Success Case: How Recipe Schema Markup Increased Organic Traffic

By Erik Bonsaksen on 2023-08-25

Success Case: How Recipe Schema Markup Increased Organic Traffic

We A/B split tested Recipe Schema Markup for SEO, revealing an estimated causal effect of:

+35.8%

Impression Uplift

+94.3%

Click Uplift

What Is Recipe Schema Markup?

Recipe structured data is a must-have for food bloggers and recipe websites. It's a schema markup type that feeds(!) search engines information such as ingredients, cooking time, nutrition and images they should care about.

Beyond the obvious benefits of enriching Google's knowledge graph with a better semantic understanding of your content (of course, right?), the schema markup for recipes has perhaps the most visually appealing rich snippets among its brethren. Decide for yourself.

Example of Recipe Schema Markup:

Example Recipe Schema Markup 1

And this:

Example Recipe Schema Markup 2

And that:

Example Recipe Schema Markup 3

How Does Recipe Schema Markup Work?

Simple. You add this type of code to the specific recipe's landing page. Naturally, the code should mirror the content of that page.

This is how you make a tasteful SEO tool:

{
  "@context": "https://schema.org/",
  "@type": "Recipe",
  "name": "Whimsical SEO Tool",
  "image": ["https://example.com/photos/1x1/seo-tool.jpg"],
  "author": {
    "@type": "Person",
    "name": "John CodeChef"
  },
  "datePublished": "2023-08-24",
  "description": "A delightful blend of code and creativity to cook up the perfect SEO tool.",
  "prepTime": "PT60M",
  "cookTime": "PT240M",
  "totalTime": "PT300M",
  "keywords": "SEO, coding, software baking",
  "recipeYield": "1 tool",
  "recipeCategory": "Software",
  "recipeCuisine": "Tech",
  "nutrition": {
    "@type": "NutritionInformation",
    "calories": "0 calories, but high in data richness!"
  },
  "recipeIngredient": [
    "2 cups of Python code",
    "1 cup of JavaScript for a smooth user interface",
    "A pinch of HTML and CSS for presentation",
    "3 tablespoons of API integrations",
    "1 quart of data analytics",
    "2 teaspoons of keyword research algorithms",
    "A dash of backlink analysis"
  ],
  "recipeInstructions": [
    {
      "@type": "HowToStep",
      "name": "Setup Environment",
      "text": "Preheat your IDE to its optimal settings. Ensure all plugins and extensions are installed.",
      "url": "https://example.com/seo-tool#step1",
      "image": "https://example.com/photos/seo-tool/step1.jpg"
    },
    {
      "@type": "HowToStep",
      "name": "Mix Code Base",
      "text": "In a vast digital bowl, combine Python, JavaScript, HTML, and CSS. Blend until the code is smooth and error-free.",
      "url": "https://example.com/seo-tool#step2",
      "image": "https://example.com/photos/seo-tool/step2.jpg"
    },
    {
      "@type": "HowToStep",
      "name": "Integrate APIs",
      "text": "Slowly pour in API integrations, ensuring a smooth connection with external data sources.",
      "url": "https://example.com/seo-tool#step3",
      "image": "https://example.com/photos/seo-tool/step3.jpg"
    },
    {
      "@type": "HowToStep",
      "name": "Serve with Analytics",
      "text": "Once your SEO tool is baked to perfection, serve with a side of comprehensive data analytics.",
      "url": "https://example.com/seo-tool#step4",
      "image": "https://example.com/photos/seo-tool/step4.jpg"
    }
  ],
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.9",
    "ratingCount": "527"
  },
  "video": {
    "@type": "VideoObject",
    "name": "Baking the Perfect SEO Tool",
    "description": "Discover the blend of code and strategy to cook up your SEO tool.",
    "thumbnailUrl": ["https://example.com/photos/1x1/seo-video.jpg"],
    "contentUrl": "https://www.example.com/video-seotool.mp4",
    "embedUrl": "https://www.example.com/videoplayer?video=seo",
    "uploadDate": "2023-07-20T08:00:00+08:00",
    "duration": "PT5M21S",
    "interactionStatistic": {
      "@type": "InteractionCounter",
      "interactionType": { "@type": "WatchAction" },
      "userInteractionCount": 5432
    },
    "expires": "2025-07-20T08:00:00+08:00"
  }
}

Mmmm, a hint of metal and ASCII, that's all you need.

When I say "add," I mean to embed JSON-LD into the HTML code of the page. While there are plugins available for CMS platforms like WordPress or Wix, and tools like Tag Manager for publishing schema markup, you can opt for a dedicated solution that automates and handles all the heavy lifting for integrating schemas. This solution is known as metamanager (yes, I'm biased).

Case A/B Split Test Results for Recipe Schema Markup

We made structured data from the recipe pages for one of Norway's biggest brand of sauces, soups and other yummy products. Their products are deeply ingrained in every kitchen throughout Norway, and their website receives enormous amounts of traffic. Could Recipe structured data really benefit their stats that much?

Apparently, yes.

Take a look at the A/B SEO split test we did for them:

Impressions:

SEO A/B Split Test - Impressions

Clicks:

SEO A/B Split Test - Clicks

Test configuration:

  • Period: 3 weeks
  • Training data: 1 months
  • Dataset: 50 vs 50 recipe pages.

Disclaimer: I wanted to have more training data, but there wasn't any available (new Google Search Console account).

Nevertheless, the test group, where the Recipe markup was added, saw a 35.8% increase in impressions and a 94.3% increase in clicks compared to the control group. That's insane! Our causal impact analysis framework also told us it was statistically significant due to the enormous amounts of traffic.

Can You Trust an SEO Split-Test?

After many years as an SEO consultant, I've learned that proving the impact of SEO initiatives is cumbersome. There are so many variables that make it hard to distinguish whether the SEO changes you've made can take all the credit.

Seasonality, a 50% sales campaign, or perhaps a Google algorithm change can render short-term comparisons worthless. Fortunately, split testing is a method that tries to solve this problem, and can be powerful for debunking SEO theories and attributing honour to us geeks when done correctly.

Training data: 1 months

This is the biggest caveat of this experiment. Too short, but the dataset was massive.

The key to a proper split test is:

  • Have a enough training data
  • You use only one page templates (recipe page, product page, etc.)
  • At least 50 URLs in each group
  • 30k or more organic impressions per month

In this study, we compared 50 vs. 50 URLs that had been live for a while, with the only variable being the addition of Recipe Schema Markup in group A, and nothing done to the URLs in group B.

  • Group A: Test group with Recipe Schema Markup
  • Group B: Control group without Recipe Schema Markup

It would be more precise to have 100s of URLs in each group, and 16 months of training data, but you work with what you have.

If nothing else, this is a strong indication, and I would say this experiment has enough leverage to make the ball rolling.

How Can You Add Recipe Schema Markup?

Just poke your developers and tell them to add the markup to each recipe page first thing in the morning.

Ha ha ha. SEO tasks are always left to rot in the menacing backlog of doom.

This is the reason I created metamanager after all - get shit done.

Take matters into your own hand, and deploy any schema markup you need on your website. It's a no-code tool solution that let's you pick any of the +800 schema markup types and publish. Manually, or automatically where we find the information, create the schema and publish it for you.

It's not too good to be true, it's just true.