Financial advisory system
A no-code ERP for a global network of financial advisors, built under NDA. It runs partner accounts, client contracts, commissions and internal reporting for a network of 10,000+ clients.
A no-code ERP for a global network of financial advisors, built under NDA. It runs partner accounts, client contracts, commissions and internal reporting for a network of 10,000+ clients.
The company runs a network of 95 financial consultants managing 15,000 contracts and 10,000 clients. Back-office staff tracked commissions, qualifications and payouts largely by hand, and building a report meant assembling an XLSX file one at a time.
The client is under NDA, so the system had to handle real financial data - transactions, commissions, contracts - without the company's name ever appearing in the build. It runs on Directual for the backend and Bubble.io for the front end, with security requirements to match: encryption at rest, role-based access, two-factor login.
What I owned
No-code architecture design, backend logic and the data model in Directual, the Bubble.io front end, every integration (Directual API, N8N, Google Sheets, DashaMail, GetCourse, Inssmart), and the security and compliance design.
Who else was involved
The company's own back-office team; further detail on team structure falls under NDA.
How long it ran
May 2024 to June 2025.
Built a multi-branch N8N workflow that automates the monthly import of 5,000-6,000 transactions from 20+ providers. Each provider sends an XLSX file of 1,000-3,000 rows; N8N reads it from a Google Sheets staging layer, routes it across 11 provider-specific branches, applies the transformations each source needs, and posts the result to Directual through its API.
Every run carries its own error handling, status tracking and import log, so a failed batch shows up on its own instead of going missing in the numbers.
Five report types - payment registry, qualifications, commissions, revenue and expenses, partner accruals - used to mean opening an XLSX template and filling it manually. Now Directual hands the data to N8N, which sorts it, converts the numeric fields, fills single- and multi-page templates, and saves the output. One click in Bubble is all the back office does now.
Some back-office work still ran through spreadsheets, so clients, contracts, transactions and consultant records export to Google Sheets automatically the moment a new record is created in Directual - no separate sync step for anyone to remember.
Data sits in Directual on MongoDB and S3, encrypted with AES-256 at rest and TLS in transit, behind Cloudflare with strict HTTPS. Access runs through role-based access control, IAM keys and two-factor authentication, with every data flow mapped and documented for compliance.
90%
Manual partner workflow automated
15,000
Contracts under management, 10,000 clients
~100
Daily users across partners and back office
Firstly, Julia joined my project on the recommendation of my developer, which already shows that she’s someone you can rely on.
Secondly, I was impressed by the level of responsibility Julia demonstrated from the very beginning. I used to think you always had to constantly supervise people to make sure they did their job. Julia showed me that there are people who can carry out tasks without anyone standing over them with a stick.
Another strength is her willingness to learn new things and listen to feedback. I believe we worked very well together by exchanging constructive feedback and sharing knowledge - not only technical, but also general. She always listens carefully and adapts her work accordingly.
We faced many technical challenges on the project, and I believe Julia learned a great deal, since the ERP system we developed was far more complex than anything she had worked on before. She accepted the challenge, asked many questions, dove deep into technical details, and learned the principles of development and basic systems analysis.
We stopped working together because the next stage for my client required moving to a code-based product instead of no-code or low-code solutions. When I have another project in the no-code space, I will definitely return to Julia, and I can confidently recommend her for your projects.