The product page already shows 12 900 ₽, but the JSON-LD still shows 11 900 ₽. JSON-LD is structured product data recorded in HTML for search engines. The price shown to shoppers is correct, but the markup tells search engines something else. This is usually caused by a separate markup generator, an outdated cache, or data for a different product variant being included in the JSON-LD.
Start with a specific product: save its URL, SKU, selected variant, price, and availability as seen by a visitor who is not signed in. Then find the corresponding Product and Offer objects in the markup. You cannot compare the first number you find with the entire page: a recommended product or a different configuration may appear nearby.
Put data for the same product in a table
Suppose you find the following after importing the catalog:
| SKU | Catalog data | Visible product page | JSON-LD |
|---|---|---|---|
| A-101 | 12 900 ₽ | 12 900 ₽ | 11 900 ₽ |
| A-102 | Out of stock | Out of stock | InStock |
| A-103 | Sale: 6 000 ₽ | 6 000 ₽ | 7 000 ₽ |
This shows that the problem is not just the price. For A-102, search engines are being sent incorrect availability information. Add the time of the check to the table: prices may have changed between opening the product page and reading an old file.
Find out which price the site shows to everyone and which it shows only to signed-in customers, wholesalers, or program members. Do not transfer a personalized price into public markup without checking the page rules and the selected offer type.
Read all JSON-LD blocks
JSON-LD is usually located in script with the type application/ld+json. It may be generated by both the template and an SEO plugin. You can read each block separately in the browser console:
[...document.querySelectorAll('script[type="application/ld+json"]')]
.map((node, index) => {
try {
return { index, data: JSON.parse(node.textContent) };
} catch (error) {
return { index, error: error.message };
}
});An error in one block will not stop the others from being read. Find the relevant Product, including inside @graph, and match its sku, URL, and offers. If the data is generated by JavaScript, check both the source HTML and the result after the page loads.
Fix the source of the value
Ask a developer to find out why the product page and markup use different data. For example, the product page may use the current sale price, while the JSON-LD uses the regular price from a separate field. The correct fix must account for when the promotion starts and ends, not just the product's price today.
If the cache is responsible, check that it refreshes after a regular import. Clearing it manually once may show the new result, but it does not prevent the problem from recurring the next day. Check availability in the same way after changing stock levels.
An Offer snippet for A-101 might look like this:
{
"@type": "Offer",
"url": "https://shop.example/product/a-101/",
"price": "12900.00",
"priceCurrency": "RUB",
"availability": "https://schema.org/InStock"
}This is only an Offer, not complete product markup. The required fields depend on the result type; refer to Google's Product documentation. The figures here are examples and must not be copied into a real product page without being replaced.
Check changes across different products
Choose a regular product, one on sale, an out-of-stock product, and a product with variants. For each, compare the public product page with the Offer that corresponds to it. Change the variant separately: price and availability may change, so checking only the initial page load is not enough.
Then repeat the import or make a standard change in the CMS and confirm that the discrepancy has not returned. Check whether a second plugin has left an outdated Product alongside the corrected one.
How to finish the check
Open Google's Rich Results Test, enter the public product URL, and run the test. In the product data it finds, check for errors and Offer objects. The test checks the markup, but it does not know your actual price: go back to the table and confirm the values. The result for the developer should be a corrected data source and several checked product pages.
Common problems are described in the Schema.org errors guide, and the markup is explained in the structured data guide. After updating the template, run a technical audit to check the pages along with the other changes.