Onboarding and agreeing the scope of structured data optimisation
Onboarding and agreeing the scope consists of collecting the information needed to plan the optimisation of structured data for your site. At this stage I identify the page types present on the website, the visibility goals in rich results and the CMS limitations that may affect the implementation. I also establish which templates and sections are to be covered by the implementation, so that the work can be carried out in the real conditions of the site.
To get started, starting materials are needed that make it possible to move quickly on to the audit and the design of the target markup. What is required are the URLs of key pages, access to the code or templates or guaranteed implementation support, as well as information about the data sources on the site (e.g. product, article or company modules). Based on this data, a decision is made on which page types we cover—the priority templates are selected (e.g. product, article, category, company page) according to their potential and the implementation options.
- Collecting information about the page types and sections planned to be covered by the implementation
- Establishing the visibility goals in rich results and the limitations resulting from the CMS
- Verifying access and materials: URLs of key pages, access to the code/templates or implementation support, data sources on the site.
- Choosing the priority templates for further work based on potential and implementation options
Audit of existing structured data and quality analysis
The audit of existing structured data and quality analysis is a review of the current markup for consistency with the content and compliance with the requirements for the selected rich results. In practice I check what is already implemented, in what form, and whether it is uniform across the site. This stage is carried out for the previously selected page types, so that the assessment concerns priority templates rather than random URLs.
As part of the audit I carry out a markup inventory, i.e. I check where and in what form structured data appears (e.g. JSON-LD, microformats) and whether there is any duplication or conflict between sources. I then assess the completeness and quality of the fields: the presence of required and recommended properties, the correctness of formats (URLs, dates, prices), identifiers and the consistency of the values with the content of the page. In parallel, I identify validation errors and warnings and point out possible causes, such as missing fields, wrong types or incorrect nesting.
Another element of the analysis is the mapping of entities and relationships, i.e. establishing which entities the site represents (e.g. organisation, products, articles) and how they should be connected with one another (e.g. author, brand, breadcrumbs, offer). The condition of consistency with the page content applies here: the markup must not declare data that the user does not see or that is untrue. If such a discrepancy occurs, either the content has to be changed or part of the markup for the given template has to be dropped.
- Inventory of structured data and detection of duplication or conflicts between sources
- Assessment of the completeness and quality of properties and the consistency of values with the page content
- Detection of validation errors and warnings, with their causes indicated
- Mapping of entities and relationships on the site for consistent linking of data between page types
Designing the target structured data schema
Designing the target structured data schema means planning which data types and properties are ultimately to work on the selected templates and which consistency rules are to apply across the whole site. At this stage I translate the audit findings into a specific design: what we implement, where, and how it is to be fed with data from the CMS. As part of the work I also establish which elements are to be shared between templates (e.g. organisation data), to avoid inconsistencies. The outcome is a description of the target data layout, ready to be implemented in the chosen model.
As part of the design I select the types and properties, i.e. I choose the appropriate types and set of fields (required and recommended) and assign data sources on the site to them. In parallel, I set the rules for identifiers and consistency: how we identify entities, how we connect them between pages and how we keep URLs consistent (e.g. canonical URLs, images, organisation data). Part of this stage is also the decision: template-based or manual implementation, chosen depending on the implementation options (a scalable implementation in the templates or a targeted one on selected pages). Everything is organised in the form of documentation, which is the basis for implementation.
- Designing the target data types and consistency rules for the selected templates
- Selecting types and properties and defining the data sources on the site for each field
- Setting entity identifiers, connections between pages and URL consistency (including canonical URLs, images, organisation data).
- A decision on the implementation approach: template-based (scalable) or manual (targeted)
- Preparing the documentation package: scope, mapping of page types → schema, list of fields and data sources, consistency rules, test checklist, maintenance tips.
Implementing changes to structured data
Implementing changes to structured data consists of delivering the code and instructions, followed by implementation on the site by the implementation team or the client. In practice I hand over ready-made markup fragments and a description of where they are to be embedded in the templates or components. I attach guidelines to the implementation that help map fields correctly to data in the CMS and control the conditions under which they are rendered on pages. The implementation stage also supplements the previously prepared documentation package with the technical elements related to the implementation.
At this stage I hand over the structured data snippets together with information on which fields are required and where and how to place them. I also deliver implementation guidelines covering rendering rules, display conditions and a list of places that may generate conflicts between data sources. A template-based implementation usually requires support on the site’s side, including changes to templates or components and the ability to publish in the test and production environments. As a result, the implementation can be carried out consistently across the selected page types.
- Delivering structured data snippets to be embedded in templates or components
- Handing over implementation guidelines: mapping fields to data in the CMS, rendering rules, display conditions, potential sources of conflict.
- Implementation on the side of the client or the implementation team on the basis of the materials handed over
- Taking into account the requirements of a template-based implementation: changes to templates/components and publication in the test and production environments.
- Supplementing the documentation package with implementation-related elements
Verifying that the structured data implementation is correct
Verifying that the structured data implementation is correct means checking, after implementation, whether the markup is embedded in the right places and works without errors. At this stage I check whether the data is complete and whether generating it causes any conflicts on the rendering side. The verification is carried out on the implemented templates, to confirm that the solution works in practice and not just on a single example. In parallel, I check whether the declared values remain consistent with the content visible to the user on the page.
As part of the work I run tests on representative URLs from each template covered by the implementation. I validate the correctness of the data and nesting and the consistency with the content, and identify potential problems that need correcting. This stage culminates in a list of fixes and the implementation status, which sets out what has been implemented, what needs changing and what the priorities for further fixes are. Such material makes it possible to move on to corrections efficiently, without guessing where the cause of the errors lies.
- Tests on representative URLs from each implemented template
- Checking completeness, correct nesting and the consistency of the data with the content visible on the page
- Detecting errors and conflicts appearing on the rendering side
- Handing over the list of fixes and the implementation status: implemented elements, problems, priorities and recommended improvements.
Monitoring and maintaining the stability of structured data
Monitoring and maintaining the stability of structured data means periodically checking whether the implemented markup remains correct despite changes to content and templates. In practice it is about quickly catching situations in which the implementation has been overwritten or is no longer complete. As part of the maintenance I also check whether new or modified templates inherit the right structured data rules. This stage also includes responding to new errors and to changes in the requirements for rich results.
The core activity is monitoring errors and regressions, i.e. regular verification of the stability of the implementation and the quality of the data. If gaps or inconsistencies appear, I point out the areas that need correcting in order to restore the completeness and consistency of the markup. One limitation must be kept in mind at the same time: correct structured data increases the chance of qualifying for rich results, but whether rich results are actually displayed depends on the search engine’s decision, the quality of the page and the indexing process. Monitoring is therefore geared towards keeping the implementation correct and minimising the risk of regression, not towards declaring an effect in the SERP.
- Periodically checking whether the implementation has been overwritten after changes to content or templates
- Checking whether the data remains complete and whether new templates inherit the correct rules
- Identifying new errors and regressions and pointing out the places that need correcting
- Taking into account that the display of rich results depends on the search engine, the quality of the page and indexing
