Skip to content

The Accuracy Index

A sighting rate you can actually trust.

Anyone can claim a perfect record after one lucky trip. The Accuracy Index won’t let them. Here’s exactly how the numbers are built — and why being honest about what we don’t know is the whole product.

Two-sided proof

Every sighting on Lyferr is backed by evidence from both sides — the guide’s geotagged, timestamped photo and the traveler’s own confirmation. Neither side can fake it alone.

We cross-check every claim against eBird, GBIF, and iNaturalist occurrence data, so a species claimed where or when it doesn’t occur gets flagged automatically.

What a rate looks like

Found on 47 of 52 trips · 90% (range 79%96%)

The denominator (52 trips) and the confidence range are always shown. We never publish a bare percentage.

Then we do the honest math.

We rank by the Wilson lower bound of each rate, not the headline number. That keeps small samples humble: a brand-new guide can’t leap to the top on a single lucky outing. Below about ten trips we lead with the raw count and label it early data.

“A guide at 1-for-1 won’t outrank one at 47-for-52.”
Transparency is the product. If we can’t back it up, we won’t show it as fact.

The three tiers

Every rate carries a tier so you always know how much to trust it. The tiers are colour-coded — and never red. We’re honest, not alarmist.

Verified
Every sighting is backed by two-sided proof and cross-checked against occurrence data. The strongest signal we publish.
Reported
Guide-reported and plausible, but not yet corroborated by two-sided proof. Our launch guides are seeded at this tier.
Unverified
Early or thin data, shown honestly and kept off by default — never mixed in as if it were proven.

Questions about the numbers

Ready to put it to work? Search a species →

No guide guarantees a wild animal. These are verified track records, shown to improve your odds — not to promise an outcome.