Inspiration
Adviance was inspired by a simple problem: advisors already use many tools, but those tools do not work together. Client notes may be in a CRM, schedules in a calendar, CPD courses in a learning platform, and specialist contacts in spreadsheets or personal messages.
The issue is not the lack of tools. The issue is fragmented information.
We wanted to create one connected workspace where an advisor can manage client context, learn based on real needs, and find the right authorised partner without manually connecting different systems.
What it does
Adviance is a connected advisory workspace built around one main idea: Client Memory.
Each client has one persistent profile containing meeting notes, follow-ups, life events, referral history, and previous interactions. This information is stored in the backend database.
The platform has four main modules:
- Client Memory Dashboard: manages client profiles, meetings, follow-ups, expenses, and notes.
- Learning Hub: tracks CPD progress and recommends approved courses based on client needs, completed learning, and advisor specialisation gaps.
- Partner Finder: helps advisors find suitable authorised partners for needs such as tax, estate planning, Takaful, retirement, and business succession.
- Growth Dashboard: shows transparent metrics for Client Engagement, Deadline Reliability, Partner Activity, and Practice Score.
Gemini supports the advisor by organising rough notes, recommending approved learning content, explaining partner matches, and drafting introduction messages. The advisor reviews every result before acting.
How we built it
We built Adviance as a full-stack web application.
The frontend was developed using React and Vite. The backend was built using Node.js and Express, with REST APIs connecting the frontend to the backend.
For the MVP, we used a local JSON-based database to store client records, notes, referrals, user accounts, and learning data. This allowed us to demonstrate persistent Client Memory safely using fictional data.
We integrated Google Gemini through the backend only. The Gemini API key is stored in the backend .env file and is never exposed in the browser.
We also added login and registration so each advisor can access their own workspace.
Challenges we ran into
One of our biggest challenges was deciding which ideas were realistic to build within the hackathon time limit.
During brainstorming, we came up with many possible solutions, but some were too complex to implement properly in a short time. We had to focus on the features that created the clearest value: Client Memory, relevant learning recommendations, authorised partner matching, and transparent growth tracking.
Another challenge was balancing AI support with advisor control. We wanted Gemini to save time, but we did not want it to make financial decisions, give investment advice, or automatically change client data.
We solved this by making every AI feature review-first. Gemini creates a draft, summary, recommendation, or suggestion, but the advisor always makes the final decision.
We also faced technical challenges while connecting the frontend, backend, login system, data storage, and Gemini API. These challenges helped us improve our teamwork, debugging process, and feature prioritisation.
Accomplishments that we're proud of
We are proud that we created a working MVP instead of only a concept.
Our strongest accomplishment is building one shared Client Memory that connects multiple advisory activities. A saved client conversation can support meeting preparation, CPD recommendations, authorised partner matching, and growth insights.
We are also proud that we used AI responsibly. Gemini is not treated as the source of truth. The backend database stores the real information, while Gemini only supports the advisor when requested.
Another achievement is that we made our AI recommendations safer by limiting them to approved CPD courses and authorised partners instead of open internet results.
What we learned
We learned that AI is most useful when it supports people instead of replacing their judgment.
We learned that the database should remain the source of truth, especially when dealing with client information. AI should help organise, summarise, and recommend, but it should not make final professional decisions.
We also learned that transparency matters. That is why our Growth Dashboard uses clear activity-based metrics instead of hidden black-box scores.
Most importantly, we learned that a strong product is not about adding many features. It is about connecting the right features in a way that solves a real workflow problem.
What's next for Adviance
The next step for Adviance is to move from a local MVP into a scalable production platform.
Future improvements include:
- Migrating from the local JSON database to a secure cloud database
- Adding role-based access for advisors, managers, and administrators
- Integrating with real CRM, calendar, and CPD systems
- Adding audit logs and stronger data privacy controls
- Creating organisation-level analytics dashboards
- Allowing advisory organisations to manage their own approved courses and partner directories
- Running a pilot with a small advisor team to measure time saved, CPD engagement, and referral outcomes
Our long-term goal is to make Adviance a trusted connected intelligence platform for advisory organisations.
One Client Memory. Smarter Advisory Decisions.
Built With
- api
- css
- git
- html
- javascript
- react
- vite
Log in or sign up for Devpost to join the conversation.