BloomDash Flowers Case Study

The Challenge

Mei Chen opened BloomDash Flowers because she loved flowers. Six years later, she spent more time planning delivery routes than she did arranging bouquets.

On a normal day, BloomDash handled about 50 deliveries across Portland. Mei’s eight regular drivers knew the city well, and the handwritten route cards she prepared each evening worked adequately. Each card listed the driver’s stops in order, with addresses, recipient names, and any delivery notes. The system was manual but manageable for 50 stops split across eight people.

Holidays were a different story. Valentine’s Day volume jumped to 200 orders. Mother’s Day hit 250. BloomDash hired 12 temporary drivers to handle the surge, bringing the total fleet to 20. And every route for every driver was planned by Mei, by hand, the night before.

Valentine’s Day 2025 was the breaking point. Mei started building routes at 8pm the night before and finished at 1:30am. She sorted 200 orders by neighborhood, assigned them to 20 drivers, wrote out route cards, and tried to account for the one factor that made floral delivery uniquely time-sensitive: workplace deliveries had to arrive before 5pm, and many needed to arrive before recipients left for lunch.

The results were predictable:

• 18% of Valentine’s Day deliveries arrived late: 36 arrangements missed their time windows. Workplace deliveries arrived after the recipient had left for the day. Surprise bouquets were delivered to empty offices.
• Three one-star reviews posted in a single day: “Ordered flowers for my wife at her office. They arrived at 5:45pm. She’d been home for an hour.” | “The arrangement was beautiful but showed up three hours after the window I paid for.” | “Not using this florist again.”
• Temporary drivers getting lost and calling for directions: Temp drivers unfamiliar with Portland’s quirks (one-way streets downtown, bridge traffic patterns, addresses in the West Hills) called Mei repeatedly for help. She spent the day answering her phone instead of managing the shop.
• No confirmation that deliveries were completed: Several customers called asking if their order had been delivered. Mei had no way to confirm without calling the driver, who may or may not have answered.

The one-star reviews stung because they were avoidable. The flowers were beautiful. The arrangements were exactly what customers ordered. The delivery logistics were what failed, and for a gifting business where timing is the entire point, that failure was existential.

Mei knew she couldn’t survive another holiday season running routes by hand. Mother’s Day was three months away, and she expected even higher volume. She needed a system that could handle 20 drivers, enforce time windows, and work for temporary staff who had never driven a Portland delivery route before.

The Solution

Mei found Upper in March, six weeks before Mother’s Day. She tested it with her regular daily deliveries first, importing her 50 daily orders via CSV with columns for address, recipient name, delivery notes, and time window.

The time windows were the critical piece. Mei categorized every delivery into one of two windows: workplace deliveries (deliver before 5pm, ideally before 2pm) and home deliveries (deliver before 8pm). Upper’s route optimization sequenced stops so that workplace deliveries were prioritized in the earlier part of each route, with home deliveries filling the afternoon and evening.

Scaling to 20 Drivers for Valentine’s Day

When Valentine’s Day planning started, Mei created driver profiles for all 20 drivers: 8 regulars and 12 temps. She uploaded the full order list, tagged each delivery with its time window, and ran the optimization. Upper distributed 210 deliveries across 20 drivers, respecting every time constraint and balancing the stop count so no driver was overloaded.

The route planning that had consumed Mei’s evening until 1:30am the previous year took 45 minutes. She reviewed the routes, made two minor adjustments for customer-specific access instructions, and dispatched them to all 20 drivers through the app.

The temporary drivers were the biggest concern. These were gig workers and part-time staff who didn’t know Portland’s delivery patterns. In previous years, they relied on Mei’s handwritten cards and their own phone GPS, making wrong turns, missing building entrances, and calling for help.

With Upper’s driver app, every temp received a sequenced route with turn-by-turn navigation built in. They followed the app from stop to stop without needing to interpret handwritten cards or figure out their own routing. The app guided them to each address, displayed the recipient name and delivery notes, and moved to the next stop automatically after completion.

Delivery Photos as a Competitive Advantage

Mei initially implemented Upper’s proof of delivery to solve the confirmation problem. When a customer called asking if their bouquet had been delivered, she wanted a timestamped photo to reference instead of calling the driver.

What she didn’t expect was the marketing value. Drivers photographed each arrangement at the recipient’s door, desk, or reception area. The photos showed BloomDash’s flowers in real-world settings: on office desks, at restaurant hostess stands, on front porches in the rain. Mei started sharing these photos (with customer permission) on BloomDash’s Instagram, and they became some of the account’s most-engaged content.

The delivery notifications added another layer. Senders received an automatic notification when their order was delivered, along with a tracking link showing the driver’s progress. For gift purchases, knowing exactly when the bouquet arrived was part of the experience.

The Mother’s Day Test

Mother’s Day 2025 was the real validation. Volume hit 250 deliveries, the highest single day in BloomDash’s history. Mei uploaded the orders, ran the optimization across 20 drivers, and dispatched routes in under an hour. Every delivery was completed within its time window. Zero complaints. Zero one-star reviews.

The temp drivers performed at the same level as regulars. Their routes were sequenced identically, their app experience was identical, and their delivery times were comparable. The performance gap between experienced and temporary staff, which had been a major source of quality issues, disappeared when everyone followed the same optimized route.

The Impact

BloomDash Flowers’ transformation centered on one shift: replacing manual, overnight route planning with algorithmic optimization that took minutes. The impact was felt most intensely on peak days, but the daily operation improved as well.

Valentine’s Day went from 18% late deliveries and three one-star reviews to zero late deliveries and 12 new five-star reviews. The reviews specifically mentioned delivery timing and the photo notification feature, both of which were direct results of the routing system.

Mei’s workload changed dramatically. Route planning dropped from a multi-hour evening task to a 45-minute morning routine. On peak days, the time savings were even more significant. The midnight sessions before holidays were eliminated entirely.

The temporary driver problem was solved not by finding better temps but by giving them better tools. When every driver follows an optimized, app-guided route with built-in navigation and delivery instructions, experience level matters less. A first-day temp and a five-year veteran produce comparable results.

The delivery photo feature created an unexpected competitive advantage. BloomDash now includes a delivery photo with every order as a standard service, something none of their local competitors offer. Customers choosing between Portland florists see it as a differentiator, and several corporate accounts cited it as a reason for switching to BloomDash.

Mei is planning for the next Valentine’s Day with confidence rather than dread. She’s already reserved 15 temporary drivers and will scale routes in Upper the same way she did this year. The process that used to generate her worst workday of the year is now one of her most straightforward.

Performance Metrics