Data Integrity

What we do today to keep state data right, and the standard we are building toward.

This page used to describe a monthly audit, automatic stale flags, and a versioned changelog as things already running. Some of that was in place and some of it was the plan. Telling them apart matters more than sounding finished, so each section below is split: Today is what the system does right now, and Building toward is work in progress with nothing shipped yet. If you find a number that is wrong, the last section is the fastest way to get it fixed.

1. Authoritative Data Registry

Today.

  • Fee data, deadlines, and draw systems come from official state wildlife agency sources.
  • Every state entry carries a sourceUrl pointing at the agency page it came from.
  • Scraped documents are hashed at fetch time, so we can tell when an agency page has changed.

Building toward.

  • Storing the agency document itself, not only its address, so a value can always be traced back to the page that stated it, on the date it stated it.
  • A page number and an exact quote attached to every individual fee, deadline, and point rule. Today the source is recorded at the level of the state, not the level of the number.

2. Verification Cadence

Today.

  • State data is re-checked against official sources ahead of application season, and after any agency notice we catch.
  • Every state shows when its data was last pulled, with a freshness indicator that turns over as that date ages.
  • Audits are dated and recorded in the state record, so a correction can be traced to the review that found it.

Building toward.

  • A scheduled re-check of every source document on a fixed cadence rather than an as-needed one.
  • A regression suite of known-answer cases that has to pass before any data change goes live, so a fix in one state cannot quietly break another.

3. Fee Reconciliation

Today.

  • Fees are itemized line by line: qualifying license, application fee, point fee, and tag cost.
  • Resident and nonresident fees are resolved from your home state, not averaged.
  • Where a cost is an estimate rather than a published figure, the app says so instead of showing it as fact.

Building toward.

  • A verified source on every fee line. The fields exist and a minority of lines are filled; the rest inherit their state’s source rather than carrying their own.
  • A per-value review step, so a changed fee is approved by a person with a name and a date before it reaches your plan.

4. Deadline Integrity

Today.

  • Every application deadline carries an IANA timezone designator, for example America/Denver.
  • Deadline warnings surface ahead of close with urgency-tiered badges.
  • Draw result dates are tracked per species per state.
  • Where an agency publishes no date, we show no date rather than an invented one.

5. What We Monitor

Today.

  • Source-document changes, detected by comparing content hashes between fetches.
  • Scrape volume, so a source that comes back short is held for review instead of published.
  • Fee discrepancy and deadline reports from users.
  • Coverage gaps, so a state or species with data we do not trust is withheld rather than estimated.

Building toward.

  • Numbered dataset versions, so a plan records which version of the data produced it and can be recomputed against a later one.
  • A published changelog of what changed between versions and why.

6. How to Report an Error

  • If you find incorrect data, email support@tag-planner.com.
  • One wrong fee is a big deal to us. We treat every report as a priority.
  • Tell us the state, the species, and the number you expected. That is usually enough for us to find it the same day.