Transparency
What the data layer actually costs
Every IRIS licence, including the entry IRIS Market tier, includes a standardised European data layer: resident demographics (including purchasing power), daytime population, visits, traffic, and points of interest, harmonised to H3 resolution 10 hexes (~120 m) across markets and kept current. IRIS Market scores and ranks sites on that layer; the forecast itself starts at the Validation Sprint. We say that layer would cost real money to assemble yourself. This page is us showing our work, so you can check the claim rather than take it on faith.
How we researched this
We priced each data type the way a retailer actually would. The supplier had to be commercial, and had to cover ten European markets (Netherlands, Germany, France, Spain, Italy, Portugal, Poland, Belgium, Austria, Sweden). It also had to deliver the granularity a site decision needs: roughly an H3 resolution-10 hexagon (~120 m), a building, or the finest national postcode.
Six rules keep this honest:
- Full-market coverage only. The data has to cover the whole market, not a sample of points. A network of a few street sensors, or a free government counting-station feed, does not qualify, because it cannot tell you about the street corner you actually care about.
- Rich, not a single number. Footfall has to carry the detail a decision needs: time of day, day of week, visit duration, cross-shopping. A bare pedestrian count does not qualify.
- Micro granularity. Coarse province or municipality data does not count. We need roughly H3 resolution 10 (~120 m hexes).
- Conservative on purpose. Where a supplier quotes a range, we take the low end, and across the ten markets we take the lowest we found. Every figure below is a floor. If anything, sourcing it yourself costs more.
- Confirmed or estimated, labelled. Some suppliers publish a price; most quote case by case. A confirmed price has a source you can open. And card-spend / transaction data is licensed separately, it is not part of the standardised layer, so we do not count it.
- Annual-equivalent, because data goes stale. Some of these are annual licences and some are one-off purchases. But a one-off snapshot ages, and a stale number quietly misleads a site decision, so we count what it costs to keep each layer current, not just to buy it once. That is the only version worth having.
What is free, and what is not
The surprise, and the honest part, is that some of this is genuinely free at a fine grid, if you know where to look and you only need part of the picture:
- Resident data is free at a fine grid in several markets: Germany (Destatis, Zensus 2022, 100 m grid), Belgium (Statbel, 125 m grid), the Netherlands (CBS, 6-digit postcode), Austria (WorldPop, 100 m), Spain and Italy (census-section). Purchasing power is free at census-section level in Spain.
- Boundaries and the grid itself are free everywhere: national open geodata plus the open-source H3 hexagon system.
Then it stops. The data that actually moves a site decision, at full-market coverage and the richness a decision needs, has no free source at all: daytime population, footfall and visits, and verified points of interest. And the free resident layers, where they exist, are scattered across dozens of national portals in different formats and vintages, and are missing for much of Europe. Assembling all of it into one current, uniform H3 resolution-10 layer (~120 m hexes) is the work you are paying for.
The numbers, per data type
Lowest conservative annual figure we found across the ten markets, per data type, at micro granularity and full-market coverage:
| Data type | Lowest we found | Free public option? | If you license it (example) |
|---|---|---|---|
| Resident data (population, age, income, purchasing power, education) | ~€5,000/yr | Yes, at a fine grid in DE, BE, AT, NL; free purchasing power in ES | Michael Bauer Research, GfK/NIQ: 100 m grid demographic + purchasing-power package (quote) |
| Daytime population | ~€5,000/yr | No | Telco mobility (Vodafone, Proximus, Telia), Michael Bauer Research (quote) |
| Footfall & visits | ~€9,000/yr | No | MyTraffic Pro, full-market GPS with time of day, dwell and cross-shopping (published, ~€9,000/yr) |
| Traffic flow | ~€5,000/yr | Sparse government sensors only, not a full network | PTV, HERE, TomTom, INRIX: full road-network flows (quote) |
| Points of interest | ~€2,000/yr | OpenStreetMap, free but weaker categorisation | Geolytica POIData.xyz (published, from ~$2,000/country/yr) |
| Boundaries / grid | €0 | Yes, everywhere (national open geodata + H3) | Not needed |
Add the paid rows and, even at these conservative floors, the data foundation runs to roughly €26,000 per market per year to source yourself, ready to use and covering the whole market. IRIS Market, which includes it, is €6,000. That comparison is the backstop, not the pitch: the real value is that we assemble the fragmented pieces, harmonise them to one grid across Europe, and keep them current, so you do not have to.
The honest caveats
- Because our layer is standardised across markets and built partly on official and modelled sources, on some variables it is less granular or less fresh than a single bespoke premium feed. Where a premium source genuinely beats it and you want it, we scope it and pass it through at cost, with the invoices.
- A few figures (daytime population, traffic-flow network licences) are quote-only estimates: real named suppliers, but no public price. We label them and are firming them with real quotes. Footfall and points-of-interest prices are confirmed from published rate cards.
- No supplier here is a competitor. They are data vendors and open-data sources. We name them so the comparison is checkable, which is the whole point of this page.
See how this feeds the licence on the plans page, or toggle the layers yourself in the interactive data-cost breakdown.