Your Warehouse Is Automated. So Why Are You Still Flying Blind?

Automation can move goods efficiently. It does not guarantee that anyone knows what is where.

A call comes in from the sales desk. A customer wants a firm answer: how much of one product is available, and when can it ship?

The warehouse manager opens the system. It shows pallets of that product in a storage lane. The warehouse has automated storage, so pallets move in and out of the racking with little handling. Then production asks what exactly is in that location right now.

The manager hesitates. A transfer last week may not have been confirmed. A pallet was moved during a busy shift, and nobody is sure the record was updated. To be certain, someone will have to check the records line by line or arrange a physical retrieval to look at the pallet.

This is an illustrative scenario, not an account of any specific company or customer.

The question is not whether this warehouse has technology. It plainly does. The question is why a simple operational question still needs an investigation.

This article is for the person who has to answer questions about stock but cannot answer them without checking first. It is also for the leaders who assume automation has already solved that problem.

Automation moves goods. Visibility tells you what is happening.

Warehouse automation and inventory visibility solve related problems. They are not the same problem.

Automation improves how physical work gets done: moving, storing, retrieving, transporting. Visibility depends on something else. It depends on whether the business has reliable information about the goods, their locations, their movements and their status.

A warehouse can have sophisticated equipment and still rely on disconnected records, manual confirmations or processes that miss some movements. When that happens, the equipment does its job and the manager is still unsure.

In my recent video, Stop Automating Inefficiencies, I challenged the decision to automate a process that is already broken. This article explores a different case. Here the automation may be working exactly as designed. The information needed to manage the operation can still be inadequate.

One technical point matters. Radio shuttle systems do not inherently lack inventory visibility. Whether a warehouse can say what is in a lane depends on its identification process, its warehouse management system, its operating procedures and how these connect to the equipment. The equipment is one link in the chain. It is not the whole chain.

Where the gap begins

Uncertainty about stock is a symptom. The cause can sit anywhere along the chain that links a physical pallet to its record. These are the failure points we would investigate first:

• Identification. The system cannot reliably associate a physical pallet with the product information recorded against it.
• Location tracking. A movement happens physically, but the matching location record is missing, late or wrong.
• System integration. The equipment, the warehouse software and the enterprise inventory records do not share information reliably enough.
• Process discipline. People perform movements or handle exceptions outside the normal recording process.
• Exception handling. When a movement fails or a transaction is interrupted, nothing reliably reconciles the physical state with the recorded state.

These are potential failure points, not findings from a survey of warehouses.

The important point is that a manager can discover the problem at a storage location without the failure having started there. The pallet in front of them is where the gap becomes visible. It is rarely where the gap was created. This connects to The Inventory Manager Is Held Responsible for Numbers the System Cannot Verify: a count reveals a gap but does not explain where it began. Here we follow the gap further upstream, to the equipment and the information systems around it.

Research supports the broader point that records and physical stock drift apart, in settings with and without heavy technology investment. It does not prove that automation causes poor visibility, or that automated warehouses are generally inaccurate.

In retail, a study by Nicole DeHoratius and Ananth Raman, published in Management Science in 2008, examined nearly 370,000 inventory records from 37 stores of one retailer. It found about 65 per cent of them to be inaccurate. An earlier article by Ananth Raman, Nicole DeHoratius and Zeynep Ton, in California Management Review, reported that accuracy varied substantially between stores in the same chain that used identical information technology.

In warehouses, a 2026 study at an oil and gas warehouse in Indonesia compared physical stock with ERP records for all 75 items that had a recorded receipt, issue or relocation during one month. Discrepancies appeared in 25 of them, or 33.33 per cent. The researchers associated several of them with delayed transaction postings, incomplete documentation, temporary material placement and movements not yet reflected in the system. Their field observations indicated that most discrepancies originated during movements, not during counting. The study covered one warehouse for one month with no control group, so it describes what was found there and is not a benchmark.

Automated storage is harder to evidence. We did not find independent research that measures record accuracy in automated or shuttle based warehouses. Vendors argue that automation can improve accuracy, for example by creating a bin level record as goods move. One vendor's own 2026 guide to its bin based system concedes that accuracy depends on how goods are decanted against the receiving plan, and warns that otherwise you get a very precise record of a sloppy event. That is vendor material about a different storage type, so we treat it as perspective, not proof.

Taken together, the evidence is narrower than a statistic and still useful. Records and physical stock can diverge in stores and in warehouses, with and without strong systems. Where researchers looked for causes, they pointed to practices and processes around the movement of goods, not to the equipment alone. For automated storage, that is a hypothesis to test in each operation, not a proven finding.

