Skip to content
Log in

Test fulfillment stations

On this page

Misconfigured station routing never raises an error. An item that cannot be routed is simply left unassigned, and a docket that cannot print is simply not printed. That makes deliberate testing the only way to find a configuration gap before service does.

Use this process in a sandbox or a test outlet, in the order given. Each test builds on the one before it, so a failure early on explains the failures after it.

Before you start

:::tip Test with two devices side by side wherever you can. Many station behaviors, including cover, fallback recipients, and printer assignments, are only observable from a second device. :::

Test 1: Confirm the configuration reached the POS

  1. Sign in to a register at the outlet and run a sync.
  2. Check the sidebar for a Stations entry.

The entry appears automatically once the outlet has at least one Fulfillment Station record. If it is missing, the station records have not synced to the device yet, or they are attached to a different outlet.

  1. Open Stations and confirm the stations listed are the ones whose Register is this register.
  2. Open a station and confirm its Fallback, Printer or Printers, and the product categories that route to it all read as you configured them.

Test 2: Print a sample ticket

  1. Go to Settings, then under Printers select Printer templates, then Fulfillment template.
  2. Confirm a template is selected, with a green tick beside it.
  3. Select Run test print.

A sample docket prints, headed Test Station, with two test products. This proves the template renders, the printer is reachable, and the printer profile matches the hardware, before any real order is involved.

:::note The sample is not a real fulfillment, so it writes nothing to any Shipment record. :::

Test 3: Route one item per category

Take a separate small order on the order-taking register for one product from each fulfillment category, one at a time, and fulfill it.

Using the restaurant example, that means:

Product Category Should print at
Wagyu burger Grill Hot station
Grilled Cos Caesar Cold Food Cold station
Glass of shiraz Wine Wine Tasting Cellar
Pale ale Drinks Bar front station or Bar back station

For each one, confirm the docket prints at the right station without anyone touching the station screen, and that the ticket appears on that station’s screen with status Requested.

Test 4: Fan one order out to several stations

Take a single order containing items from every category, then fulfill it.

Confirm that:

  • One ticket appears per station, not one ticket for the whole order.
  • Each ticket lists only the items for its own station.
  • Every ticket prints at the same moment, on its own printer.
  • The order record on the order-taking register shows a prep status caption under each line.

This test exposes a missing category mapping, because the unmapped line is silently absent from every station.

Test 5: Check what happens to an unmapped product

  1. Pick a product with no Product Fulfillment Category record, or create one for the test.
  2. Sell it and fulfill the order.
  3. In Salesforce, open the resulting Shipment record.

Fulfillment Station is blank. That is correct behavior, not a fault: nothing is lost and no error is raised. What matters is that you know which products this applies to, so decide deliberately whether every sellable product needs a category.

Run a report for fulfillments where Fulfillment Station is blank to find the full set.

:::note A virtual product, such as a fixed-price bundle’s wrapper, is expected to route nowhere. It is not a fulfillable good. Its components route normally, so check the components rather than the bundle anchor. :::

Test 6: Check group load balancing

Only applies if you configured a station group.

  1. Sell several items in the group’s category in separate orders, fulfilling each one.
  2. Watch which child station each ticket lands on.

Tickets spread across the available children rather than piling on one. The child chosen each time is the one with the most spare capacity, measured as its outstanding tickets divided by its Weight, so the split follows the weight ratio rather than strict alternation.

  1. To test weighting, set one child’s Weight to 200 and the other’s to 100, sync, and repeat. The heavier child should take roughly twice the work.

Test 7: Check the fallback chain

  1. On the station’s register, open the station and select Degrade station, then confirm.
  2. Check the banner on the station details screen. It should name where new tickets are going.
  3. Sell an item in that station’s category and fulfill it.

The new ticket appears at the fallback station, and prints there. Existing tickets stay where they are, which is intended: degrading a station never re-routes work already in progress.

  1. On the fallback station’s device, confirm the degraded station now shows as a station it is serving.
  2. Select Restore station and confirm new tickets return to the original station.

Repeat for each station that has a fallback. Follow the whole chain if a station’s fallback has its own fallback.

:::warning A station with no fallback keeps its tickets when it is degraded, and the banner says so. Confirm that is what you want for every station before going live, because those tickets have no other screen to appear on. :::

Test 8: Check a register override

Only applies if you configured a Register Fulfillment Category Station.

