Our Methodology
SafariPlnr is built on verified data from authoritative sources. This page explains exactly how we collect, process, and present information so you can trust what you read.
Wildlife Data
Species-park associations come from three primary sources:
- GBIF (Global Biodiversity Information Facility) - Occurrence records with geographic coordinates matched to park boundaries
- eBird (Cornell Lab of Ornithology) - Bird species checklists from over 6,300 species with 400,000+ park-level associations
- iNaturalist - Community observations used for lodge-level wildlife enrichment (species within 20km radius)
Conservation status comes directly from the IUCN Red List API, covering 94% of species in our database.
Weather & Climate
Monthly climate data uses Open-Meteo's historical weather archive, averaging 5 years of data for each park location. This gives reliable monthly averages for temperature, rainfall, and humidity. 7-day forecasts are fetched on demand from open weather APIs with a 6-hour cache.
PlnrScore Algorithm
Every lodge receives a PlnrScore from 0 to 100, calculated from:
- Google Places rating and review count (weighted by recency and volume)
- Expert Africa rating percentage and review count
- Nearby wildlife density (mammal and bird species counts within 20km)
- Elevation and climate suitability
The score is algorithmic and cannot be purchased or influenced by lodge operators.
Lodge Data
Lodge information is aggregated from multiple public sources including Google Places, Expert Africa, and Go2Africa. We cross-reference data points (name, coordinates, ratings) to build comprehensive profiles. Lodges flagged as duplicates or non-operational are hidden but not deleted.
Entry Fees
Park entry fees are sourced directly from official government systems where available. Currently: Kenya Wildlife Service (KWS) via the eCitizen API (27 parks with fees by visitor category). We are expanding to Tanzania (TANAPA), Zimbabwe (ZimParks), Uganda (UWA), and Rwanda (RDB).
Content Generation
Some content panels (FAQs, visitor tips, safari calendar, ecosystem descriptions) are AI-generated using Google Gemini, following a consistent editorial voice. All generated content is based on real data from our database - not hallucinated. Content is reviewed and can be regenerated with updated voice guidelines.
Update Frequency
Our enrichment pipeline runs nightly, processing data in batches across 27 scheduled tasks. This includes lodge descriptions, wildlife data, weather updates, ratings, and content panels. The full pipeline is managed via our admin interface with job tracking.