The cost of not knowing

Uncertainty turns an information gap into an operational cost. Depending on the operation and the severity of the gap, the consequences can include:

• Staff spend time investigating stock that should be simple to locate.
• Picking or production decisions wait while inventory is verified.
• People carry out extra checks to compensate for records they do not trust.
• Managers hesitate to commit to customers on information they cannot rely on.
• Different teams make decisions from different versions of the same inventory position.

Research points the same way. A 2023 simulation study of distribution centres found that record inaccuracy affected the performance of every warehouse it modelled, measured through picking productivity, lost sales and use of warehouse capacity. It is a simulation, not a field measurement.

Not every warehouse with limited visibility will experience all of these. One pattern does appear whenever people stop trusting operational information: they create workarounds. They call colleagues. They check spreadsheets. They recount. They retrieve a pallet to look at it. They keep personal records.

This is the dynamic we described in The Operational Mistakes That Keep Complex Businesses Running Below Their Potential. When systems do not give people what they need, people become the integration layer. The warehouse still ships. The cost is paid in time, attention and confidence, and it rarely appears on any report.

Why buying more technology may not fix it

When visibility is poor, the instinctive response is to look at more technology: another software module, more scanners, a new storage system, further automation. Sometimes that is the right answer. It is the wrong first step. Before investing, the business needs to find where the information chain breaks.

Six questions help:

1. Can we identify the physical goods reliably?
2. Are inventory movements recorded when they happen?
3. Can we establish the current location of each relevant handling unit?
4. Do the warehouse system and the equipment reflect the same operational state?
5. What happens when a transaction fails or a movement differs from the plan?
6. Can we trace how the current inventory record was produced?

These questions are more useful than a technology shortlist. The point is not that businesses should avoid investing in technology. It is that they should understand what the investment is meant to solve before deciding what to buy. That was the argument of the video, and it applies here too.

What a better approach looks like

Before the next warehouse technology investment, a business should establish five things.

Reliable identification. A dependable link between each physical pallet or handling unit and its record.

Accurate movement capture. Transactions that reflect what actually happened, when it happened.

Location integrity. A credible record of where goods are stored now, not where they were last recorded.

Connected systems. Consistent information across the equipment and every software system that depends on it.

Exception reconciliation. A clear process for resolving differences between physical activity and system records.

The right solution depends on the warehouse's actual operating model. One operation may need better procedures and tighter transaction controls. Another may need improved identification, deeper integration or further automation. A third may need a combination. Not every warehouse needs a major technology deployment.

That is how we approach it at TENNXT. We identify the operational gap first. Then we decide which mix of processes, software, equipment and integration will close it.

Can you know what is happening in the warehouse you already have?

A warehouse does not become intelligent simply because its equipment is automated. It becomes easier to manage when the people responsible can reliably establish what is there, where it is, what has happened to it and what needs to happen next.

Before investing in more warehouse technology, ask a more basic question: can we reliably know what is happening in the warehouse we already have?

If you manage a warehouse, where do you lose visibility most often: identifying stock, tracking its location, recording movements or connecting your systems?

---

---

Further reading:

• The Inventory Manager Is Held Responsible for Numbers the System Cannot Verify
• The Operational Mistakes That Keep Complex Businesses Running Below Their Potential
• Why Enterprise Software Projects Fail Before Development Begins
• Why Operational Visibility Must Come Before Automation
• How to Detect Engineering Bottlenecks Before They Become Operational Problems

Sources:

• DeHoratius, N. and Raman, A. (2008), Inventory Record Inaccuracy: An Empirical Analysis, Management Science, 54(4), pp. 627-641
• Raman, A., DeHoratius, N. and Ton, Z. (2001), Execution: The Missing Link in Retail Operations, California Management Review, 43(3)
• Rambe, A. W., Hartini, S. and Handayani, N. U. (2026), Development of a Movement-Triggered Inventory Checking Method for Earlier Detection of Inventory Record Inaccuracy, Jurnal Teknik Industri, 12(2), pp. 201-210
• Destro, I. R., Staudt, F. H., Somensi, K. and Taboada, C. (2023), The impacts of inventory record inaccuracy and cycle counting on distribution center performance, Production, 33, e20220077
• Rapyuta Robotics (2026), Digital Transformation for Warehouses: How ASRS Creates a Live Bin-Level Inventory Record (vendor material, cited as perspective only)

Research note: The cited studies examine retail inventory, a conventional warehouse, and simulated distribution centres. Their findings do not establish inventory accuracy rates for automated or radio-shuttle warehouses. The scenario described in this article is illustrative, not a case study of a specific company.

Joshua Otomo is Founder and Chief Executive Officer of Tennext Core Ltd (TENNXT), an enterprise systems integrator focused on intelligent software, industrial automation and operational analytics. He writes about how large organisations get technology projects right and, more often, why they don't.

Contact: hello@tennxt.com · www.tennxt.com