Why Aren't My Barcodes Scanning: Troubleshooting Common Read Errors

When a barcode will not scan, the instinct is to fault the label or the scanner and move on. The more effective approach is to monitor the read rate. A read rate is a formula that represents the accuracy of scanning systems calculated by the number of barcodes that are read divided by the number of attempts. Read rates are generally expressed as a percentage and the closer to 100%, the better.  

Read rates matter because they are directly related to costs. The more no-reads, or unsuccessful attempts at reading a barcode, the more costs increase.

For example, if your read rate falls from 99.8 to 94.0%, it might seem like an acceptable decrease. But that translates to 5.8% more no-reads. Every time that happens, an operator is pulled from their task to resolve the issue, and on a busy line scanning millions of parts or packages per year, costs add up quickly.

The table below demonstrates the approximate cost of a read rate decreasing for 99.8 to 94.0%: 

The cost of no-reads

Why does a barcode fail to scan in the first place?

A barcode fails for one of four reasons:

• the symbol is poorly marked
• the reader is wrong for the code
• the environment interferes with the image
• the software mishandles the decoded data

Most no-reads trace back to a single domain. Identify that domain first, and the correct fix follows quickly.

The mistake most teams make is treating every failure as either a damaged barcode problem or a faulty scanner. A no-read is rarely that simple. It sits inside a full data capture chain that runs from the mark on the part to the record in your software.  

The four leading domains responsible for no-reads

No-reads and misreads are two different problems

A no-read and a misread are different failures with very different consequences. A no-read is a failed attempt to read a code, so nothing gets recorded and the line waits. A misread is worse. The reader decodes the code but returns the wrong data. No-reads cost time while misreads corrupt data.

A no-read is direct; it halts a line, flags an issue, and someone fixes it. A misread lets the errors silently compound, the line keeps moving while the wrong lot number, part ID, or destination flows into your systems. In regulated sectors like medical device manufacturing, aerospace, and consumer goods, misreads put audit trails and traceability at risk, and it can surface later as a fine, a chargeback, or a recall. When you tune a reading system, optimize read rates and prevent misreads with checksums and structured data validation.

When asked which cost customers more, Cognex application engineering supervisor Pankhil Soni said generally, misreads are harder to identify, causing issues to compound and costs to rise as the problem goes unnoticed.

No-reads were more straightforward; you can usually determine very quickly if a code did not scan. It’s still costly, but it’s a more manageable problem and faster to fix it.  

Daniel Lapidus, a senior applications engineer at Cognex, added passing along the wrong information – regardless of where it happens in your manufacturing or logistics process – is often a worst-case scenario.

“If a product or package has incorrect data, it can lead to line stoppages, recalls, and directly harm your customers,” he said. “When I and most of our team design systems, we always design the fail-safe so that if there is ever a question about the data, we will preferentially fail that rather than give false or incorrect data.”

Lapidus added most customers have systems that serialize products, so they will be able to halt systems if they see inconsistent data and evaluate what caused the misread.  

Is it the label or the mark?

Start with the code itself, determine if the marking method is inefficient, the mark itself is damaged, or the environmental conditions that make a properly marked code hard to read. For label-based codes, most unscannable barcode problems come from low contrast, incorrect sizing, poor placement, smudging, physical damage, or a missing quiet zone. A label can look fine to your eye and still sit below the quality threshold a reader needs. Print quality and mark quality are where reliable reads begin.

On the print side, the most common issues are mechanical: worn or improperly tensioned ribbon, dirty or uneven substrates, and print speeds that are too high to produce clean edges. Handling can also damage barcodes, while torn or creased labels and codes placed across seams or too close to an edge can cause failures even with a reliable scanner. Fix these issues at the source. No reader setting can compensate for a poorly printed or damaged code. 

A quiet zone is the clear margin around a barcode that allows the reader to identify where the code begins and ends. It’s a design setting, not a print variable, so the quiet zone either meets specification or it doesn’t. Even a clean, well-printed barcode can fail if its quiet zone is insufficient.

Because the quiet zone rarely changes during a run, this failure hides in plain sight. The code prints, it looks complete, and it still no-reads because the reader cannot find its boundaries. Check the quiet zone in your label template first when a batch fails. 

