Houston Restaurant Health Inspections

Get Houston (TX) restaurant and food-establishment health inspection records as clean, structured JSON — establishment name and address, inspection date and type, numeric score, PASS/FAIL result, facility risk profile, and (in violations mode) the full itemized list of violations written up by the inspector, straight from the City of Houston's official open-data portal (data.houstontx.gov).

No HTML scraping, no proxies, no brittle selectors: this actor reads the city's documented CKAN datastore API, pages through it politely, and normalizes every record into one consistent schema.

What data you get

Two official Houston Health and Human Services (Consumer Health Division) datasets:

Mode	Coverage	Volume	What's in a record
inspections (default)	FY2011 – FY2015 (Jul 2010 – Mar 2015)	~201,000 inspections	Facility name/type/address/ZIP, inspection date & type, score, PASS/FAIL, risk profile, permits, follow-up date, citation number
violations	FY2012 – FY2014	~84,000 violation rows, grouped into one record per inspection	Everything above plus an itemized violations[] array: city ordinance code, demerit weight, the inspector's written comments, correct-by date

Freshness disclosure (read this): Houston's machine-readable inspection publication covers fiscal years 2011–2015 — this is the complete official archive the city exposes via API, and the city has not published newer machine-readable inspection data (its old consumer lookup site now redirects to the software vendor's homepage). That makes this dataset the only structured, bulk-accessible record of Houston restaurant inspections in existence. It is ideal for historical analysis, research, and enrichment — not for checking last week's inspection. Coverage was re-verified against the live API on 2026-07-22 (~201,255 inspection records reachable).

Example output record

{
    "source": "data.houstontx.gov",
    "dataset": "food-service-facility-inspections",
    "fiscalYear": "FY2011",
    "inspectionId": "1269554",
    "accountNumber": "413443",
    "establishmentName": "TAQUERIA JALISCO MEX - Texas",
    "establishmentType": "Mobile - Conventional, Unrestricted, Motorized",
    "cuisine": null,
    "address": {
        "street": "2322 1/2 WIRT RD",
        "suite": null,
        "city": "HOUSTON",
        "state": "TX",
        "zip": "77055"
    },
    "inspectionDate": "2010-12-04",
    "inspectionType": "Routine Inspection (001)",
    "score": 3,
    "result": "PASS",
    "followUpDate": "2011-03-05",
    "citationNumber": null,
    "riskProfile": "HIGH",
    "riskScore": 6,
    "numberOfEmployees": 2,
    "permits": "H",
    "facilityStatus": "Active",
    "violations": []
}

In violations mode, violations[] is populated, e.g.:

"violations": [
    {
        "code": "20-21.21(a)",
        "weight": 3,
        "description": "Effective measures intended to minimize the presence of cockroaches on the premises not utilized. (correct immediately)",
        "correctByDate": "2011-10-18"
    }
]

Input

Field	Type	Default	Description
mode	select	inspections	inspections or violations (itemized violations, FY2012–FY2014)
establishmentName	string	—	Case-insensitive substring match on facility name ("taqueria", "whataburger")
zipCode	string	—	Exact 5-digit ZIP filter ("77036")
dateFrom / dateTo	string	—	Inspection date range, YYYY-MM-DD
maxRecords	integer	500	Cap on records pushed (each record = one billable event)

Use cases

• Due diligence & risk scoring — pull a location's full inspection and violation history before a lease, acquisition, or franchise deal in the Houston market.
• Food-safety journalism & research — 200k+ inspections with inspector-written violation narratives (roach findings, condemned food, temperature abuse) is rich raw material for investigative and academic work on enforcement patterns by ZIP and risk tier.
• Local consumer apps & enrichment — join by name/address/ZIP to add a health-history signal to restaurant directories, POI databases, or real-estate datasets.
• ML training data — labeled (score, PASS/FAIL) records paired with free-text violation descriptions.

Pricing

Pay-per-event — you pay only for what you actually get, no subscription:

• $0.005 per actor start
• $0.0008 per record ($0.80 per 1,000 records)

Priced against the comparable city actors on Apify Store: the NYC restaurant-inspection actors (opaldata/nyc-restaurant-inspections, copious_atoll/nyc-restaurant-inspections) charge $1.00 per 1,000 records on the same pay-per-result model. Houston is priced 20% below those comps because its official archive is historical (FY2011–FY2015) rather than continuously refreshed. The full ~201k-record archive costs about $161 — assembled by hand from the city's five fiscal-year extracts, that's weeks of work.

How it works (politeness & reliability)

• Uses the city's documented JSON API (/api/3/action/datastore_search), never HTML.
• Descriptive User-Agent, sequential paging with delays, exponential backoff on 429/5xx.
• Server-side full-text narrowing plus authoritative client-side filtering, so ZIP+4 quirks in the source data never drop matching records.
• Fiscal-year resources outside your requested date range are skipped entirely.