persistence so student applications can continue beyond individual server sessions instead of being lost when the backend restarts.
Added a student confirmation stage to improve the quality of information handed over to the admissions team.
Established a direct integration between WhatsApp Business, Meta's Cloud API, the ASP.NET Core backend and the SWS admissions database.
Created the technical foundation for AI-generated advisor assessments and future Germany and Italy university or programme recommendations.
Established the foundation for an internal admissions workspace where SWS can manage student profiles, assessments, recommendations and application progress from one system.
Conclusion
The SWS Admissions Assistant demonstrates how automation can improve a real business process without forcing customers into an unfamiliar experience. For prospective students, the interaction remains simple: open WhatsApp, answer a series of questions, review the information and submit the assessment. Behind that conversation, however, the system transforms each response into structured admissions data, maintains the student's progress, stores the application persistently, and prepares the profile for the SWS team. This reduces the amount of repetitive information gathering required from advisors and creates a more consistent starting point for every consultation. The project is being developed as more than a chatbot; it is becoming the intake layer for a broader AI-assisted admissions workflow. Ongoing development is focused on advisor-side AI assessment, programme and university recommendations, and an internal workspace that will allow SWS to move from incoming enquiry to informed admissions guidance through one connected system.