  1. Sell the overridden category on the overridden register and fulfill.
  2. Sell the same product on a different register and fulfill.

The first goes to the override station. The second goes to the outlet default. If both go to the same station, check the override’s register and that its station belongs to the same outlet as the order.

Test 9: Walk a ticket through the prep lifecycle

  1. On a station screen, open a ticket with several lines.
  2. Confirm the status pill reads Requested and the waiting time is counting.
  3. Tap the footer button and confirm it advances to the next stage in that station’s flow.
  4. Select Select items, tick one line, and advance.

That line splits onto a new ticket at the next status and the rest stay on the original. Confirm both tickets are visible and the counts on the station card add up.

  1. On a station with a shortened Status Flow, confirm the footer button skips the stages you excluded. A station set to requested;complete should read Mark "completed" on a new ticket.
  2. On the order record, open an item row’s trailing menu and select Mark "<status>". It should open the ticket detail with that line preselected rather than changing the row directly.

Test 10: Check printer assignments on a multi-printer device

Only applies to a device with two or more printers of the template’s printer type.

  1. Open a station, then Assign printers, and assign it one printer.
  2. Assign the other printer to a second station on the same device.
  3. Fulfill an order containing items for both stations.

Each docket comes out of its own printer. If both come out of the same one, check the Printer box on each station’s details screen and confirm the printers were told apart by serial number, using Settings, Manage hardware, then the printer.

  1. Uncheck every printer for one station and confirm its dockets return to the device’s default printer.

Test 11: Check what happens with the printer off

  1. Turn off or unplug a station’s printer.
  2. Fulfill an order for that station.
  3. In Salesforce, confirm the Shipment record has Auto Printed At blank.

Nothing prints and the ticket stays in the queue, which is correct: a docket is never sent to a printer that cannot be confirmed, so it is never printed at the wrong station.

On browser POS with a USB printer, a Reconnect button may appear on the station and printer screens, and a Star mPOP needs it because it reports no USB serial.

What happens to the held ticket when the printer returns is covered in the printer reliability tests below, along with the report that catches a ticket that never printed.

Test 12: Cover a station from another device

  1. Pick a station with no Fallback Station, then degrade it so it needs cover. A degraded station with a fallback sends new orders to the fallback, so steps 4 and 6 would not reach it. Alternatively, clear the station’s Register and set Offline Behavior to accept.
  2. On a different device, select Cover a station and pick it.
  3. Confirm the station joins that device’s list, tagged Covered by you, and that its tickets are visible.
  4. Fulfill a new order for that station and confirm nothing prints on the covering device.

Covering shows tickets, it does not print them.

  1. Select Auto-print new tickets on the covered station.
  2. Fulfill another order for it.

The new ticket prints on the covering device. The tickets that already existed do not, because switching auto-print on stamps a starting point and leaves the backlog alone. That is the behavior to confirm, because it looks like a fault and is not one.

  1. Confirm the covering device’s choices are invisible on every other device.

Test 13: Move tickets between stations

  1. From a station’s details screen, select Move tickets, pick some tickets, and choose a destination.
  2. Confirm the tickets appear at the destination and an alert names how many moved.
  3. Confirm no docket printed at the destination.

A moved ticket does not auto-print again. Reprint it from the ticket detail if the new station needs a docket, then confirm Last Printed At on the Shipment record moved while Auto Printed At did not.

Test 14: Check the print audit data

Run a report on Shipment records from your test orders and confirm each field behaved as expected.

Check Expectation
Fulfillment Station Populated on every routed ticket, blank only where you expect it
Auto Printed At Populated once per ticket, never twice
First Printed At and First Printed Register Populated, and unchanged by later reprints
Last Printed At and Last Printed Register Updated by each reprint
Manual reprint Moved Last Printed At and left Auto Printed At alone

See Track station ticket printing for what each field means.

Printer reliability tests

A station ticket that does not print is an order nobody is preparing, and nothing on the screen shouts about it. Run this set before go-live and again after any change to printers, stations, or templates.

Each test below has a known-good result. Anything else is worth raising.

R1: One ticket, one copy, under load

  1. Place several orders at the same moment on different devices, covering items for more than one station.
  2. Count the dockets at each station.

Exactly one copy of each ticket at its assigned station. Concurrent orders on separate devices do not produce duplicates.

R2: Adding to an order does not reprint what already went

  1. Submit an order with items for two stations.
  2. Add more items for both stations and submit again.

Each station gets one new docket covering only the new items. The first dockets are not reissued.

