How KalQLater works
We do not publish a conclusion first and search for numbers later. The research question, source definitions and limitations come before the narrative.
1. Start with a falsifiable question
Good KalQLater topics can be answered with evidence. “Are Indians really switching to EVs?” is measurable. “Why EVs are the future” is already a conclusion.
2. Prefer primary public data
We prioritise government statistical agencies, official registries, international organisations and first-party datasets. Secondary reports are useful for context but should not replace the underlying source when the source is available.
3. Keep unlike measures separate
Registrations are not sales. Job postings are not jobs filled. Visitor arrivals are not unique travellers. Search interest is not demand. When two datasets use different definitions, we label the distinction rather than averaging them into a cleaner number.
4. Calculate transparently
Derived statistics are labelled as KalQLater calculations. Downloadable CSVs are provided where practical so readers can recreate the arithmetic.
5. Separate fact from interpretation
Observed values are written as findings. Explanations are written as possible explanations unless a source establishes causation.
6. Publish limitations
Every flagship analysis includes a section on what the data does not prove. This is not legal boilerplate; it is part of the analysis.
7. Update visibly
Substantive data changes update the visible modified date. Corrections are handled under our corrections policy.
Rendering and SEO
Public research pages are static or server-rendered HTML. Core article copy, tables, metadata and structured data do not depend on client-side API calls. This makes the page faster for users and gives crawlers the same substantive content a reader sees.