Why does my reader miss codes that look perfectly fine?

Often the reader was never matched to the code. A reader has to be set for the right symbology, carry enough resolution to resolve the smallest element, and sit within its focal range at an angle that avoids glare.  

Three configuration issues account for most of these cases:

• Scanning technology: laser scanners can only read one-dimensional (1D) label-based codes, while image-based scanners can read 1D and 2D label-based codes and direct part marks (DPMs). When a code suddenly stops scanning, confirm the scanner can read the symbology.  
• Resolution: Every reader can only resolve elements down to a certain size, so a high-density code with a small X-dimension may fall below what a standard reader can capture.
• Geometry: Each reader has an optimal focal range and mounting it square to the code at 90 degrees can bounce light straight back into the sensor. A slight tilt removes that mirror effect.

Match the device to the job. Compact fixed-mount barcode scanners suit high-speed, hands-free lines, while handheld barcode scanners fit variable, operator-driven work. The point is to specify the reader for the application and code type. 

Customer success story: how Beyonics stopped no-reads from halting production

Beyonics, an electronics manufacturer, saw no-reads forcing operators to reload PCBs and stall its surface-mount technology (SMT) lines. “This caused the SMT machines to come to a halt until a manual intervention took place,” said Shanker Kaneson, an engineering specialist at Beyonics.  “With the DataMan fixed-mount readers installed, we witnessed a production throughput increase of about 10%.”

The limiting factor was the reader, not the code.

The Beyonics engineers wanted a plug-and-play swap with no re-cabling or re-programming. Image-based readers delivered exactly that while lifting read rates on the same codes. This is the pattern to watch for: when unacceptable read rates clear up after a reader upgrade with no change to the marking process, the reader was the constraint all along.

Read the Beyonics success story.

Why does the same code scan on one line but not another?

The code and the reader can both be fine while the environment obscures the code. Glare from overhead lighting, dark racking aisles, dust on the lens, and other variables can lower read rates without any obvious hardware fault. This is the classic case of a code that reads on the bench and fails on the floor.

Environment is the domain teams check last and should often check first, because the reader shows no fault. It simply works in some conditions and fails in others. Controlled machine vision lighting and the right machine vision lenses turn an unstable read into a stable one by fixing contrast and focus at the source. Before you fault the reader, replicate the actual line conditions, not the lab conditions, and watch what the read rate does.

Customer success story: how a brewery scanned creased and torn labels

At Flensburger Brewery, forklift labels arrived creased and partially torn from a harsh logistics environment. Mast-mounted DataMan barcode scanners enable employees to activate the scanner from the forklift instead of aiming for a handheld scanner, decreasing no-reads.  

“The large depth of focus of the optics and the very high read rates, even under unfavorable conditions, are advantageous,” said Christian Ludwig, a business solutions specialist at Jetschke, a machine vision system integrator that partnered with Flensburger.

The environment did not change. The reading approach did. By moving from handheld scanning to fixed image-based readers built for tough conditions, the brewery cut time on every pick and eliminated the recurring expense of fixing damaged, unread codes. Robust optics and decoding did the work that a cleaner label would have required otherwise.

Read the Flensburger brewery success story. 

The data layer is where invisible failures hide

A clean decode does not guarantee clean data. Readers can append stray prefix or suffix characters that make a valid scan get rejected. Worse, a reader that captures a GS1 code, but does not parse its application identifiers, can pass the whole raw string to your system, filing the lot number or expiry date in the wrong field.

This is the misread that looks like a successful read. The scan looks fine, the operator moves on, and the error only surfaces downstream when the data does not add up. Reading software and edge processing that parse structured GS1 fields correctly, rather than dumping a raw string, close this gap. 

What makes DPMs so hard to read?

DPMs often have inherently low contrast. The code and the surface are the same material, separated only by microrelief or reflectivity, so contrast is faint. Add curved or reflective metal, glare, tiny module sizes, and wear over the life of the part, and DPMs become the toughest reading job on the floor.

