Stream.Estate vs Casafari

Casafari has no pricing page. casafari.com/pricing redirects to the homepage, /en/pricing returns 404, and no pricing URL appears in their sitemap. Their FAQ is explicit about the route in: "To buy a subscription, you should click on Request a demo, where you will find a contact form… recognised by our sales department". The API page says pricing "depends on the type of API, data volume, and integration needs".

The commercial terms they do publish matter more than the missing number. Their FAQ states a "Permanence period: 12 months (automatically renewed)", with payment monthly or annually by credit card or SEPA direct debit. If you are evaluating a data source for a product that may not survive a year, an auto-renewing annual commitment is a different kind of decision from a monthly plan.

There is a self-service signup form, and a "Start free trial" call to action sitewide — but no page on their site states how long the trial lasts, so we do not either.

Stream.Estate publishes everything: €0.01 per item pay-as-you-go with no minimum commitment, Starter €99/month (15,000 items), Growth €399/month (150,000 items), Scale €799/month (350,000 items), Enterprise custom. New accounts get €5 in credits. You create the key yourself.

One API term is worth reading before you architect around Casafari: their API page states that "API results may not be stored indefinitely for reuse." If your product depends on building your own persistent copy of the data, ask them where that line falls.

Casafari's geographic reach is real and it is the clearest thing it has over a France-first API. Their FAQ enumerates a dedicated database per territory for Andorra, Austria, Belgium, France and French Overseas Territories, Germany, Greece, Italy, Luxembourg, Monaco, Poland, Portugal, San Marino, Spain, Switzerland and the United Arab Emirates. If you compare Lisbon to Madrid to Paris in one product, that matters and Stream.Estate cannot serve it.

Be careful with their headline numbers, though: on the day we checked, their own pages carried "20+ countries" (home and API page), 15 enumerated territories (FAQ), "23 countries" (Data Export page title) and 9 (area guide). Same for database size — 50 million unique properties analysed (home), 60 million properties with history (API page), "more than 200M listings" (FAQ). We name the page rather than pick a number.

The consistent figure across home, FAQ and API page is sources: "more than 30.000 sources that have properties listed, including agencies and marketplaces" — spread across all their markets. Stream.Estate's 1,500+ sources are concentrated on France and Belgium. Neither number tells you French depth, and Casafari publishes no France-specific figures at all; their area guide for France says only that they are "present all around France".

France is genuinely supported: their OpenAPI spec carries FR in both its zip-code boundary country enum and its sale-comparables country enum, alongside French overseas territories.

Casafari's API page describes update cadence as "once per day or even more frequently" depending on the API, and their bulk Data Export product as "daily delta-sync and weekly full-sync". No absolute daily listing count is published.

Stream.Estate is built around change events rather than snapshots: saved searches carry webhooks that fire on property.ad.create, ad.update.price, ad.update.surface, ad.update.pictures and ad.update.expired, so you learn about a price cut when it happens rather than at the next sync.

On deduplication, both claim it. Casafari claims "100% deduplication accuracy" and "one complete record per property" — an unusually absolute number with no published methodology behind it; the word "duplicate" does not appear anywhere in their 406 KB OpenAPI specification. Stream.Estate reconciles listings published on several portals into a single record and publishes no accuracy percentage, which is its own kind of admission.

On history, Casafari is ahead where closing prices matter: their home page states "45 Million registered property sales with closing prices" and "7+ years data history on closing prices", with transaction comparables available for France, Italy, Portugal and Spain per their spec. Their advert history is more conditional — "up to one year of historical records or complete historical data from the start of collection", depending on the API. Stream.Estate tracks per-advert change events rather than notarised closing prices; in France, closing prices come from the public DVF dataset.

Three honest cases.

You operate in more than one country. Fifteen enumerated territories against France and Belgium is not a framing problem, it is a fact. If your portfolio spans Iberia and France, Casafari serves it and Stream.Estate does not.

You want a suite, not a feed. Casafari ships a CRM (with a free-forever plan covering client, property and lead management plus website templates), market analytics with time series, valuations, location intelligence from its 2023 Targomo acquisition, commission splitting between agencies, and a lead-generation widget. Stream.Estate has no dashboard at all.

Your buyer is an agency, not an engineering team. Casafari's whole shape — demo, sales conversation, annual contract, dashboards — fits an agency network. Stream.Estate's shape fits a team that wants a key this afternoon and will build the interface themselves.

And the flip side: if you need French and Belgian listings behind a self-provisioned API, with a published per-item price and no annual commitment, that is what Stream.Estate is built to be.