Walkthrough · No. 05

Data Maps: Guided discovery across 990 datasets, seed to market

How a keyword-searchable catalogue of agriculture datasets became a ten-stage map from seed to market.

Data Maps agriculture sector map: stage cards above a stippled illustration of a farmer with bullocks and a cart.
Fig. 1 The agriculture Data Map on dataful.in. Factly product

§ 01 Problem

The problem

Dataful holds hundreds of agriculture datasets. Keyword search suits people who know what they want; for everyone else the catalogue was a wall.

§ 02 Information and data

What made it hard

  1. Datasets differ in level (all-India, state, district), frequency and time coverage.

  2. Relevant data sits in other sectors: bank credit, trade, rural crime.

  3. The catalogue keeps growing, so any hand-built structure goes stale.

§ 03 Workflow and design

The decisions that shaped it

  • D1

    A narrative taxonomy, not a filing system

    Ten stages that follow the farming year, from Land & Soil to Trade & Economy.

  • D2

    Three shapes of collection

    A state grid for state-wise data, one card per crop or commodity, a plain list otherwise.

  • D3

    Cross-sector by design

    Datasets from other sectors are included and flagged.

  • D4

    Generated, versioned, static

    Produced from the catalogue as a dated bundle; it regenerates as the catalogue grows.

§ 04 Shipped, not planned

What it actually does

  • Sector → stage → collection → dataset drill-down, in English and Hindi
  • State grids for state-wise data; one card per crop or commodity
  • Every dataset links to its Dataful page: preview, metadata, citation, downloads

§ 05 Input → steps → output

Walkthrough

  1. Input

    The Dataful agriculture catalogue: several hundred datasets plus related datasets from other sectors.

  2. Step 1: Start at the sector

    The root node shows the sector and its ten stages. A newcomer picks “Rain & Water” and is right.

    Data Maps "Agriculture and Allied" knowledge map: a root node branching into ten stage cards from land and soil to trade.
    Fig. 2 The root of the map: one sector, ten stages.
  3. Step 2: Open a stage

    Rain & Water: 113 datasets in seven collections, with their source sectors.

    Data Maps drill-down into the "Rain & Water" stage with its dataset collections.
    Fig. 3 A stage and its collections.
  4. Step 3: Open a collection

    Irrigated Area by Source is state-wise, so it renders as a grid of states.

    Data Maps drill-down to the "Irrigated Area by Source" collection with an India state tile-grid map.
    Fig. 4 A collection as a state grid.
  5. Step 4: Land on the dataset

    Each tile opens the dataset’s Dataful page.

  6. Output

    From “what do we know about irrigation in Telangana?” to a documented table in four clicks.

§ 06 Evaluation

How we check it

  • Every agriculture dataset is reachable from a stage; cross-sector datasets are flagged.

  • The map regenerates from the catalogue, so coverage is re-checked on each release.

§ 07 Evidence

The figures

  • 10[1]

    stages, seed to market

  • 63[1]

    collections

  • 990[1]

    dataset records, including cross-sector data

    Catalogue snapshot of 16 October 2026.

§ 08 Limitations and lessons

Limitations and lessons

read this part first

Limitations

  • Only the agriculture map is live so far.
  • A dataset that fits two stages appears where it is most useful: a judgement.

Lessons from operation

  1. Taxonomy is a product decision. The seed-to-market story makes the map usable by newcomers.
  2. A growing catalogue needs a generated structure, not a hand-edited one.
  3. Show the shape of the data before the data.

Sources (3)

  1. [1] Agriculture Data Map catalogue (bundle of 2026-10-16) · datamaps.dataful.in (opens external site)
  2. [2] Data Map: Agriculture and Allied · dataful.in (opens external site)
  3. [3] Dataful: a dataset page · dataful.in (opens external site)

Sources reviewed 15 November 2026.

Discuss a project

Have a catalogue people cannot navigate?

We design the taxonomy and build the map or portal around it.