Recipe JSON-LD for small food websites: a practical checklist

Łatwe Przepisy — Sat, 12 Sep 2026 15:27:16 +0000

Recipe websites often start with a simple publishing workflow: a title, a photo, ingredients and preparation steps. That is enough for readers, but search engines and recipe tools need a little more structure to understand the page clearly.

JSON-LD is usually the least intrusive way to add that structure. It sits in the HTML as machine-readable metadata, while the visible recipe can stay clean and useful for people.

What a basic recipe object should include

A small food site does not need a complicated schema setup on day one. Start with the fields that are stable and easy to maintain:

recipe name
short description
author or publisher
ingredient list
preparation time
cooking time
serving yield
image, when the image is original and public

Those fields are enough to describe most everyday recipes without creating a maintenance burden.

Keep the visible recipe and JSON-LD aligned

The structured data should describe what is actually on the page. If the page says the recipe serves four people, the JSON-LD should not say six. If the article lists ten ingredients, the metadata should not include a copied ingredient list from another version of the recipe.

This sounds obvious, but it is where many small recipe archives become messy over time. The easiest habit is to generate the JSON-LD from the same notes used to prepare the visible recipe.

Example

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Recipe",
  "name": "Simple yogurt cake",
  "description": "A quick everyday cake with yogurt and apples.",
  "recipeYield": "6 servings",
  "prepTime": "PT15M",
  "cookTime": "PT40M",
  "recipeIngredient": [
    "2 cups flour",
    "2 eggs",
    "1 cup yogurt",
    "2 apples"
  ],
  "author": {
    "@type": "Organization",
    "name": "Recipe publisher"
  }
}
</script>

A small workflow that prevents errors

Draft the human-readable recipe first.
Normalize serving size and ingredient quantities.
Add preparation and cooking time in minutes.
Generate the JSON-LD.
Check that the metadata and page content still match after editing.

For Polish recipe publishers, the same workflow works well for everyday cooking categories such as dinners, desserts and chicken recipes. I keep a simple recipe site at Łatwe Przepisy and use this kind of checklist when thinking about clean recipe pages and category hubs.

Common mistakes

Do not mark up a category page as a single recipe. A dessert category, for example, is usually a collection page, not one recipe.

Do not add fake ratings or review counts. If users have not left visible reviews, the structured data should not claim that they have.

Do not copy JSON-LD from another recipe and forget to change the ingredients, time or yield. That creates confusing signals and makes future cleanup painful.

A tiny helper approach

For small teams, even a plain static tool can be enough: input the title, servings, times and ingredients, then copy the JSON-LD snippet into the publishing workflow. The useful part is not the tool itself, but the repeatable habit it creates.

Structured data works best when it reflects a real editorial process. Keep the recipe useful for people first, then make it easier for machines to understand.