Realie delivers nationwide and state-level property datasets in bulk to AWS, Google Cloud, or Azure. Contact hello@realie.ai with the geography, attributes, and refresh cadence you need and we will scope delivery; bulk is licensed separately from the self-service API plans.
Bulk delivery makes sense when you need the whole dataset resident in your own warehouse — for analytics over millions of parcels, for model training, or when per-request API access would be the wrong shape for the workload. Realie collects and hosts its own data rather than reselling someone else’s, so bulk pricing reflects the geography and attributes you actually want instead of a fixed national package. You can take a single state, a set of counties, or full national coverage, and set a refresh cadence that matches how often the underlying county records change.
Choose bulk when you need to scan or aggregate across the whole dataset — market-wide analytics, model training, or joins against large internal tables. Those workloads are awkward and expensive to express as per-record API calls.
Choose the API when access is sparse and interactive: looking up the properties a user asks about, enriching records as they arrive, or checking a parcel inside a live request. Plenty of teams run both — bulk for the warehouse, API for real-time lookups against the same schema.
Geography, first: national, a list of states, or specific counties. Then attributes — the full record or a subset such as parcel geometry and ownership only. Then refresh cadence, which should track how often the underlying counties actually publish rather than how often you would like new files.
Also settle the delivery target (AWS, GCP, or Azure) and the licence terms covering internal use, redistribution, and retention. Those terms shape the price as much as the row count does.