Cut manual booking work 90% and time-to-bid 12x with a custom staffing platform
I built a platform that monitors several hospital portals, aggregates every open shift, matches and books nurses automatically against the agency's criteria, runs nurse communication on WhatsApp, and gives the team one place to manage everything else.
Concept
The client is a UK nursing agency from the 90s. They supply temporary nurses and healthcare assistants to public hospitals and private facilities across the country.
A large part of the job is booking nurses onto open shifts across several hospital platforms. Each booking has to fit the nurse’s availability, specialties, location, preferences, and experience.
They found me on Upwork. They wanted to know how much of this could be automated, and to build a platform for it.
Challenge
The work was manual, and it ran non-stop:
Approach
Before writing a single line of code, I invested time in truly understanding the business.
-
Discovery. I interviewed the stakeholders and team members doing the bookings. I wanted to understand the process, pain points, current workarounds, and edge cases in detail.
-
Research. I spent a week navigating hospital platforms, checking what each one exposed, and testing different scraping and browser automation approaches. I mapped out which features would deliver the most impact and which weren’t worth automating.
-
Prototype. Within a week I built a frontend prototype. The team clicked through it so we could check the flow before I built the full system.
-
Iterative build. Based on that feedback, I finalized features and design. Then I built the platform in close collaboration with their team, changing things as we went.
Solution
I built a custom platform that covers every stage of the booking process - from finding a shift to talking to the nurse.
flowchart TB
P[Several hospital<br/>portals] --> I[Per-portal<br/>integrations]
I --> S[(All open shifts<br/>in one place)]
S --> M{Match}
M -->|Full match| A[Booked<br/>automatically]
M -->|Partial match| W[WhatsApp bot<br/>offers the shift]
M -->|Rest| T[Team books,<br/>single or bulk]
A --> N[WhatsApp thread,<br/>email notifications]
W --> N
T --> N
style P fill:#e3f2fd,stroke:#1565c0
style I fill:#e3f2fd,stroke:#1565c0
style S fill:#ede7f6,stroke:#4527a0
style M fill:#fff8e1,stroke:#f57f17
style A fill:#e8f5e9,stroke:#2e7d32
style W fill:#e8f5e9,stroke:#2e7d32
style T fill:#fce4ec,stroke:#ad1457
style N fill:#eceff1,stroke:#37474f
Automated shift discovery
The platform checks every connected hospital platform every 5 minutes, using custom integrations, browser automation, and scraping. New shifts show up immediately in one dashboard, filterable by date, hospital, department, profession, and status. Nobody sits there clicking through portals.
Smart nurse matching and auto-booking
For every shift, the platform shows relevant nurses from their internal system. It also shows each nurse’s surrounding shifts to see whether they’re free.
If a nurse is a 100% match (availability, specialty, location), the platform books them and notifies the team.
One-click and bulk booking
When staff still book by hand, they pick nurses and book with rates prefilled from historical data. Each hospital platform is handled behind the scenes, with a dedicated integration for each.
To book nurses across several shifts at once, the bulk view shows availability and experience on every selected shift, and a button to book them all in one click.
Full WhatsApp chat system
I built a WhatsApp communication system and replaced the messaging software they were already using.
The built-in chat has templates for availability checks and shift confirmations, automatic notifications, file and voice messages, and AI help for drafting replies and summarizing threads. Shift confirmation messages come prefilled and can be sent from the shift page.
There’s also a WhatsApp bot that messages nurses who aren’t a 100% match. It tells them about the opening, asks them to confirm, and lets them book with one tap. That cuts the manual communication and gets bids in much faster.
Automated notifications
The right people on the team get emails on new shifts, bid updates, confirmations, and status changes. Nobody has to watch for these or forward them.
Impact
Before: 3 people working 24/7, in shifts. Checking portals, matching nurses, copy-pasting between systems, and hopping between chat, CRM, and hospital sites.
After: 1 person spends 1-2 hours a day, mostly checking that everything is running smoothly and dealing with edge cases.
-
~90% less manual staffing work. Shift discovery, nurse matching, and most of the booking happen automatically. The team is no longer on 24/7 manual work.
-
12x faster time-to-bid. Average time-to-bid went from about 1 hour to under 5 minutes. The old delay was slow detection and copy-pasting bid details. Detection and submission are now automatic.
-
More shifts won. Qualified nurses get submitted faster, and more of them get submitted, so the platform wins more shifts than the manual process did.
-
One place to work. Shifts, nurses, bookings, and chat live in one platform. Staff don’t jump between hospital portals, CRM, and a separate messaging tool.
* Client and company details have been anonymized on their request.