Skip to content
PhotoWeather

Rule Previewer

You spent 20 minutes building the perfect fog rule. Fog probability above 80%, humidity above 90%, wind below 5 km/h. Looks good. You save it, enable it at your locations, and wait.

Three days later: nothing. Maybe fog conditions are rare? Or maybe your thresholds are too strict and you’re missing every opportunity?

Five days later: still nothing. You lower the fog probability to 70%. Wait another three days.

Finally, an alert. You drive out. Conditions are marginal at best. Too much wind, fog’s patchy. Was your rule too loose? You edit again. Wait more days.

The Rule Previewer ends this cycle. Test your rule against 30 days of historical weather data in seconds. See exactly when it would have triggered, adjust thresholds, re-test instantly. No more waiting days to find out if your rule works.

How It Works

Section titled “How It Works”

The Rule Previewer runs your rule against archived weather forecasts to show when it would have matched. You get a timeline of all matching periods, how long each lasted, and how many total hits you’d have gotten.

This tells you immediately if your rule is:

  • Too strict (zero matches in 30 days)
  • Too loose (matches every day, multiple times)
  • Just right (matches when conditions are actually worth photographing)

Using the Rule Previewer

Section titled “Using the Rule Previewer”
  1. Build your rule in the rule builder
  2. Click “Test Rule” and pick one of your saved locations
  3. Choose your lookback period (up to 30 days)
  4. Review the timeline of matches
  5. Adjust thresholds if needed and re-test instantly
  6. Save when you’re confident the rule is dialed in

No waiting. No wasted alerts. No missed opportunities because your thresholds were off by 5%.

What You See

Section titled “What You See”

Each matched period shows:

  • Start and end time - Exact timestamps when conditions matched
  • Duration - How long conditions persisted (minutes or hours)
  • Date - Which day the match occurred

Plus summary stats:

  • Total match count - How many separate periods matched your rule
  • Evaluation time - How fast the query ran (usually under 1 second)

You can immediately assess rule quality. If you see 47 matches in 30 days, your rule is probably too loose. If you see zero matches in 30 days, your thresholds are too strict (or the conditions you’re after are genuinely rare).

Refining Your Rules

Section titled “Refining Your Rules”

The instant feedback loop is the killer feature here. Old workflow: build rule, wait days, adjust, wait more days. New workflow: build rule, test, adjust, test, adjust, test, save.

Example: Dialing in a Golden Hour Rule

First test: golden_hour_potential > 90 Result: 2 matches in 30 days

Too strict. Only catching absolute peak conditions.

Second test: golden_hour_potential > 80 Result: 8 matches in 30 days

Better, but still conservative. Let’s see…

Third test: golden_hour_potential > 70 Result: 23 matches in 30 days

Now we’re talking. Roughly every other evening, which matches your local climate. Duration averages 45 minutes per match. Perfect for your workflow.

Total time spent refining: 90 seconds.

The alternative? Three weeks of waiting for enough triggers to figure out if your rule works.

Rate Limits

Section titled “Rate Limits”

To keep the system responsive:

  • 15 tests per minute per user
  • Limit applies across all test queries
  • Tests don’t count against your location or rule limits

Hit the limit? Wait one minute. You’re probably iterating too fast anyway—review the results between tests.

Availability

Section titled “Availability”

Available to all users

Free, Plus, and Pro subscribers all get preflight testing. No restrictions. Everyone benefits from being able to validate rules before activating them.

Limitations

Section titled “Limitations”
  • 30-day lookback - Historical data goes back one month
  • Existing locations only - Test against locations you’ve already added
  • Forecast data, not observations - Tests use archived forecasts, not actual recorded weather

The forecast data limitation means test results show what your rule would have predicted, not necessarily what actually happened. But forecasts are what you’ll be working with for future alerts anyway, so this is the relevant accuracy measure.

Next Steps

Section titled “Next Steps”