Warehouse tracking system: know where pallets, forklifts and stock actually are
Most warehouses do not have a visibility problem everywhere. They have it in three or four specific places: staging areas, returns, high value stock and equipment that moves between buildings. A warehouse tracking system is worth buying only for those places. We help you work out which ones need a continuous position, which ones only need a read when something passes a door, and which ones need nothing at all. Then we implement it.
Not sure whether you need RFID gates or a real time system? That is the normal starting point, and it is the first thing we help you decide.
The problems a warehouse tracking system is bought to fix
None of these are technology problems. They are the reasons an operations manager starts looking for a warehouse tracking system in the first place. If you recognize two or three of them, the rest of this page is for you.
Pallets that are in the building, somewhere
The system says the stock is on site. Nobody can say which aisle, which rack or which trailer. Someone walks the floor looking for it while a truck waits at the dock. Pallet tracking is usually bought after this happens once too often.
Staging areas that quietly turn into storage
Inbound goods land in a staging lane and stay there. Nobody measures how long, so nobody notices. The lane fills up, receiving slows down, and the only warning sign is that the floor looks busy.
Inventory counted by walking the aisles
Cycle counts and stock takes cost hours of labor and stop other work. The count is accurate the day it is done and starts drifting the day after. This is the problem an rfid warehouse project is normally aimed at.
Forklifts and equipment nobody can account for
Forklifts, tuggers, scanners, trailers and returnable containers move between zones, shifts and sites. When one is missing, the search is manual. Forklift tracking is also often asked for as a safety project, which is a different problem with a different answer.
Returns and phantom stock
Returned goods sit in a buffer waiting for a decision. The stock record says they exist, the picker cannot find them, so the item gets ordered again. Inventory accuracy drops for reasons that have nothing to do with the software.
Three questions to answer before you buy anything
These three answers decide the technology and most of the cost. They are also the questions that most vendors skip, because two of the three often lead to a cheaper system than the one they sell.
Do you need a continuous position, or only a passage?
This is the most expensive question in the project. If you need to know that a pallet left receiving and entered the racking area, a fixed reading point at the door answers it. If you need to know where it is right now while it sits somewhere in a large hall, you need a real time system. In warehouses, the first case is far more common than the second, and it costs much less. A hypothetical example rather than a project of ours: a distribution center that keeps losing pallets between staging and loading needs to know that a load left staging and never reached the dock. That is a passage question, answered by a read point on two doors, not by a live map of the whole hall.
How precise does the position really have to be?
Zone level, aisle level and meter level are three different budgets. Be honest about the decision the data will drive. If the picker still has to scan the label to confirm the item, knowing the zone is often enough, and paying for meter level accuracy buys nothing.
How many items, and who acts on the data?
Tagging a few hundred pieces of equipment is a different project from tagging every pallet that leaves the site and never comes back. Reusable assets justify a reusable tag. Consumable loads usually do not. And if nobody owns the alert when something sits too long in staging, the data will not change anything.
Which technology fits which warehouse problem
We are not tied to one technology. Two of the systems below are products we build ourselves, the Bluetooth Beacon Tracker and the Sentinel Video Tracker. The rest we did not develop, and we supply them all the same: we compare them, specify them, supply the hardware and run the project. That is the whole point of the arrangement: we have nothing to defend when we tell you which one fits. These are the options, with what each one is genuinely good at. The right answer is often a combination, and quite often the simplest one on this list. We run this today for a company that tracks forklifts and people across warehouses.
Passive UHF RFID at the gates
Tags with no battery, read automatically when a pallet or a tote passes a door, a dock or a portal. You get the event, the time and the direction, not a live position. For receiving, dispatch, returns and stock takes this is very often all that is needed, and the hardware and tag cost per item is the lowest of any option here. Start here before you consider anything else.
Our Bluetooth Beacon Tracker for live position
Battery beacons on the assets, fixed nodes on the ceiling, position calculated by trilateration. Accuracy is 1 to 3 meters in 90% of cases. Nodes detect beacons up to 30 meters away and are installed at the same height, on ceilings up to 7 meters. Positions update in the order of seconds. A zone or proximity mode is also available and costs less. Useful for equipment, returnable containers and the few pallets that really need to be found live.
