case studies
Legacy System Integration
Problem: A health insurance broker was spending up to two hours per PDF manually extracting data and re-entering it into a proprietary legacy system.
Solution: A fully self-hosted setup with a custom integration layer to extract data from incoming PDFs, generate reports, and file documents automatically - with no data leaving the client's own infrastructure.
Result: Full data compliance, minimal costs with no subscription fees, and 10+ hours of manual work saved each week.
Airtable Operations & Finance System
Problem: An alternative provision school was running its entire operation across 12-15 Excel spreadsheets, manually entering the same data in multiple places. Nothing was connected, everything was duplicated.
Solution: A single Airtable system covering student sessions and timetables, staff scheduling and availability, invoicing, pay runs, and campus capacity - with operations and finance interfaces, management dashboards, and 20 automations handling the work that used to be done by hand.
Result: One source of truth for the entire operation, with manual data entry almost entirely eliminated.
Bi-directional Notion Sync
Problem: Before Notion introduced a more granular permissions structure, a client needed each team member to work in their own private Notion workspace while the owner maintained a single source of truth across all of them.
Solution: Two Make scenarios built to sync data both ways via the Notion API, with strict loop prevention to ensure changes propagate without cascading or duplicating.
Result: Full team autonomy with centralised visibility - private workspaces, zero data conflicts.
End-to-End Lead Discovery System
Problem: A growing agency was spending hours each week manually finding, verifying, and logging prospects before any outreach could begin.
Solution: A scalable lead discovery pipeline that sources prospects from public web data, verifies contact details automatically, and pushes structured records directly into the CRM - with Airtable managing the query layer and routing logic throughout.
Result: A fully automated research layer that runs continuously, with no manual prospecting required.
Have a system that needs fixing?
Let's Talk