Research methodology and data standards
How BankCreds Research sources, stores, calculates, reviews, updates and corrects the data and statistics it publishes about Indian consumer finance.
What counts as research
A page is published under BankCreds Research only if it contains original data or measurable analysis: historical or time-series analysis, statistics, market or consumer analysis, a proprietary calculation or index, geographic analysis, financial modelling, or a comparison based on measurable data. Guides, opinion and product explainers belong in the blog and product sections.
Where the data comes from
Official publishers first: the Reserve Bank of India, government ministries and statistical offices, and lenders’ own published rate and fee pages. Aggregators, comparison sites and unsourced figures are never used as a source. Each dataset records its publisher, the exact table or series, the URLs used and the date it was retrieved.
How data is stored
Values are stored exactly as published, with the source of each value. Periods the publisher has not released are left blank — never interpolated, estimated or filled from another series. Every dataset has a version number and a changelog, and every earlier version is retained so past figures can be reproduced.
How numbers are calculated
Every statistic in a report is computed from the stored dataset by code at the moment the page is built; no figure is typed by hand. The page, its charts and its PDF therefore always agree. Each report’s methodology section states its formulas, windows and rounding.
Review before publication
Reports are written by the BankCreds Research & Data Team, reviewed for financial accuracy by the BankCreds Financial Experts and fact-checked by the BankCreds Content & SEO Team. A report is published only when every figure traces to its dataset and every placeholder resolves.
What we will not publish
Forecasts or price predictions; “best time to buy” or “guaranteed return” claims; guarantees of loan approval or outcomes; historical correlation or seasonality presented as a prediction; statistics without a named source; rounding that changes a conclusion.
Updates and versions
Living datasets are updated in place at the same URL when the publisher releases new data, and the dataset version is raised. Annual reports receive a new URL each year and earlier editions are kept unchanged. A methodology change is versioned and noted on the page. A report replaced by a better one is redirected to it; a discontinued dataset stays online, marked as archived.
Corrections
If a published figure is wrong we correct it in place, record the change in the report’s version history and, where material, list it on our corrections page. Report errors through our contact page.
Related policies: editorial policy · fact-checking policy · corrections · use our data.