Automation case studies

The problem. Order details landed in an Outlook mailbox and still had to be keyed into the ERP. After an order number existed, the same record had to be stored in Oracle.

What we automated. UiPath reads order data from the mailbox, creates the order in the ERP, and saves the resulting record in the Oracle database.

The problem. New vendors arrived as mailbox requests and still had to be created as business partners in SAP. The client also needed a report of what was processed.

What we automated. UiPath reads vendor data from Outlook, creates the business partner in SAP, and emails the vendor report back to the client.

The problem. Purchase orders and goods receipts both ended in SAP, but the inputs lived in vendor spreadsheets and in Odoo. Someone still created the SAP documents and then emailed the reference numbers.

What we automated. The PO bot creates purchase orders in SAP from vendor Excel files and emails the PO number. The GRN bot creates goods receipt notes in SAP from the Odoo Coil Receipts section and emails the material-document number.

The problem. Checking a large number of shipments on a tracking website and retyping each status into the database was slow and easy to mistype.

What we automated. The robot reads shipment details from the database, enters tracking numbers on the carrier site, retrieves the latest status, and updates the database. The run also produces an updated shipment report.

The problem. Each simulation meant opening PVSYST, importing a CSV, running the model, and exporting Excel. The proof of concept asked whether that sequence could run unattended.

What we automated. UiPath reads the CSV, extracts the required information, launches PVSYST, imports the file, runs the simulation, and generates the output Excel file.

The problem. Market data, open positions, stop-loss orders, and futures approaching expiry all needed attention on a schedule. Doing that by hand meant constant monitoring against rules that were already written down.

What we automated. Four UiPath bots share the work. One maintains a 20-day market report. One places orders and stop-loss instructions when predefined conditions are met. One monitors the market every two minutes in the morning window. One rolls eligible futures into the next month. Trading decisions follow the configured rules.

The problem. Insurance documents arrived as combined PDFs. Splitting them, scanning the contents, and retyping the result into Brio was a handoff-heavy process.

What we automated. The first bot splits each PDF on barcode separators. The second uploads files to the Metamaze portal through an API and pulls extracted information back through an API. The third reads that Excel extract and enters the details into Brio.

The problem. Teams needed editable copies of client documents that arrived as archives or individual files. Manual conversion was slow and inconsistent across formats.

What we automated. A Power Apps upload starts a UiPath process that identifies the file type, uses ABBYY FineReader for scanning and conversion, handles bulk ZIP processing, and returns editable outputs.

The problem. Learning reports lived in six different portals. Someone had to log in, download each file, clean it, and merge the set before the business could use it.

What we automated. UiPath signs in, opens each report section, downloads the files, cleans and standardizes the data, consolidates them, and uploads the processed reports to SharePoint.

The problem. Ripples certificates were built and sent by hand. At a few hundred records a month, creation time and missing status updates both added up.

What we automated. UiPath reads employee records, creates a PowerPoint certificate for each one, sends it, calls an API, and writes a successful status to the database so distribution can be tracked.

The problem. Booking meant reading a sheet, deciding if the patient was new, finding a centre and a free slot, then texting a confirmation. Every step was repetitive and easy to get wrong.

What we automated. UiPath reads patient details from Google Sheets, identifies CDBS, PYWC, or HF, checks whether the patient is new or existing, creates a file if required, finds a free slot at the stated centre and date, books it, sends a Twilio SMS for that patient type, and generates a report.

The problem. Preparing a report meant opening the patient file, checking allergies and earlier reports, and applying the doctor’s notes. The format drifted, and the review took time.

What we automated. UiPath reads the sheet, opens the patient file, collects allergies, earlier reports, and doctor instructions, sends that material to an approved AI service for a structured draft, checks the draft against predefined rules, saves it on the patient file, and updates the tracking sheet. The draft is not a final clinical record until a person reviews it.

The problem. Someone had to scan the appointment list, work out which visits were 7, 3, or 1 day away, and send each reminder. Missed messages were easy to produce.

What we automated. UiPath reads the Excel appointment report, selects the reminder window, finds matching visits in the portal, retrieves patient or parent contact details, sends the message, and updates the sheet with success and failure logs.

The problem. Discharges can be due from the insurance company before the payment arrives. Booking those pending items in Brio by hand was easy to delay and easy to miss.

What we automated. UiPath reads the Excel input and books pending discharges in the Brio application, covering records that are expected from the insurer but have not been received.

The problem. Reminders needed the correct attachment for each row in an Excel file. Assembling those emails inside Brio was repetitive clerical work.

What we automated. UiPath reads reminder details from Excel, attaches the required files, and sends the reminder through the Brio application.