Video for people, vehicles and flows
Our Sentinel video tracker runs on existing CCTV cameras and reaches 20 cm spatial precision. In a warehouse it is used for counting, dwell time in an area and flow patterns, not for identifying individual staff. It is often the cheapest way to measure a bottleneck, because the cameras are already there.
UWB when you need sub second and sub meter
Ultra wideband is the technology for tight accuracy and fast reaction, which is what forklift anti collision and worker proximity alerts require. It costs more per anchor and per tag, and it needs denser infrastructure. We will tell you when your case genuinely needs it instead of stretching the system we own to cover it. Neither UWB nor passive UHF RFID is a product we developed, and we supply both: we assess them, specify them against the alternatives, supply them and integrate them end to end. Which of them your site needs, and in which areas, is what we settle on the discovery call and confirm in the pilot.
Long range positioning for multi building sites
For logistics hubs spread over several buildings, where installing indoor infrastructure everywhere is not practical, we work with our partner Wheere. According to Wheere, its low frequency technology passes through concrete and reaches sub meter accuracy with gateways placed around the site rather than anchors in every room.
What a warehouse tracking system will not solve
This is the section most vendors leave out. Read it before you budget anything, because two of the points below are the most common reasons a warehouse project disappoints.
Forklift anti collision
Be careful here, because forklift tracking is sold as if it were one thing and it is two. Knowing which zone a forklift is in, how long it idles and how often it crosses a pedestrian area is something our Bluetooth system does well. Preventing a collision is not: that needs a reaction in a fraction of a second, and our BLE positions update in the order of seconds. For anti collision and proximity alerts you need UWB, and we will say so rather than stretch our own product to cover it.
Telling you which rack slot an item is in
BLE is not a sub meter technology. If you need bin level or slot level certainty, either you scan the label, which is cheap and reliable, or you look at UWB or long range sub meter options. Large metal structures and racking also degrade radio accuracy through reflection, which is exactly what a warehouse is full of. That is why we validate accuracy in a pilot on your site instead of promising a number in a proposal.
The yard, the trailers and everything outside
Radio positioning is reliable inside the perimeter formed by the outermost nodes. Outside it, positions fall on the edge of the covered area. Yards, outdoor storage and trailers in transit are a different problem, solved with GPS or long range technologies, not with an indoor node network.
A process that nobody owns
Tracking makes a mess visible, it does not clean it up. If goods are not labeled consistently, if the staging rule is informal, or if no one is responsible for acting when an alert fires, a tracking system will produce very accurate data about a process that stays broken. We would rather say this on a discovery call than after installation.
Replacing your WMS
This is not inventory management software. It tells you where things are and when they moved. Your WMS or ERP still owns stock records, orders and picking logic. We integrate with it, we do not replace it.
What it costs
We work with a USD 5,000 minimum commitment. That is the floor for a project we can run properly, including the assessment and a pilot. What sits above it depends on a small number of factors, all of which we can estimate with you on the first call.
What drives the cost up
The area to cover and its layout, because racking and metal mean more nodes for the same square meters. The number of items to tag, and whether tags come back to you or leave with the load. Continuous position instead of zone level. Ceiling height and power availability at the mounting points. Integration work with your WMS.
What keeps it down
Covering only the areas where the problem actually is, instead of the whole building. Using fixed read points at gates where a passage event answers the question. Zone mode instead of meter level. Re-using CCTV cameras you already own for counting and flow. Starting with a pilot so that the rollout is priced on measured facts.
Licensing
Two models, and in warehousing the second one comes up more often than people expect. You can run the system as a cloud subscription, hosted by us and kept up to date by us, or buy a one-time license and install it on-premise on your own servers. On-premise is the answer for sites where the uplink is unreliable, for buildings that sit on a network deliberately isolated from the corporate WAN, and for cold stores or outbuildings where coverage stops at the wall. The floor is the same either way: a USD 5,000 minimum commitment. Which model fits your site is something we settle with your IT team during the site assessment, not before it.
