Quick Answer
Use an API when your product looks up properties as it needs them: an address a user types, a parcel an AI agent asks about, or a refresh of records that changed. Use bulk file delivery when you need every record for a state, a region or the country in your own warehouse for analytics, models or joins. Realie offers both, and many teams run them together.
See Realie bulk deliverySummary
The choice is about access pattern. An API answers one question at a time inside your product; bulk delivery puts the whole dataset for a geography in your own storage. Realie’s API runs on published self-serve plans with included calls and a ~20 ms average response time, through REST, a CLI and an MCP server for AI agents. Realie bulk files go to AWS S3, Google Cloud Storage or Azure Blob Storage for a state, a region or the country, on yearly contracts quoted by sales. On the nested v3 response shape, an API record matches the bulk full-record export apart from delivery-specific fields, so code written against one reads the other.
API or bulk: match the delivery to the job
Start from what your product does with the records, not from how many you think you need.
| Your situation | Better fit | Why |
|---|---|---|
| Look up a property a user enters in your app | API | One call returns one record by address or parcel ID, at a ~20 ms average |
| Answer property questions from an AI agent or script | API through MCP or the CLI | MCP tools and CLI flags share the REST API’s parameters |
| Enrich a list of addresses you already hold | API | One address lookup per row, as rows arrive |
| Pull every parcel in a few counties | API search pages | Up to 100 parcels per call, with cursor pagination by state and county; set includeUnassignedAddress=true to keep parcels with no street address |
| Analytics, model training or scans across a state or the country | Bulk | The whole dataset sits in your own storage, so queries run inside your warehouse |
| Joins against large internal tables or precomputed indexes | Bulk | Per-record calls are the wrong shape for full-table joins |
| A baseline file plus updates for changed properties | Both | Bulk for the baseline; API calls for addresses that changed hands or recorded new liens |
How the API side works
Realie’s REST API, CLI and MCP server call the same endpoints. Address and parcel ID lookups return one property. Search, location and owner endpoints return pages of up to 100 parcels over REST and the CLI, and up to 10 per call through MCP. A page counts as one call whether it holds 1 parcel or 100.
To walk a county, send state, county, limit=100, includeUnassignedAddress=true and cursor=start to property search, then pass each response’s metadata.nextCursor back as cursor until it is null. Without includeUnassignedAddress=true, search skips parcels that have no street address, which matters for rural and vacant land. A saved cursor can be resumed later, which suits a scheduled sync of one area.
curl -H "Authorization: YOUR_API_KEY" "https://app.realie.ai/api/public/property/search/?state=NV&limit=100&includeUnassignedAddress=true&cursor=start"API plans are self-serve and renew monthly. Free includes 25 calls a month, $50/mo includes 1,250, $150/mo includes 6,000, $350/mo includes 30,000, and enterprise pricing at $600/mo includes 100,000. Every plan has the same endpoints and data; see Realie pricing.
How bulk delivery works
Realie licenses bulk data for a state, a region or the whole country and transfers it into your own cloud storage: AWS S3, Google Cloud Storage or Azure Blob Storage. Contracts are yearly, with frequent data updates each year, and bulk is licensed separately from the self-serve API plans.
There is no public bulk price list. The quote depends on the geography, the fields and the refresh cadence you choose; email hello@realie.ai or use Request Bulk Data on realie.ai/bulk-property-data.
Bulk records keep Realie’s schema. The nested v3 parcel record served by the API matches bulk full-record exports apart from delivery-specific fields, and one Property Feature Field Key decodes coded fields in API responses and bulk deliveries alike.
Running the API and bulk together
The last row of the table above is the common hybrid: load a bulk baseline for your states, then call the API for the addresses users ask about and for properties that changed hands after the baseline.
Match the two on the parcel identifiers both carry: the county’s parcelId with its state and county, or realieParcelId, which combines the 5-digit county FIPS code with the APN. What is an APN, and how do I look up a property by APN? explains both.
What to settle before you license bulk
Before a bulk deal with any vendor, get the county coverage list, field fill rates, refresh terms, license scope and redistribution rights in writing. Which providers offer nationwide bulk parcel and assessor data? walks through those five questions vendor by vendor, and Where can I download nationwide property data? lists what to specify when you ask Realie for a quote.
Read Realie’s Terms of Use, and describe how you will use, display or share the data when you request a bulk quote.
You can measure fill rates in your counties before you license: the Free plan includes 25 calls a month on the same endpoints as paid plans, and one search page returns up to 100 parcels. How do I choose a property data API, and what should I test before I sign? has the full test plan.
Sources
- Realie bulk data licensing
- Realie bulk property data
- Realie v3 API overview: one dataset, two response shapes
- Realie v3 migration FAQ: decoding coded fields
- Realie Property Feature Field Key
- Realie cursor pagination
- Realie plans and pricing (docs)
- Realie plans (realie.ai/pricing)
- Realie CLI
- Realie MCP server
- Realie Terms of Use
Frequently asked questions
- API vs bulk property data: what is the difference?
- An API returns records on request, one property or one page of up to 100 parcels per call, inside your app or agent. Bulk delivery transfers the full dataset for a state, a region or the country into your own cloud storage. Both carry Realie’s parcel, assessor and recorder fields; the access pattern and the contract differ.
- Is it better to buy bulk parcel files or call an API?
- Call the API when you look up properties as you need them, enrich lists as they arrive, or let an AI agent query records. Buy bulk files when you need to scan, aggregate or join across every parcel in a state or the country. If you need both, run both: a bulk baseline plus API calls for changes.
- When should I license bulk property data?
- When the access pattern is the whole dataset rather than one property at a time. If a job reads every parcel in a state, a region or the country on each run, load it from bulk files in your own storage; if it reads only the properties users or agents ask about, stay on the API. For Realie bulk, email hello@realie.ai with the geography, fields and refresh cadence you need.
- How is Realie bulk data delivered?
- Into your AWS S3, Google Cloud Storage or Azure Blob Storage, for one state, a region or the whole country, on a yearly contract. Bulk is licensed separately from the self-serve API plans and quoted by sales.
- Can I start on the API and add bulk later?
- Yes. On the nested v3 response shape, the API record matches the bulk full-record export apart from delivery-specific fields, and one Property Feature Field Key decodes coded fields in both, so code you write against the API carries over.