Tools / Development Activity

Development Activity for Indian Plots

Is this plot vacant? How much of the parcel is built-up? What's the neighbourhood density? See 8 years of building footprint data — real measurements, no assumptions.

What does this tool tell you?

Before committing to any plot, three questions get asked in order: Is the plot itself vacant? Is the neighbourhood built-up enough to support what I want to build? Is the area still growing or has it saturated? This tool answers all three from satellite data — no surveys, no site visits, no broker claims.

Parcel footprint: how much is actually buildable

We measure three things about the plot itself:

Total parcel area — the area of the polygon you drew or the 50 m circle around your pin.
Built-up footprint — how many m² of that parcel already have building structures on it, as detected by the VIDA satellite dataset.
Open / unbuilt area — what remains after subtracting the footprint. This is the gross upper bound on what can be built. Note: road setbacks are not yet subtracted.

Neighbourhood density: total built-up area vs coverage percentage

The 500 m buffer analysis gives you two complementary numbers:

Total built-up area (m²) — the absolute amount of construction in the neighbourhood. A coverage of 25% in a dense urban zone can mean 200,000 m² of buildings — that tells you far more than the percentage alone.
Coverage percentage — what fraction of the 500 m area is covered by footprints. Useful for comparing across different geographies.
Typical building size (median footprint) — the median m² footprint of all buildings within 500 m. A raw measurement from the VIDA dataset, directly from the satellite data with no classification applied.

Construction CAGR and height growth — what FSI is doing

Construction CAGR is the compound annual growth rate of built-up area in the 500 m buffer over 2016–2023, computed via log-linear regression across all 8 annual data points. Height growth is the change in average building height over the same window. They reveal different things:

High CAGR, low height growth — the area is densifying horizontally. New plots are being built out for the first time. Typical of peri-urban expansion edges.
Low CAGR, high height growth — the area is already built-out and is now adding floors. FSI is being consumed. Typical of mature urban cores like Bandra-Kurla extensions or Gachibowli.
High CAGR + high height growth — the area is in the middle of a development boom. The classic Tier-2 city growth corridor pattern.
Low both — saturation or limited demand. Look elsewhere for growth plays.

Why parcel-level vacancy and 500 m neighbourhood matter together

A vacant plot in HITEC City Hyderabad is a fundamentally different opportunity from a vacant plot of the same size in a sparse village. The HITEC plot has water, sewage, roads, schools, hospitals, and an IT-park demand pool within 2 km. The village plot has none of those. Vacancy alone does not tell the full story — the 500 m neighbourhood context is what separates infill from greenfield.

What do these numbers mean?

The result card shows three groups: what is on the plot, what is around it, and how the neighbourhood has changed over 8 years. Every number is a direct measurement from the satellite dataset.

Parcel breakdown

Vacancy status, total parcel area, how much is already built-up (VIDA footprint), and how much remains open. The open area is the gross upper bound for new construction — road setbacks are not yet subtracted.

Example: “2,500 m² total · 300 m² built-up · 2,200 m² open” means an 88% unbuilt parcel with some existing structures.

Neighbourhood density (500 m)

Total built-up area in m² and the equivalent coverage percentage within 500 m. Both numbers together give you scale — the absolute area tells you how much construction exists; the percentage tells you how densely packed it is relative to available land.

Example: “186,000 m² · 23.7% coverage” means substantial construction but still room to grow.

Typical building size

The median footprint of all buildings within 500 m, in m². Directly from the VIDA dataset — the middle value when all building footprints in the neighbourhood are sorted by size. No classification threshold is applied.

Example: “68 m² median footprint” means half the buildings in the neighbourhood are smaller than 68 m² and half are larger.

Construction CAGR (8-yr)

Compound annual growth rate of built-up area in the 500 m buffer from 2016 to 2023. Computed via log-linear regression across all 8 annual data points — not just comparing 2016 to 2023 — which makes it more robust against any single anomalous year.

Year-by-year chart is also shown so you can see exactly which years drove the growth and which were flat.

Building height (avg + growth)

Average building height in the 500 m buffer in 2023, and how much it has grown since 2016. Height growth tells you whether the area is adding floors (vertical densification, FSI consumption) rather than just spreading outward.

Example: “14.8 m avg · +1.3 m since 2016” suggests modest upward densification in an otherwise stable neighbourhood.

How the analysis works

Three steps. First-time areas take 30 s to a few minutes; cached areas return in under 1 s.

You mark the plot

Draw the boundary, drop a pin, or paste an address. Pin inputs are buffered to a 50 m circle. We extend the analysis to a 500 m buffer around the plot to capture the surroundings.