How we work
Four steps. You can stop after any of them, and after step two you will already know whether the project is worth doing.
1. Discovery call
30 minutes, free, with an engineer. What do you lose track of, how often, and what decision would change if you knew. At the end of it we tell you which of the technologies above fits, including the case where the answer is fixed RFID read points and nothing else.
2. Site assessment
We work from a scaled floor plan with the areas to monitor marked on it, a few photos and, ideally, the position of the power outlets. We check ceiling heights, mounting points, Wi-Fi reachability and metal structures, then size the system.
3. Pilot
A real deployment on a limited area: at least 6 nodes, covering roughly 300 square meters. The pilot is where the accuracy you will actually get in your building is measured, rather than quoted from a datasheet.
4. Rollout and integration
Full deployment, then data into your systems. The platform exposes a REST API for sites, nodes, beacons, levels and floor plans, push callbacks for events, and daily CSV files with positions on request. Cloud by default, on-premise supported. It is your data.
Frequently asked questions
Do I need a real time system, or is RFID enough for my warehouse?
For most warehouse questions, fixed RFID read points are enough and cost considerably less. If what you need to know is that an item arrived, moved between areas or left the building, a read at the gate answers it. You need a real time system when the item sits somewhere for a long time inside a large area and has to be found while it is stationary. We will tell you which case you are in on the first call, even when the answer means a smaller project for us.
How accurate is BLE tracking in a warehouse full of metal racking?
The confirmed claim is 1 to 3 meters in 90% of cases, and the final figure depends on the environment and on node density. Large metal objects and reflective surfaces degrade accuracy by reflecting the signal, and racking is exactly that. There is a useful paradox here: BLE often performs better in buildings with walls than in wide open halls, because signals attenuate more predictably. The honest answer is that the accuracy in your building is measured in the pilot, not promised before it.
Can I use this to stop forklifts colliding with pedestrians?
Not with our Bluetooth system. Anti collision needs a reaction below one second and our positions update in the order of seconds. For that specific use case the technology is UWB. What our system does do is tell you which zones forklifts use, how long they stand idle and how often they enter areas they should not, which is a traffic and layout question rather than a split second safety interlock.
Do I have to tag every pallet?
Usually not, and this is where budgets are saved. Tag what comes back to you: returnable containers, roll cages, totes, trolleys, equipment. For loads that leave the site and do not return, a passive RFID or barcode label at the gate is the sensible option, because a battery beacon on a one way pallet is money that walks out of the door. BLE beacons are standard iBeacon devices with a battery life measured in years, and they can be supplied by us pre-configured or bought by you.
Will it work with our WMS or ERP?
Yes, through the REST API and push callbacks, or with daily CSV files if you prefer a simple file drop. Positions come out as WGS 84 coordinates or as metric X and Y values, so you can map them onto your own layout. We do not replace the WMS: we feed it location data it does not have today.
Our racking and ceilings are very high. Does that change anything?
Yes. Nodes are mounted on the ceiling, all at the same height, on ceilings up to 7 meters. Above that, we look at mounting on structures, columns or racking uprights at a workable height, or at a different technology. High bay warehouses are one of the cases where we sometimes recommend fixed read points at ground level instead of full area coverage, which is also the cheaper answer.
Specifications
BLE accuracy: 1 to 3 meters in 90% of cases
Node range: Beacons detected up to 30 meters
Ceiling height: Up to 7 meters, nodes at the same height
Update rate: In the order of seconds
Video precision: 20 cm, on existing CCTV
Pilot: At least 6 nodes, about 300 square meters
Integration: REST API, push callbacks, daily CSV
Minimum commitment: USD 5,000
Tell us what you keep losing
Bring the floor plan and the problem. We will tell you which technology fits, what it will cost and, when it applies, that you do not need a real time system at all. No obligation and no sales pitch.