Daylytix automatically generates valid JSON-LD schema markup for every page it crawls — no configuration required. During the audit, the crawler inspects each page and assigns it a schema type based on a set of detection signals. The result is a ready-to-paste code block you can drop straight into your <head>.

How page type detection works

Daylytix uses a priority-ordered detection system to decide which schema type fits each page. The system checks the following signals in order and stops as soon as a confident match is found:

1

Existing schema on the page

If the crawler finds a <script type="application/ld+json"> block already present, it reads the declared @type and uses that as the ground truth. The generator will still suggest improvements but it will not contradict the type you have already chosen.

2

Open Graph type tag

The og:type meta tag is checked next. A value of article maps to BlogPosting, product maps to Product, and website maps to WebSite or WebPage depending on the URL depth.

3

URL path patterns

Common path segments are matched: /blog/, /news/, /posts/ → BlogPosting; /product/, /shop/, /store/ → Product; /faq/, /faqs/ → FAQPage; /events/ → Event; /recipe/, /recipes/ → Recipe; /services/ → Service.

4

Title and heading hints

The page title and H1 are scanned for keywords such as "price", "buy", "review" (→ Product), "contact us", "get in touch" (→ LocalBusiness), "about us", "our team" (→ Organization), or numbered FAQ-style patterns.

5

FAQ item count

If the crawler detects structured question-and-answer blocks on the page (typically <details>/<summary> patterns or repeated heading+paragraph sequences) and finds three or more, it assigns FAQPage regardless of URL or title.

Supported schema types

Daylytix can generate JSON-LD for ten schema types. The table below lists each type along with the key properties included in the generated output.

Schema type Key properties generated Typical page
WebSite name, url, description, potentialAction (SearchAction) Homepage
WebPage name, url, description, inLanguage, isPartOf Generic interior page
BlogPosting headline, author, datePublished, dateModified, image, publisher Blog post or article
Product name, description, image, offers (price, currency), brand Product or shop page
FAQPage mainEntity (Question + acceptedAnswer pairs) FAQ page, accordion pages
LocalBusiness name, address, telephone, openingHours, geo, url Contact / location page
Organization name, url, logo, sameAs (social profiles), description About page
CollectionPage name, description, url, hasPart Category or archive page
Service name, description, provider, areaServed, serviceType Services page
Event name, startDate, endDate, location, organizer, description Event listing page
Recipe name, author, recipeIngredient, recipeInstructions, cookTime Recipe page

Accessing schema recommendations after an audit

Once your audit finishes, open the Schema Recommendations tab inside the audit dashboard. You will see a grid of all crawled pages with their detected page type and a status indicator showing whether schema is currently present or missing.

1

Open the Schema Recommendations tab

From the audit results screen, click the Schema tab in the tab bar. The page loads a table showing each crawled URL, its detected type, and whether schema markup was found on the live page.

2

Find a page that needs schema

Pages flagged with a red "Missing" badge are the highest priority. Use the filter box to narrow by page type or search for a specific URL.

3

Click "View Code"

Clicking the View Code button on any row opens a modal with the full generated JSON-LD block for that page. The code is pre-filled with data extracted from the page during the crawl — title, description, author, dates, prices, and more wherever available.

4

Copy and paste into your site

Click the Copy button in the top-right corner of the code modal. Paste the block inside your page's <head> element, or inject it via Google Tag Manager as a Custom HTML tag targeting the relevant page URL.

AI Enhancement with Claude

The rule-based generator produces solid, valid JSON-LD, but the AI Enhancement feature takes it further. Clicking the robot button (labeled Enhance with AI) in the View Code modal sends the page's crawl data to Claude AI, which rewrites and enriches the schema with:

  • More complete and naturally-worded descriptions tailored to the actual page content
  • Additional relevant properties the rule-based system may have left blank
  • Nested entities (e.g. adding author details to BlogPosting, or aggregateRating hints to Product)
  • Better keyword-aligned text in name and description fields to improve rich snippet relevance
⚠️
AI enhancement requires an Anthropic API key. Go to Settings → AI → Anthropic API Key and paste your key to enable this feature. The rule-based schema generator works without any API key. See Configuring your API key for step-by-step instructions.

Deploying the generated schema

There are three common ways to add the JSON-LD block to your site:

Paste directly into <head>

The simplest method. Copy the generated code block and paste it inside the <head> section of the relevant page template in your CMS theme or HTML file. This approach gives search engines the fastest possible access to the markup.

Use a CMS plugin or field

Most CMS platforms (WordPress, Webflow, Squarespace) have a per-page "Custom Code" or "Header Scripts" field. Paste the JSON-LD block there — it will only be injected on that specific page, keeping your templates clean.

Inject via Google Tag Manager

Create a new Custom HTML tag in GTM, paste the JSON-LD block, and set the trigger to fire on the specific page URL (use a Page URL regex trigger). This is the best approach for sites where you do not have direct template access. Remember to wrap the block in a <script type="application/ld+json"> tag when using GTM.

💡
Validate after deploying. Use Google's Rich Results Test or the Schema Markup Validator to confirm the deployed markup parses correctly before submitting your pages for indexing.

Reading the type counts summary

At the top of the Schema Recommendations tab, a summary grid shows how many pages Daylytix detected for each schema type during the crawl. This gives you a quick inventory — for example, if you have 40 BlogPosting pages but only 5 have schema markup, you can see at a glance how much work remains.

The grid also shows an overall schema coverage percentage — the share of crawled pages that already have valid JSON-LD. Aim for 100% on your most important page types (Product, BlogPosting, LocalBusiness) before working on lower-priority pages.

Was this article helpful?

Let us know so we can keep improving our documentation.