AI assistants and search decision
One Place
A useful subset (search + dedupe + visual similarity for a handful of sources) is realistic for a small team to build and run, but reproducing the live, pan‑European index, massive image corpus, and continual ingestion across 660+ portals requires large-scale data collection and infrastructure that are durable advantages for the vendor.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- User describes desired property → system searches indexed listings and images → deduplicate/merge matches → rank & return results with explanations → user saves/boards favorites and receives alerts.
What it still won’t have
- Pan-European live index at the scale claimed (millions of listings, hundreds of millions of images)
- Coverage and continual ingestion from 660+ portals and 20 countries
- Proprietary price-history and continuous tracking across many markets
- Agentic features built on their full-market dataset (research agent that reads whole market)
What remains hard
- Infrastructure at scale
LIVE 5.0 M Active listings
- Infrastructure at scale
LIVE 292 M Property images indexed
- Infrastructure at scale
LIVE 660 + Source portals, unified
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 11 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal One Place replacement using: Node.js + Express backend, PostgreSQL for canonical listings, Elasticsearch/OpenSearch for text search, Milvus (or Pinecone) for image vectors, a Python worker to run CLIP embeddings and a simple LLM-based intent parser (OpenAI or local LLM). Scope: ingest CSV/HTTP feeds from 5 sample portals and normalize; dedupe/merge into canonical records; index text and vectors; implement NL-to-query parsing; provide a web UI to run free-text searches, show merged listing pages with images, save listings to boards, and one saved-search alert email. Out of scope: crawling 660+ portals, full production-scale ingestion pipeline, multi-country currency normalization, and advanced agentic explainers. Include error handling, basic tests for ingestion/dedup/indexing, and deployment scripts (Docker + Terraform) for a small cloud instance.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 4 cited sources+3
- Price verified on pricing page+3
- Hard moats found in the evidence-3
- Evidence score60
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productOne Place — product
- official pricingPricing | One Place
- official docsFeatures | One Place
- open sourceorangecoding/fredy
Integrity checks
What held up, and what did not.
