Franchise-site location intelligence on open government data
FranchiseIQ
A 0 to 100 franchise-viability score for any neighborhood, with a live pillar breakdown and a reverse mode that ranks every format for a location. Two live markets: Philippines (Quezon City) and Malaysia (Kuala Lumpur).
provenance · Philippines production pipeline (12 pilot barangays, keyed on PSGC (10-digit))
- Population
- PSA POPCEN 2024 / CPH 2020 (OpenSTAT)
- Income
- PSA small-area poverty estimates
- Traffic
- DPWH AADT (data.gov.ph)
- Points of interest
- OpenStreetMap Overpass
- Join key
- PSGC (10-digit)
Choosing a franchise site in Metro Manila is usually done on instinct. FranchiseIQ turns open government data into a defensible score, and shows its work for every pillar.
What it does
Pick a barangay and a franchise format and FranchiseIQ returns a single 0 to 100 viability score, a verdict, and a breakdown across four weighted pillars. Switch to reverse mode and it ranks all four formats for one location, so you can ask both “is this site good for a coffee shop?” and “what should go here?”
- DemandPopulation and density, normalized within the pilot city.
- Spending powerA poverty-inverse income index, with a soft floor penalty for premium formats.
- Foot trafficMeasured road traffic blended with format-specific anchor gravity.
- Competition gapResidents served per existing outlet: how much room is left.
How it is built
Each market joins every dataset on one canonical government identifier: the 10-digit PSGC (Philippine Standard Geographic Code) in the Philippines, the DOSM administrative code in Malaysia. Population, income, road traffic, and points of interest all key on that one code. Name matching is forbidden outside the single step that owns the code to name to geometry mapping, which is where most naive location tools quietly corrupt their results.
The scoring engine is a transparent weighted dot product, identical across markets, not a black box. It is deterministic and designed to port verbatim from this frontend into a Postgres or edge function, so the demo math and the production math are the same math. Adding a country is a data-adapter and weight-matrix change, not a rewrite; extending a country from one pilot city to the whole nation is an ingestion job, because the schema and the join key already assume national scale.
Honest limitations
Each live pilot runs on a clearly labeled seed dataset (12 Quezon City barangays for PH, 12 Kuala Lumpur zones for MY). Income is city-resolution in the pilots, foot traffic is a modeled proxy (road traffic plus anchor gravity), and informal-vendor competitor counts undercount. Those caveats travel with the product rather than hiding in a footnote, which is the whole point: a real-data product that is honest about what its data can and cannot say. Use the provenance panel above each map to see the exact sources per market.