Manufacturers in automotive, aerospace, and electronics choose direct part marking with laser, dot peen, or chemical etch because the code has to withstand harsh manufacturing environments and be readable throughout the part’s entire lifecycle.  

You cannot reprint a bad DPM, so the quality has to be right the first time, which makes verification a requirement rather than an option. Reading these codes reliably calls for image-based barcode readers, the right lighting to raise contrast on metal, and optics matched to very small codes.

Explaining why a clearly visible mark can be difficult for a reader to scan and decode can be difficult. But a major benefit of image-based barcode scanners are the ability to recall images of misreads and no-reads, which help teams identify and diagnose scanning issues. 

“Since we are using machine vision, oftentimes, the easiest thing to do is show them images from the camera,” Lapidus said. “That helps show our customers why things are happening the way they are, or why certain barcode scanning setups work better than others. It helps illustrate concepts like the geometry of the setup, the field of view, and lighting.”

When should you add a barcode verifier?

Add a verifier when a passing scan is not enough proof. Scanning tells you a code reads today. Verification grades it against an ISO standard and tells you how much margin it holds before it fails. In regulated and high-volume work, that margin is the difference between reliable compliance and relying on luck.

Under ISO grading, a code that cannot be decoded receives an automatic F. For codes that can be decoded, the overall grade is determined by the lowest individual parameter grade, so a single weak parameter can bring down the grade for the entire code.

Most industries require a specific grade to guarantee the code stays readable across the supply chain. That’s why a code can scan on your line and still be one bad print run away from failing at your customer.  

Common Barcode Verification Standards

Inline barcode verifiers bring this grading onto the line so you catch poor-quality codes as they are produced, not after a shipment gets rejected.

GS1 Sunrise 2027 is the global move to accept 2D barcodes at retail point of sale. The effort already spans 48 countries and about 88 $ of global GDP. Until roughly 90% of point-of-sale systems can read 2D codes by the 2027 deadline, products will carry both a 1D and a 2D code, which multiplies the codes that can fail. Products already marked with 2D codes will not have to carry both types of symbologies.

Dual marking is the near-term reality, and it doubles your exposure to reading errors during the transition. It also raises the payoff for getting reading right, because 2D codes such as GS1 Data Matrix carry far more than a product number. They hold lot, expiry, and links to live product data, which feeds traceability, Industry 4.0 workflows, and IIoT visibility across the supply chain. Reading infrastructure that handles both symbologies cleanly today is the infrastructure that carries you through 2027. 

Most manufacturers underestimate the smaller symbologies that can come with Data Matrix and GS1 digital link initiatives, and the hardware required to read them, according to Soni.

“More data, and more encoded data, can necessitate higher-density 2D codes,” he said. “To read 2D symbologies, you may have to invest in a higher-resolution barcode scanner, modify your lighting setup, and ensure you have compatible software.”

The GS1 Sunrise 2027 initiative might also necessitate reconfigurations like data reformatting, the way a PLC collects data, and data management, added Lapidus. 

Build a repeatable barcode troubleshooting workflow

When a code fails, work through the four domains systematically. Inspect the symbol, confirm the reader configuration, check the environment, then trace the data path. A consistent sequence turns barcode scanner troubleshooting from trial and error into a repeatable process that anyone on the team can run.

• Inspect the symbol. Check contrast, size, quiet zone, placement, and damage. Grade a sample against the standard if you can.
• Confirm the reader. Verify the correct symbology is enabled, the resolution suits the X-dimension, and the code sits inside the focal range.
• Examine your environment. Look for specular glare and ambient light, adjust your lighting setup accordingly and ensure the lens is clean.
• Trace the data. Look for stray prefix or suffix characters and confirm the software parses application identifiers into the right fields.

Reliable reads start with matching the system to the application

Reliable barcode scanning is a systems outcome. Match the reader to the code, control the environment, and confirm the data path to make no-reads less frequent and easier to diagnose. The teams that treat reading as a full chain, from mark to software, are the ones that encounter the fewest unexpected issues.

The next time someone asks why a barcode won’t scan, you’ll have a better answer than “try a new scanner.” You have a domain to check, a sequence to run, and a way to tell a passing scan from a code that is quietly running out of margin.