EN
Inventory and Products

How Do You Find Product Code Matching Errors? A Course on the Code Rule, Match Checks, Error Types and a Correction Plan

How do you find product code (SKU) matching errors? Code rule, match rate across channels, error types and fix order: an 8-lesson marketplace guide.

Contents 15 sections
  1. What will you be able to do by the end of this training?
  2. Lesson 1: Why does product code matching matter?
  3. Lesson 2: How do you write a good product code rule?
  4. Lesson 3: How do you run a match check?
  5. Lesson 4: What types of errors are there?
  6. Lesson 5: How do you calculate the cost of a matching error?
  7. Lesson 6: In what order are fixes made?
  8. Lesson 7: What do you do so errors do not recur?
  9. Lesson 8: How do you monitor matching regularly?
  10. How do you catch codes with spelling differences in a spreadsheet?
  11. What are the most common mistakes in product code matching?
  12. How does NetKasam make product and inventory matching easier?
  13. Mini quiz: Test what you have learned
  14. Conclusion: Fix the code and inventory and prices will flow correctly on their own
  15. Frequently Asked Questions About Product Code Matching

A product that shows 12 units in your inventory system says "sold out" on the marketplace. Another product, at 0 units in the system, keeps selling on your own site. Most of the time the cause is not inventory but the code: the same product is recorded in two places with two different spellings, and the system does not recognize them as the same product. When product code (SKU) matching breaks, the inventory, price and order flow break silently, and nobody sees an error message.

This post is not a narrative but an 8-lesson hands-on course. We will answer the question how to find product code matching errors using a sample store with 600 products and 1,200 listings across two channels (your own site and a marketplace): the code rule, the match check, error types, impact calculation, fix order and lasting prevention. At the end there is a mini quiz and an answer key. All figures are sample values chosen to demonstrate the method; use your own catalog data.

What will you be able to do by the end of this training?

In short: You will write a code rule, calculate the match rate across channels, sort errors by type and order the fixes by their impact.

Course mapThe six steps of a product code matching check
  1. 01
    Set the ruleDefine a single product code format.
  2. 02
    ListPull each channel's product and code list.
  3. 03
    MatchCompare the codes with the master record.
  4. 04
    ClassifySort the unmatched items by error type.
  5. 05
    FixStart with the one that has the biggest impact.
  6. 06
    ProtectMake the code mandatory on new products.

The 8 lessons of the course: (1) why matching matters, (2) the code rule, (3) the match check, (4) error types, (5) the cost of an error, (6) fix order, (7) lasting prevention, (8) regular monitoring.

Recommended ReadingHow Do You Calculate the Cost of Stockouts? A Course on Lost Sales, Indirect Loss, Loss by Product Class, the Extra Stock Decision, Early Warning Rules and Limiting Losses

Lesson 1: Why does product code matching matter?

In short: Inventory, price and order information travel between products by code; if the code does not match, the system does not recognize the product and does not make the update.

In multichannel selling, every product lives in at least two places: in your master record and in the channel's listing. Inventory dropping when an order arrives, channels updating when inventory changes, and the right product being picked when an invoice is issued all depend on the code. If the code is "TSH-0142-M" on one channel and "tsh 0142 m" on another, a person sees the same product but the system counts two separate products.

The result goes both ways: the product is in stock but shows as "unavailable" on the channel and a sale is lost, or inventory has run out but the channel keeps selling and cancellations follow. For the cost of the second, see the post how to prevent overselling on marketplaces. For how inventory updates should flow, the post how to synchronize marketplace inventory is a good complement.

Lesson 2: How do you write a good product code rule?

In short: The code is written in one format, without spaces, in capital letters and with sections whose meaning is fixed; each variant (size, color) gets its own code.