We check parcel + 500 m, 8 years deep

VIDA Combined Buildings (Google + Microsoft + OSM) gives us parcel-level footprint measurements — vacancy, built-up area, building count, density, and typical building size. Google Open Buildings Temporal V1 gives us annual built-up area and average building height for every year from 2016 to 2023.

You see the raw measurements

Parcel area, built-up footprint, open area, building count, total neighbourhood built-up area, coverage %, median building size, 8-year area chart, CAGR, and height growth. Every number is a direct measurement from the dataset.

Methodology and data sources

We use two complementary satellite building datasets, both publicly available with no API key.

Where the data comes from

VIDA Combined Buildings is a deduplicated merge of Google Open Buildings + Microsoft GlobalMLBuildingFootprints + OSM, distributed as a 64 GB GeoParquet for India alone. We pull individual parcels and buffers on demand via DuckDB bbox queries, cache results in PostGIS by S2 cell, and filter to confidence > 0.65 at fetch time. This gives us parcel-grain footprint geometry for vacancy, built-up area, density, building count, and median footprint size.

Google Open Buildings Temporal V1 publishes annual mid-year composites of building presence and height for 2016 through 2023 — 8 years total. Each composite is a set of 12.5 km × 12.5 km GeoTIFFs at 0.5 m resolution streamed on demand from Google Cloud Storage via HTTP range reads (no local raster storage). We sample all 8 area years and 2 height years (2016 and 2023) in parallel. CAGR is computed via log-linear OLS regression across all 8 area data points.

VIDA Combined BuildingsGoogle + Microsoft + OSMGoogle Open Buildings Temporal V12016–2023 (8 annual composites)

Coverage and limitations

VIDA confidence threshold. We filter to > 0.65 confidence — a moderately conservative cut that removes most false positives in agricultural land while keeping nearly all real buildings.
Temporal pixel scale. 0.5 m/px Google composites are detailed enough to catch most buildings, but very small structures (under 9 m²) may be missed.
Cold-fetch latency. First queries on uncached areas can take 30 s to a few minutes because the 64 GB VIDA file is streamed over HTTPS. Once an area is cached in PostGIS, repeat queries return in under 1 s.
Water body pre-check pending. Querying a point in a lake or reservoir will currently return “vacant + sparse surroundings” — technically correct but not useful. Bhuvan LULC integration is in progress to flag water bodies up-front.
Road setbacks not subtracted. Net developable area does not subtract road right-of-way buffers. Treat it as a gross upper bound.
Height data at 2016 and 2023 only. We sample average building height at the two endpoints of the 8-year window. Intermediate height values are not available in the current dataset.

What we don’t do

We do not certify title, zoning compliance, or lending eligibility. All numbers are direct measurements from satellite-derived datasets. No interpretation labels — such as “Informal”, “Formal”, “Rapid”, or “Accelerating” — are applied to the output. Any interpretation of what the raw numbers mean for a specific transaction is the user’s own judgement.

Frequently Asked Questions

Where does the building data come from?

Two complementary satellite-derived datasets. VIDA Combined Buildings is a deduplicated merge of Google Open Buildings + Microsoft GlobalMLBuildingFootprints + OSM — used for parcel-level vacancy, footprint area, density, building count, and median building size. Google Open Buildings Temporal V1 provides annual built-up area and average building height for 2016–2023 — used for the 8-year growth trajectory and CAGR. Both are publicly accessible with no API key.

Why does the 500 m neighbourhood matter so much?

A vacant plot inside a dense neighbourhood is fundamentally different from a vacant plot in a sparse village. The first has utilities, roads, schools, hospitals, and a demand pool within walking distance. The second has none of those. Same vacancy, completely different opportunity. The 500 m surroundings is what tells you which one you're looking at.

What does Construction CAGR tell me?

Compound Annual Growth Rate of built-up area in the 500 m buffer over 8 years (2016–2023). It is computed via log-linear regression across all 8 annual data points — not just comparing the 2016 and 2023 endpoints — which makes it more robust if any single year had an anomalous reading. A positive CAGR means the neighbourhood has been gaining built-up area year-on-year; a negative CAGR means it has been shrinking. The year-by-year bar chart shows you exactly which years drove the growth.

What does the median building footprint tell me?

The median footprint of all VIDA buildings within 500 m is the middle value when every building footprint in the neighbourhood is sorted by size. It tells you the typical scale of construction in that area — whether buildings are small and dense or large and spread out. It is a raw measurement from the dataset with no classification applied on our end.

Why does the first query take so long?

The VIDA building dataset is a 64 GB GeoParquet file streamed over HTTPS. The first time we see an area, we fetch and cache the relevant section — typically 30 seconds to a few minutes depending on density. Once cached in our PostGIS database, repeat queries on that area return in under 1 second. The 8-year Google Temporal rasters are also streamed on demand from Google Cloud Storage for each new query area.