Skip to content
inroads market

Research methodology

How we collect, verify, date and update the data in our reports.

1. Defining the market

Every report starts with a written definition of the market:

  • Product or service scope: what is included and excluded. For example, the California grape crush includes raisin and table grapes that are crushed, so it is not a wine-grape-only figure.
  • Geography: the state, and where relevant the counties or districts that matter.
  • Value chain position: growers, processors, distributors, retailers, service providers.
  • Time period: the most recent complete data year, stated explicitly.

A clear definition prevents the most common research error: combining numbers that measure different things.

2. Data collection

We use sources in this order of preference:

  1. Official statistics: USDA National Agricultural Statistics Service, USDA Economic Research Service, the U.S. Census Bureau, the Bureau of Labor Statistics, the Bureau of Economic Analysis, the Alcohol and Tobacco Tax and Trade Bureau, and state agencies such as the California Department of Food and Agriculture.
  2. Regulatory filings and registries: permit lists, licensing data and public company filings.
  3. Academic and extension research: university extension services and peer-reviewed studies.
  4. Trade associations and industry bodies: used with attribution and labeled as such, because their methods and interests vary.
  5. Company information: company websites and announcements, used for describing who operates in a market, never for unverified market sizes.

We do not reproduce paid commercial research, and we do not use commercial market-size estimates whose methodology is not public.

3. Source evaluation

For each figure we record:

  • the source document and URL;
  • the metric definition, exactly as the source defines it;
  • the data period and geographic scope;
  • the date we retrieved it;
  • methodology notes (preliminary versus final data, sample-based surveys, revisions).

We check figures against the source document itself, not a press summary. Trade-association and survey figures are kept, when useful, but marked as such.

4. Observed facts, calculations and interpretation

Our reports distinguish three kinds of statement:

  • Observed facts: published values from a source (for example, tons crushed in a crop year).
  • Calculations: values we derive from published data, such as a state's share of a national total. We show the inputs and the method.
  • Interpretation: our analysis of what the data means for a business decision. It is written as analysis, not presented as data.

5. Geographic relevance

State boundaries matter, and so does how data is collected. Two measurements are often confused:

  • data about a state (for example, production located in California), and
  • data from people in a state (for example, consumers located in California).

When national data is the only available measure, we say so and explain how it relates to the state market rather than presenting it as a state figure.

6. Limits of market estimates

Official statistics rarely measure a "market" exactly as a business defines it. Where no reliable revenue estimate exists, we do not invent one. We may provide a transparent, assumption-based estimate, clearly labeled, with each assumption stated so you can replace it with your own.

7. Publication dates and editions

Each report shows:

  • an edition label (for example, "2026 edition");
  • the publication date of that edition;
  • a last-verified date, the most recent date on which its figures were checked against sources.

We do not relabel older content as current. A report is marked as updated only when its underlying facts have actually been reviewed.

8. Updates

We schedule reviews when the official sources behind a report publish new data. Most agricultural statistics are annual (for example, final grape crush data is released each spring), so most reports are reviewed at least once a year. Facts that pass their review date are flagged internally and shown with their original period until re-verified.

9. Corrections

If you believe a figure or statement in our research is wrong, tell us. We review every correction request, record the outcome, and correct the report and its page when we find an error. Material corrections are noted on the report page.

10. Use of AI tools

We may use software tools, including AI-assisted drafting and data processing, to help organize research. Nothing is published without editorial review against the standards above, and no figure is published unless it can be traced to a source.