SectionExampleRule
Product groupTSH3 capital letters
Model number01424 digits, leading zeros preserved
ColorLC2 capital letters
Size/variantMCapital letter or number
Full codeTSH-0142-LC-MSeparated by hyphens, no spaces

This is an example; you can change the sections to fit your own catalog. What matters is that the rule is written down, kept in one place and applied the same way by everyone. Avoiding Turkish characters, spaces and special symbols in the code reduces transfer errors. Keep the relationship between barcode and code separate; for setting up barcodes, see the post how to set up a barcode system. For how to keep product information in the master record, the post how to prepare a product card is useful.

FormulaMatch check formulas
  1. Matched listings = Listings linked one to one with the code in the master record
  2. Unmatched listings = Total listings − Matched listings
  3. Match rate = Matched ÷ Total listings
  4. Error share = Number of an error type ÷ Unmatched listings

Lesson 3: How do you run a match check?

In short: Compare the master product list and each channel's listings by code; codes that are on one side but not the other, or spelled differently, are unmatched.

In the sample store, 600 products are live on two channels, so there are 1,200 listings. Export each channel's list and put it side by side with the master record. Do the comparison in two stages: first an exact match, then a "cleaned" match (ignoring spaces, hyphens and letter case). Codes that do not match exactly but match once cleaned are spelling differences; those that still do not match are empty, duplicate or wrongly linked.

ResultCountRate
Exactly matched listings1.104%92
Unmatched listings96%8
Total1.200%100

92% looks good at first glance, but each of the 96 listings means a broken inventory and price flow for one product. Repeat the check monthly; the rate drops as new products are added. To combine the same check with a stock count, see the post how to do a stock count.

Lesson 4: What types of errors are there?

In short: There are five types of errors: spelling difference, empty code, duplicate code, wrong variant link and passive product; each has a different fix.

DistributionBreakdown of 96 unmatched listings by error type
  • Spelling difference (space, letter, zero)₺3839.6%
  • Empty code₺2425%
  • Duplicate code₺1414.6%
  • Variant linked to the parent product₺1212.5%
  • Passive or old product₺88.3%
MistakeExampleSolution
Spelling differenceTSH-0142-M and tsh 0142 m and TSH-142-MStandardize the code to the rule
Empty codeThe code field has not been filled inAssign the code, prevent leaving it blank
Duplicate codeTwo products carry the same codeGive one a new code, separate the inventory
Variant linkSizes M and L are linked with a single codeA separate code for each variant
Inactive productA product withdrawn from sale is still listedClose or delete the listing

The 38 spelling differences and 24 empty codes make 62 listings, or 64.6% of the unmatched ones. These two groups mostly come from manual entry; a rule and a mandatory field largely close this group. For link errors in products with variants, see the post how to manage products with variants on marketplaces.

Lesson 5: How do you calculate the cost of a matching error?

In short: Calculate three items: overselling cancellations, sales lost while inventory is available, and manual correction time.

In the example, 18 orders caused by unmatched listings were cancelled because there was no inventory; at ₺85 per cancellation, that is 18 × 85 = ₺1,530. The 24 listings that show "unavailable" on the channel although the product is in stock miss an average of 2 orders a month, and with an average basket of ₺300, the lost revenue is 24 × 2 × 300 = ₺14,400. Manual correction at 5 minutes per listing takes 96 × 5 = 480 minutes, or 8 hours.

ItemAccountEffect
Overselling cancellations18 × 85 ₺1.530 ₺
Lost sales (in stock, unavailable on channel)24 × 2 × 300 ₺14.400 ₺
Manual correction time96 × 5 min480 min (8 hours)

Lost revenue and profit are not the same; to see margin by product, see the post how to analyze profitability by product. For the cancellation process, the post how to cancel an order is useful. These amounts are examples; enter your own cancellation and basket values.

Lesson 6: In what order are fixes made?

In short: First fix the listings that cause lost sales (in stock, unavailable on channel) and cancellations, then the error types that are highest in number.

