In brief: Structured data on a website describes the meaning of content using a machine-readable vocabulary, often with Schema.org and JSON-LD. It helps search engines understand entities, but does not guarantee a rich result.
How I use structured data in practice
I incorporate structured data only once it is clear who collects the data and who acts on it. First, I verify the event in the browser and source system, then its transfer to the report, and finally its meaning for the business. I document differences between platforms instead of forcibly reconciling them.
What to watch out for
A tool is not a source of truth in itself. With structured data, I need to know the limits of collection and the possibility of correction. Without checking duplicates and missing events, a precise calculation may rest on faulty input.
Questions for decision-making
- How exactly do we calculate the value, and from which source?
- Which decision will change if the number rises or falls?
- What can distort or duplicate the measurement?
- What do we compare the value with so that it has commercial meaning?