How we score 52 weeks.

Factors

For each destination, every ISO week of the year carries five normalised scores (0–100, higher = better):

• Weather weather
  Comfortable temperatures, low rain, pleasant conditions.
• Quiet streets crowds
  Fewer tourists, less queueing.
• Affordability cost
  Lower flights, accommodation, and on-the-ground costs.
• Long days daylight
  More daylight hours for activities.
• Events & culture events
  Festivals, holidays, and cultural moments.

Weights

You tell kairoto what matters. Each slider sets the relative weight of one factor; weights are normalised to sum to 1 and the recommendation for each week is the weighted sum of its factor scores. The math is linear, transparent, and tweakable in the source.

for each week w in 1..52:
  score(w) = Σ  weights[f] · factorScore[f][w]
            f∈factors

normalise weights so Σ weights = 1

Archetypes

Five preset weight profiles to start from. Tap one and the sliders move to match. Archetypes for this PoC:

• Balanced
  An even mix of factors.
• Photographer
  Dramatic light, low crowds, scenic conditions.
• Budget backpacker
  Cheapest time to go that's still pleasant.
• Festival seeker
  When the city is alive.
• Comfort traveler
  Best weather, willing to pay, fine with crowds.

What we're honest about

Climatology gives us 30-year averages, but any given year can vary. We tell you the typical pattern, not a guarantee. For the PoC, factor scores are plausibly modelled per-city — they reflect real seasonality but are not drawn from live APIs yet. Costs are seasonal, not real-time fares.

A few specific limitations worth flagging:

• Departure agnostic: we don't know where you're flying from yet, which matters for cost and daylight comfort.
• Activity agnostic: there's no dedicated factor for surf, snow, scuba, etc. — they're partially folded into the weather/events terms.
• No real-time signal: wildfires, strikes, currency moves, and specific year quirks aren't captured.
• Reasoning text: some blurbs are placeholders during the PoC; they'll be edited or replaced before launch.

Sources

The PoC uses a placeholder climatology layer modelled on typical-year averages, and a hand-curated event calendar. Real integrations on the roadmap include Open-Meteo for climatology, official tourism boards for events, and historical fare baselines for cost.