R3: A designated print device

This is worth setting up deliberately, not just testing. Rather than a device at every station, one device can print for all of them.

  1. Set every station’s Register to the designated device’s register, and assign each station its own printer on that device.
  2. Confirm no other register is named on these stations, and leave auto-print off on every device that covers them.
  3. Place an order spanning several stations.

Only the designated device prints, and each station’s printer gets its own ticket. Other devices show the tickets but print nothing. See Set up station tickets on a register for how to build this.

R4: A USB printer pulled mid-service

  1. Unplug a station’s USB printer.
  2. Fulfill an order for that station.

Nothing prints. The ticket stays in the queue at Requested, and Auto Printed At is blank on the Shipment record.

  1. Plug the printer back in and select Reconnect if a warning offers it.

The held ticket prints once the printer reconnects, and Auto Printed At is now populated on the Shipment record.

:::warning Do not reprint a held ticket by hand while it waits. A manual reprint does not count as the automatic print, so the station gets a second docket when the printer returns. A ticket completed before it printed never prints by itself, so the report in R7 is what protects you, not the reconnect. :::

R5: A network printer switched off

A network printer cannot be polled, so nothing notices until a print is attempted.

  1. Switch off a station’s network printer.
  2. Fulfill an order for that station.

Nothing prints. The print attempt runs for up to two minutes before reporting a timeout, and the ticket stays in the queue.

  1. Switch the printer back on, then reload the POS.

The held ticket prints. A network printer sends no reconnect signal, so without the reload it prints the next time a ticket changes or syncs on the device. As in R4, do not reprint it by hand while it waits.

R6: A printer removed from settings

  1. Note which printer is assigned to a station.
  2. Remove that printer under Settings, Manage hardware.
  3. Fulfill an order for that station.

The station falls back to the printer marked Default. If no printer is marked Default, it falls back to the first eligible printer set up on the device, which is not something an operator can predict.

:::tip Mark one printer as Default on every device that prints dockets. It is the only way to know where a station’s dockets go if its assigned printer is removed or stops matching the template’s printer type. :::

R7: The report that catches what the screen does not

This is the safety net, and the one test that keeps working after go-live.

  1. Build a report on Shipment where Fulfillment Station is not blank and First Printed At is blank.
  2. Repeat R4 or R5 and, while the printer is still off, confirm the held ticket appears in it. It drops off the report once it prints.
  3. Give it to whoever runs service, and have them check it during and after each shift.

Every ticket that was never stamped as printed appears here. A docket lost after the stamp, such as a browser printer out of paper, does not. See Track station ticket printing.

R8: A resync does not reprint history

  1. Clear and resync orders and fulfillments on a device, as you would after a data reload.

No dockets print. Resyncing does not reissue tickets that already printed, and does not print ones that never did.

R9: New stations and printers do not flush a backlog

  1. With unprinted tickets already waiting at a station, cover that station on another device and switch Auto-print new tickets on.

No backlog prints on the covering device. Only tickets created after auto-print was switched on are printed automatically, so switching it on never dumps history at a printer.

This holds for covered stations only. A device that is a station’s own Register prints every open ticket that has never auto-printed, including ones raised before the device was set up.

Fulfillment stations go-live checklist

Check Done
Every sellable product at the outlet has a fulfillment category, or is knowingly excluded  
Every category is mapped to a station at every outlet that sells it  
Every station has a Register, or is deliberately left for the cover picker with Offline Behavior set to accept  
Every station has a Fallback Station, or you accept its tickets stay put when it is degraded  
Every station device has a fulfillment template selected  
Every station device has a working printer, tested with Run test print  
Multi-printer devices have printers assigned per station  
Station groups have weights that reflect real capacity  
Status Flow is set on every station that does not need the full four stages  
Staff have been shown Degrade station, Cover a station, Move tickets, and Reprint  
One printer is marked Default on every device that prints dockets  
The unprinted-ticket report from R7 exists, and someone checks it each shift  
Staff know a held ticket prints by itself once the printer is back, and not to reprint it by hand while it waits  
If one device prints for several stations, it is the only register named on those stations, and auto-print is off on every device that covers them  

When every test passes, an order taken on any register at the outlet produces one printed ticket per station involved, each ticket lists only its own items, and every resulting Shipment record carries a station and a print timestamp.

Was this article helpful?

Was this article helpful?