The hardware conversation usually happens too late. The software is selected, the configuration is underway, and somebody asks in a status meeting what devices we are using. Two weeks before go live, twenty tablets arrive in a box.

Then the warehouse finds out that the back corner of the building has no signal, that the tablets cannot read a label from more than eight inches, and that a dropped device costs $600 to replace.

Hardware is not the interesting part of a WMS project. It is the part your operators touch every minute of every shift, and it is where I see money wasted most predictably.

Warehouse Wi-Fi is a different problem than office Wi-Fi

Your building probably has Wi-Fi that people describe as fine. It is fine for a laptop that sits still and can pause for two seconds without anyone noticing.

Scanning is not that. A handheld moves constantly, hands off between access points at walking and forklift speed, and every scan is a round trip to the server. A gap of a few seconds that a laptop absorbs invisibly is a picker standing there tapping the screen.

Three things make warehouses hard that offices are not:

  • Racking is not drywall. Metal and dense product absorb and reflect signal. Coverage measured in an empty building is not coverage in a full one, and coverage in January is not coverage at peak.
  • Ceilings are high. Access points mounted to a 30 foot ceiling put most of their energy where nobody works. Lower mounting or directional antennas usually beat adding more APs.
  • Roaming matters more than raw speed. A device crossing between APs needs to hand off cleanly. Consumer gear does this badly, and it shows up as scans that hang at aisle boundaries.

Get a real site survey with the racks full. A predictive survey off a floor plan is a starting point and not an answer. If your vendor or integrator will do a walk test with the actual device model you plan to buy, that is the version worth paying for.

Test the dead spots people already complain about, plus the freezer if you have one, plus the dock doors when they are open, plus the mezzanine.

Offline behavior is a real question, ask it early

Ask what your WMS does when a device loses connection mid task. The answers vary a lot.

Some systems queue the transaction and sync on reconnect. Some fail the task and make the user start over. Some appear to accept the scan and silently discard it, which is the worst outcome available.

If your building has spots you cannot fix economically (an outbuilding, a yard, a freezer), the offline answer determines whether you can work there at all.

Rugged devices versus phones

Phones and consumer tablets are cheap, familiar, and take good pictures. They also have a camera-based scanner that is meaningfully slower per scan and struggles with damaged labels, poor light, and distance.

The math is per scan seconds times scans per day. If a picker does 500 scans a shift and a camera scan is a second and a half slower than a laser or imager, that is over ten minutes a day per picker, every day, forever.

Where I land:

  • High volume scanning roles (picking, receiving, cycle counting): rugged devices with a real imager. They survive drops onto concrete, the batteries last a full shift, and the scan is instant.
  • Occasional scanning (supervisors, office, a second building with light volume): phones are fine and the savings are real.
  • Forklifts: vehicle mounted terminals, or a rugged handheld in a holster. Not a tablet velcroed to the mast, which I have seen more than once.

Buy spares. Ten percent of fleet size, minimum two. A device out for repair with no spare means someone works on paper, and paper work does not reach the system the same day.

Check the battery specification against your actual shift, including the drain from the scan engine and screen brightness, and check whether batteries are hot swappable. On a ten hour shift that matters.

Labels and printers, which nobody budgets

Location labels are cheap until you buy them for a whole building, and then they are a line item worth planning.

Get them printed on durable stock with the right adhesive for your rack finish, and include both a barcode and human readable text large enough to read from a forklift. Rack labels that need a person to walk up and squint are a permanent tax on every putaway.

For high racks, add aisle signs and level markers so a driver can find a bin from the floor. If you are labeling a whole building anyway, do the location renaming at the same time. It is the same ladder time.

For printers: check the label formats your WMS produces and whether they will run on the printers you own. Sometimes they do, sometimes the system expects a specific printer language and your existing units need replacing. Finding that out in the test phase is cheap. Finding out during go live week is not.

Charging, storage, and the mundane stuff

Decide where devices live between shifts, who plugs them in, and what happens when someone takes one home in a jacket pocket.

Assign devices to people or to workstations, and put a number on the back big enough to read across a room. When a device goes missing or gets abused you want to know whose it was, not because you want to punish anyone but because damage clusters, and the cluster usually points at a physical problem like a bad ramp.

Mount chargers where the traffic already flows. If the charging rack is in the office and the pickers start at the far dock, devices will be charged in someone's locker.

Order it early

Rugged devices routinely run six to ten weeks out, longer for anything with a specific radio or scan engine. Access points and switches have their own lead times.

Order after selection, not after configuration. You want devices in the building for testing and training, on the actual network they will run on, weeks before anyone depends on them. The first day your operators hold the device should not be the first day they are also learning new software.