ChartUnmatched listings falling after the first three fixes
  1. Unmatched listings96 listings
  2. Spelling differences standardized−38 listings
  3. Empty codes filled in−24 listings
  4. Duplicate codes separated−14 listings
  5. Remaining unmatched listings20 listings

Prepared with sample values. Calculation: 96 − 38 − 24 − 14 = 20. The remaining 20 listings are variant link errors (12) and passive product errors (8).

ComparisonMatch rate by fix stage
  1. Start92.0%
  2. After the first three fixes98.3%
  3. After all errors are closed100.0%

The first three fixes close 76 listings and raise the match rate to (1,104 + 76) ÷ 1,200 = 98.3%. The remaining 12 variant errors need attention product by product, so they are left for the second stage; the 8 passive products are solved by closing the listing. To clean up many old products in bulk, see the post how to simplify a product catalog.

Lesson 7: What do you do so errors do not recur?

In short: Make the code field mandatory, match a new product before it goes live, change the code only in the master record and push the change to the channels from there.

  • Mandatory field. A product with an empty code is not put live.
  • Single source. The code is changed only in the master product record; no manual changes on channels.
  • Match before going live. Before publishing a new product on a channel, confirm that the code is linked.
  • Code owner. Have a single person assign codes, or share the rule document.
  • Passive product routine. A product withdrawn from sale is closed on the channels within a week.

To add these steps to your daily operations list, see the post how to build an e-commerce daily operations checklist. If you want to combine it with product listing quality, you can add a "code matched" criterion to the score in the post how to measure product listing quality.

Lesson 8: How do you monitor matching regularly?

In short: Check new products, take a quick weekly look, pull a full monthly audit and compare the rate with the target.

FrequencyWhat happens
Every new productVerify compliance with the code rule and the channel link
Every weekScan products with 0 inventory that show "available" on the channel (and the reverse)
Every monthFull match check: rate, error type, fix list
Every quarterReview the code rule and the passive product list
IndicatorHow it is calculatedSample value
Match rateMatched ÷ Total listings%92 → %98,3
Product with an empty codeNumber of products without a code24 → 0
Lost salesUnavailable on channel × Orders × Basket₺14,400 / month
Correction timeListings × Minutes480 min
Takeaways from the exampleFour numbers from the product code matching example
  • 92%starting match rate
  • 96unmatched listings
  • ₺14,400monthly lost sales from products that show as unavailable on the channel while in stock
  • 98.3%match rate after the first three fixes

To add the monthly result to the management summary, you can use the layout in the post how to prepare a weekly management report. When looking for the cause of inventory differences, a code error is also one possible cause; for details, see the post how to find and close an inventory difference.

How do you catch codes with spelling differences in a spreadsheet?

In short: Compare the codes in a cleaned helper column: remove spaces and hyphens, convert letters to uppercase and keep leading zeros as text.

  1. Text format. Set the code column to "text"; otherwise the program turns 0142 into 142 and the difference creates itself.
  2. Clean code column. In a new column, write the code converted to uppercase with spaces and hyphens removed: "tsh 0142 m" and "TSH-0142-M" both become "TSH0142M".
  3. Matching. Join the master list and the channel list on this clean code; those found on both sides are the same products with spelling differences.
  4. Left on one side. Those only in the master list are products not published on the channel; those only on the channel are passive, wrong or empty-code listings.
  5. Duplicates. Codes that appear more than once in the clean code column are duplicate codes; list them separately.

This method scans a catalog of 1,200 rows in a few minutes. However, when you find a product that matches on the clean code, remember to correct the code on the channel to the rule as well; if you only match it in the audit and leave it, the system will still see two separate codes. After making the fix, check by hand on one example that the product's inventory and price are the same in both places.

What are the most common mistakes in product code matching?

