MedDirect Supply Case Study

The Challenge

Every logistics operation has a weak link. For MedDirect Supply, it was a Microsoft Access database built in 2003 by a now-retired office manager named Gerald. The database contained delivery schedules, customer addresses, driver assignments, dock instructions, and facility contact information for every hospital, clinic, and surgical center MedDirect served across the Philadelphia metro area. Gerald had maintained and updated it for 17 years before retiring. When he left, he handed Sharon Patel a spiral notebook with handwritten instructions for running the daily queries.

Sharon was the only person who understood how the database worked. She ran it every morning at 5am, exporting delivery lists that drivers printed and carried in clipboards on their dashboards. The Access database did not optimize routes. It grouped deliveries by facility and assigned them to drivers based on zones Gerald had drawn on a paper map taped to the wall in 2005. Drivers decided the actual driving order themselves.

The Access database produced delivery lists. It did not produce efficient routes. Drivers received their stop lists and self-routed based on personal preference. Some prioritized the easy suburban stops first and left the congested hospital dock deliveries for the afternoon when loading docks were busiest. Others ran their routes geographically but ignored the delivery time windows that hospitals required.

Hospital dock delivery windows were not suggestions. They were contractual obligations. Major facilities like Penn Medicine, Jefferson Health, and Temple University Hospital required suppliers to deliver within specific time windows, typically 60-90 minute slots assigned by the receiving department. Missing a window meant the delivery was turned away at the dock and had to be reattempted the following day. Repeated violations triggered financial penalties.

• 12% delivery window violation rate: Out of 150-200 daily deliveries, roughly 18-24 missed their assigned windows. Some were early, arriving when dock staff weren’t ready. Most were late, caught behind drivers who had prioritized easier stops.
• 60+ penalty incidents per month: Hospitals charged $50-$150 per violation depending on the facility and severity. Monthly penalties ranged from $3,000 to $9,000, totaling approximately $45,000 per year in avoidable costs.
• No priority differentiation: A routine supply restock for a small clinic and an urgent surgical instrument delivery for a hospital operating room received the same treatment in the Access database. Both went on the driver’s list. Both were delivered whenever the driver got around to them.
• Single point of failure: Sharon was the only person who could operate the Access database. When she took a week of vacation, a colleague attempted to run the morning export and accidentally deleted three months of delivery records. The backup was a USB drive that hadn’t been updated in six weeks.

The penalty costs were the most visible problem, but the reputational damage ran deeper. MedDirect had bid on supply contracts with two large hospital systems in the past year. Both times, the hospitals asked for on-time delivery metrics. MedDirect’s 88% on-time rate was below the 95% threshold both hospitals required. They lost both bids.

The Solution

Sharon’s search for a replacement started with two requirements. First, it had to handle strict time windows for hospital deliveries, not as a nice-to-have feature but as the core logic driving route construction. Second, it had to accept CSV imports so she could transition away from the Access database without rebuilding 20 years of customer data from scratch.

She tested Upper with a pilot group of six drivers over two weeks. The pilot confirmed that Upper’s optimizer could build routes that respected individual facility time windows while still minimizing total drive time across the fleet. She expanded to all 22 drivers the following Monday.

Replacing the Access Database Without Losing 20 Years of Data

The transition from Gerald’s Access database to Upper began with a data export. Sharon pulled the complete customer list, including addresses, dock instructions, facility contacts, and delivery window requirements, into a CSV file. She imported the CSV into Upper, mapping columns to the appropriate fields. The 3,200 facility records transferred in a single upload.

The critical addition was time window data. Sharon worked with each hospital’s receiving department to confirm their current dock windows, then entered these as hard constraints in Upper. Every facility now had a documented delivery window that the route optimizer would respect as a non-negotiable requirement.

The single point of failure disappeared immediately. The Access database ran on one machine, maintained by one person. Upper ran in the cloud, accessible to any authorized team member. Sharon trained two colleagues on the daily workflow in a single afternoon. When she took vacation the following month, routes ran without a single issue.

Priority-Based Routing for Hospital Deliveries

Not every delivery carried the same urgency, and MedDirect’s old system treated them all identically. Upper’s route optimization allowed Sharon to assign priority levels to each stop. Hospital deliveries were tagged as Critical. Surgical instrument deliveries received the highest priority. Routine clinic restocks were tagged as Standard.

The optimizer built routes that placed Critical stops early in the sequence, ensuring drivers reached hospital docks within the required windows. Standard deliveries filled the remaining time. If a route was overloaded, the optimizer pushed Standard stops to another driver rather than risk a Critical delivery arriving late.

This priority logic was the single largest factor in reducing delivery window violations. Drivers no longer decided the delivery order. The algorithm did, and it prioritized the stops where lateness carried financial and contractual consequences.

Proof of Delivery for Compliance Documentation

Medical supply delivery requires documentation that goes beyond a signature on a clipboard. Hospitals need records of what was delivered, when, and who accepted it. Insurance audits, accreditation reviews, and supply chain compliance checks all require traceable delivery records.

Upper’s proof of delivery feature captured a signature from the receiving dock staff, a timestamped photo of the delivered items, and any notes about the condition of the shipment. Every delivery record was stored in Upper’s dashboard with the facility name, driver, time of delivery, and confirmation details.

The dock notification system also improved the receiving process at hospital facilities. Dock coordinators received an alert when a MedDirect driver was 15 minutes away, giving them time to clear dock space and assign receiving staff. The notification reduced dock wait times because staff were ready when the driver arrived, rather than scrambling to process an unannounced delivery.

Sharon used Upper’s route analytics to generate monthly compliance reports for each hospital client. The reports showed on-time delivery rates, average delivery times, and any exceptions. Hospitals that had previously requested manual performance reports now received automated summaries that demonstrated MedDirect’s reliability with data, not promises.

The Impact

The numbers told the story clearly. Delivery window violations dropped from 12% to 1.8% within three months of full implementation. The remaining 1.8% consisted almost entirely of situations outside MedDirect’s control: dock closures due to emergency procedures, facility construction blocking loading areas, and weather delays. The violations caused by poor routing and driver self-sequencing were effectively eliminated.

Penalty costs fell from $45,000 annually to under $5,000. That $40,000 savings went directly to the bottom line, but the larger financial impact came from what the improved metrics enabled. Six months after implementation, MedDirect rebid on a contract with one of the hospital systems that had previously rejected them. This time, their on-time delivery rate was 98.2%. They won the contract.

Three months later, a second hospital system approached MedDirect based on a recommendation from the first. The two new contracts together represented more than $500,000 in annual revenue, business that would not have been possible with an 88% on-time rate.

Route efficiency improved across the board. The average driver saved 50 minutes per day because routes were optimized for geography and time windows rather than driver preference. Those 50 minutes translated to either earlier completion times or capacity for additional stops, depending on the day’s volume.

Performance Metrics

The operational resilience was perhaps the most underappreciated improvement. MedDirect no longer depended on a single person or a single machine to generate daily routes. Three team members could run the morning dispatch process. The customer data lived in the cloud with automatic backups. The spiral notebook Gerald left behind was archived in a drawer, a relic of a system that had served its purpose but couldn’t carry the business forward.

Sharon retired Gerald’s Windows XP machine on a Friday afternoon. She didn’t hold a ceremony, but she did take a photo. After 20 years, the blinking cursor on the Access database startup screen was finally dark.