# Watch your competitors every week

> **Casafari is the AI agent-native real estate data intelligence platform.** The most complete property index in Europe: a deduplicated, cleaned property graph of residential and commercial property, for sale and for rent, in 16 countries. Every property is one record with its full price and market history. [How the property graph is built](/docs/property-graph).

Set up a weekly competitor watch for your agency: your share in your towns, your competitors there, and which of them take on homes or start cutting prices.

## Ask your assistant

> I run Sample Homes Lisboa. Show my market share in my main towns, and every week tell me which competitors took on more homes than usual or started cutting prices.

Who it is for: Agencies; the CRM vendors and portals that build reports for them.

## Steps

1. [`agencygraph_resolve_agency`](/docs/agencygraph/resolve_agency): Your agency's agc_ id (free).

   ```json
   {
     "country": "PT",
     "q": "Sample Homes Lisboa"
   }
   ```

2. [`agencygraph_territory`](/docs/agencygraph/territory): Your main towns, with your properties, share and rank in each.

   ```json
   {
     "country": "PT",
     "agency_id": "agc_EXAMPLEHOMES00000000000000"
   }
   ```

3. [`agencygraph_market_share`](/docs/agencygraph/market_share): The competitors in each main town: keep their ids as your watchlist, in your own system.

   ```json
   {
     "country": "PT",
     "location_id": 101
   }
   ```

4. [`agencygraph_changes`](/docs/agencygraph/changes): Every week after the weekly build: for the agencies on your list, new, delisted and relisted homes and price increases and cuts, week by week, and any merge.

   ```json
   {
     "country": "PT",
     "agency_ids": [
       "agc_EXAMPLEPRIME00000000000000",
       "agc_EXAMPLEOEIRAS0000000000000"
     ],
     "since": "2026-10-19"
   }
   ```

5. [`property-sourcing_search`](/docs/property-sourcing/search): Optional: the new homes in your towns this week (Property Sourcing, sorted by newest), to see what a competitor is taking on.

## What you get

Illustrative example: made-up agencies, ids and figures, not real data.

| Week | Agency | What changed |
|---|---|---|
| 2026-W43 | Example Prime Estates | 8 new homes, 2 price cuts |
| 2026-W43 | Example Casas Oeiras | 11 price cuts, median -6.0% |

## Notes

AgencyGraph keeps no watchlist for you: pass up to 25 agency ids to agencygraph_changes each week, one call per 25. A scheduled agent (Claude, ChatGPT or your own code) can run the call and send the summary.

What it costs: `agencygraph_resolve_agency` free; `agencygraph_territory` 50 tokens per page of up to 25 rows; `agencygraph_market_share` 50 tokens per page of up to 25 rows; `agencygraph_changes` 25 tokens per call (up to 25 agencies); the other products' tools at their own prices ([Pricing](/pricing)).