Testing a food delivery app is hard. Testing an on-demand logistics app is harder.
Food delivery has a fixed pattern: one restaurant, one customer, one delivery partner, 30 minutes. Logistics apps break every one of those constraints. Multiple pickup stops. Multiple drop locations. Vehicle selection based on package size bike, auto-rickshaw, mini-truck, or 3-wheeler. Weight and dimension-based pricing that recalculates mid-booking. Driver allocation across vehicle categories. Package photo verification at pickup and delivery. OTP-based handoffs. And live tracking across routes that can stretch 3 hours across a city instead of 30 minutes across a neighbourhood.
India's on-demand logistics market is growing at 25%+ annually. Platforms are processing hundreds of thousands of parcels daily across 50+ cities from a document courier within downtown Manhattan to a 50kg furniture shipment from Los Angeles to San Francisco. In India, the same scale plays out across cities like Bangalore, Delhi, and Mumbai. The QA challenge scales with the complexity.
This guide provides 28 ready-to-use test cases covering every critical flow in an on-demand logistics app, from booking to delivery confirmation. Each test is written in both Appium and Drizz format so you can see the difference in approach and maintenance cost.
For food delivery testing, see our 30 Test Cases from Order to Doorstep. For the broader delivery app challenge, see Why Delivery Apps Are the Hardest to Test.
Key Takeaways
- On-demand logistics apps are harder to test than food delivery because of multi-stop routes, vehicle type selection, weight-based pricing, package verification (photo + OTP), and tracking across city-scale distances.
- 28 test cases cover the complete logistics flow: booking (8), driver matching (4), pickup verification (4), live tracking (4), delivery confirmation (4), payment (4), and cancellation/refund (4).
- Selector-based tools (Appium) struggle with logistics apps because of dynamic fare estimates, map-heavy flows, photo verification screens, and OTP dialogs that cross app boundaries.
- Vision AI (Drizz) validates the visual experience fare estimates displayed, map pin moving, photo upload confirmed, OTP screen rendered without depending on element IDs that change with every release.
- Region-specific challenges vary by market: in the US, challenges include apartment complex access codes, loading dock scheduling for commercial deliveries, toll route pricing, and multi-timezone scheduling for cross-state shipments. In India, additional challenges include auto rickshaw as a vehicle category, addresses without pin codes in tier-3 cities, intercity parcel flows, and COD for logistics.
How Is Logistics App Testing Different from Food Delivery Testing?
Section 1: Booking Flow (Test Cases 1-8)
TC-01: Enter pickup address
Appium:
Drizz:
TC-02: Enter drop address
Drizz:
TC-03: Add multiple stops (multi-drop)
Drizz:
TC-04: Select vehicle type
Appium:
Drizz:
TC-05: Fare estimate calculation
Drizz:
TC-06: Weight and dimension input
Drizz:
TC-07: Schedule a future pickup
TC-08: Intercity parcel booking
Drizz:
Section 2: Driver Matching and Acceptance (Test Cases 9-12)
TC-09: Booking confirmation and driver search
Drizz:
TC-10: Driver assigned with details
Drizz:
TC-11: No driver available handling
Drizz:
TC-12: Driver cancels after accepting
Drizz:
Section 3: Pickup Verification (Test Cases 13-16)
TC-13: OTP verification at pickup
Appium:
Drizz:
TC-14: Package photo at pickup
Drizz:
TC-15: Pickup address reached by driver
Drizz:
TC-16: Wrong pickup location handling
Drizz:
Section 4: Live Tracking (Test Cases 17-20)
TC-17: Map loads with route
Drizz:
TC-18: Driver pin moves on map
Drizz:
TC-19: ETA updates during transit
Drizz:
TC-20: Multi-stop tracking transition
Drizz:
Section 5: Delivery Confirmation (Test Cases 21-24)
TC-21: OTP verification at delivery
Drizz:
TC-22: Delivery photo confirmation
Drizz:
TC-23: Delivery to alternate recipient
Drizz:
TC-24: Failed delivery attempt
Drizz:
Section 6: Payment (Test Cases 25-28)
TC-25: Pay with UPI after delivery
Drizz:
TC-26: Cash on Delivery completion
Drizz:
TC-27: Corporate account billing
Drizz:
TC-28: Fare adjustment and dispute
Drizz:
The Pattern: Why Vision AI Fits Logistics Apps
Logistics apps have more visual, dynamic, and cross-boundary flows than food delivery:
- Vehicle selection is a visual grid of options with images, names, and prices not a standard dropdown Appium can easily select from.
- Fare estimates update dynamically based on distance, weight, vehicle, and time static assertions break instantly.
- Photo verification at pickup and delivery renders images inside the app that Appium can't validate (it can check if an image element exists, not if it shows the right package).
- OTP screens may be in-app or OS-level Vision AI sees both.
- Multi-stop tracking requires visual validation that the map route, status text, and ETA all update correctly at each stop transition.
- Addresses without pin codes in tier-3 cities often use landmark-based descriptions that render as long text strings Vision AI confirms the text renders without truncation.
At 28 test cases with an average of 8 selectors each, an Appium suite has 224 breakage points. A Drizz suite has zero. For logistics apps that iterate on booking UI and tracking flows weekly, the maintenance gap compounds every sprint.
After 3 months of weekly releases, a selector-based suite has faced 12 rounds of potential breakages across those 224 points. The QA team that started with 28 well-maintained tests is now spending more time fixing broken selectors than writing tests for the new features shipping every sprint scheduled pickups, intercity routes, corporate billing, package insurance. Coverage stalls. The newest, highest-risk flows ship untested.
A Vision AI suite faces those same 12 releases and needs zero selector updates. The team spends that same time expanding coverage to 50, then 80, then 120 tests covering every new feature as it ships. The architecture doesn't just reduce maintenance. It changes what the team is capable of building.
Frequently Asked Questions
How is logistics app testing different from food delivery testing?
Logistics apps add vehicle selection, weight-based pricing, multi-stop routing, package photo verification, OTP handoffs at both pickup and delivery, scheduled bookings, intercity routes, and corporate accounts. Food delivery has a simpler fixed pattern (one restaurant, one customer, 30 minutes). Logistics has variable routes, variable durations (30 min to 3+ hours), and variable package types.
Can Appium test photo verification flows?
Appium can verify that an image element exists in the element tree, but it cannot verify what the image shows. A pickup photo showing the wrong package, a blurry image, or a placeholder icon all pass Appium's is_displayed() check. Vision AI can confirm that an image is present, appears to be a photo (not a placeholder), and the surrounding context (timestamp, "Picked up" label) renders correctly.
How do you test multi-stop routes?
Test each stop transition independently: verify tracking shows the correct current destination, verify ETA updates when a stop is completed, verify the map route changes to the next stop, and verify the status text transitions correctly. Use a test API to trigger stop completions without waiting for real deliveries.
What about testing auto-rickshaw as a vehicle option?
Auto-rickshaw is a vehicle category specific to Indian logistics apps. Test it the same as any vehicle: verify it appears in the vehicle selection grid with an image and estimated price, verify selecting it updates the fare, and verify the booking confirms with the correct vehicle type. Vision AI sees "Auto" as a visual option regardless of what element ID the component uses.
How many devices should logistics apps test on?
Logistics apps should test on 25-40 devices with emphasis on mid-range and budget Android phones (Samsung A-series, Xiaomi Redmi, Realme) because delivery partners and many customers in tier-2/3 cities use budget devices. Include 3-4GB RAM devices, older Android versions (12-13), and smaller screen sizes where booking forms and tracking maps are most likely to have rendering issues.