  1. Changing the code by hand on the channel. The link to the master record breaks.
  2. Not keeping the rule in writing. Everyone uses their own spelling.
  3. Not giving variants separate codes. Size and color inventory gets mixed up.
  4. Deleting the leading zero. 0142 and 142 become different codes; spreadsheet programs can do this silently.
  5. Leaving a passive product open. An old listing shows the wrong inventory.
  6. Doing the check once and stopping. The rate drops again as new products arrive.
  7. Looking only at the overall rate. Without the error type, the fix is unclear.

How does NetKasam make product and inventory matching easier?

  • Inventory management. Inventory tracking by product, variant and warehouse is done in a single master record (inventory management).
  • Marketplace integration. Marketplace orders are invoiced automatically and posted to bookkeeping (marketplace integration).
  • e-Commerce integration. Orders from your own site enter the same flow (e-commerce integration).
  • Single-screen order management. Orders from channels appear in a single list (order management).
  • Saved reports. You save inventory and sales reports with their criteria and export them to Excel; you can prepare the matching table with this data (reports).
  • Bookkeeping is free. e-Invoice, customer and supplier accounts, inventory, cash, bank and expense management are free in every plan; see the pricing page for details.

Mini quiz: Test what you have learned

In short: Answer the 5 questions below, then compare with the answer key. 4 or 5 correct shows you have completed the training.

  1. If 1,104 of 1,200 listings are matched, what is the match rate?
  2. Of 96 unmatched listings, 38 are spelling differences and 24 are empty codes. What percentage of the unmatched ones are these two?
  3. If the cancellation cost of 18 orders is ₺85 per order, what is the total in ₺?
  4. If 24 listings miss 2 orders a month and the average basket is ₺300, what is the lost revenue?
  5. What is the match rate after the first three fixes? (76 listings were fixed)

Answer key:

  1. 1.104 ÷ 1.200 = %92.
  2. (38 + 24) ÷ 96 = 62 ÷ 96 ≈ %64,6.
  3. 18 × 85 = 1.530 ₺.
  4. 24 × 2 × 300 = 14.400 ₺.
  5. (1.104 + 76) ÷ 1.200 = 1.180 ÷ 1.200 ≈ %98,3.

Frequently Asked Questions About Product Code Matching

What is a product code (SKU)?

It is a unique code that identifies a product and its variant, set by the business itself. Inventory, prices and orders are linked to products through this code.

How can you tell there is a matching error?

Typical signs are a product that is in stock showing as unavailable on a channel, a product being sold after it has run out, or a price not updating. To be certain, the codes are compared.

How is the match rate calculated?

The number of matched listings is divided by the total number of listings. For 1,104 matches out of 1,200 listings, the rate is 92%.

Is a barcode the same as a product code?

No. A barcode is used to scan a product; a product code is the business's internal identifier. The relationship between them needs to be kept in the master record.

Why do variants get separate codes?

Each size or color has its own inventory. If they are linked with a single code, inventory gets mixed up and the wrong variant can be sold.

Can I change the code later?

Yes, but only in the master record and by pushing it to the channels. Changing it by hand on a channel breaks the link; also note the relationship with past orders.

How often should the check be done?

Check new products before they go live, do a quick check weekly and a full audit monthly. Review the code rule once a quarter.

What can be done to stop errors from repeating?

Make the code field mandatory, manage the code from a single source and match new products before they go live. Close passive products on channels regularly.

Conclusion: Fix the code and inventory and prices will flow correctly on their own

The most useful thing you can do today is export one channel's product list and compare it with the master record by code. Sort the unmatched items into five error types, fix first those that cause lost inventory sales, and put the code rule in writing. Repeat the same check next month and see the rate go up.

Bookkeeping Is Now Completely Free

Manage your bookkeeping and e-Invoice operations with NetKasam without paying a thing. Customer and supplier accounts, inventory, collections and more in one panel, completely free.

Start FreeInventory Tracking

Read similar articles

19 free tools

Commission, shipping, VAT and profit calculations in your browser, no sign-up needed.

All tools