SEO: lead destination H1 with the best time to visit money phrase

[Manuel10i]

Problem

The destination pages are the only compounding traffic source in the portfolio (dozens rank page 1, pos 6 to 11, with real clicks). But the <h1> in all four locales was When is {name} crowded? and its localized equivalents, which targets only the secondary "is X busy" intent.

Search Console (28 day window to Aug 1) shows the dominant query volume driving these pages is the "best time" family:

• "best time to go to X" / "best time to visit X" (e.g. puerto vallarta appears in 8+ query variants, prague, cancun, yellowstone, buenos aires)
• "best time of year to visit X" (ningaloo reef pos 34, barossa valley pos 42, cairns)
• "best month(s) to visit X" (best months buenos aires pos 35)
• German: "beste Reisezeit X" is one of the highest volume DE travel queries, and the DE h1 (Wann ist {name} voll?) missed it entirely

The single strongest on-page element did not contain the primary money phrase, while the title, meta description and FAQ already do.

Change

Rewrite the destination h1 to lead with the primary phrase and keep the crowd angle in a parenthetical, so both intents stay covered and no existing ranking signal is dropped:

Locale	Before	After
en	When is {name} crowded?	Best time to visit {name} (and when it is crowded)
de	Wann ist {name} voll?	Beste Reisezeit fuer {name} (und wann es voll ist)
es	Cuando hay aglomeraciones en {name}?	Mejor epoca para visitar {name} (y cuando hay aglomeraciones)
fr	Quand y a-t-il foule a {name} ?	Meilleure periode pour visiter {name} (et quand il y a foule)

The <title> is deliberately left unchanged (it already reads {name}: best time to visit and when it is crowded) to protect the existing "best time to visit X" ranking. Crowd relevance is retained across the H1 parenthetical, the intro, the "Busiest months" heading and the "Is X busy?" FAQ, so nothing is lost.

Verification

• All four message files parse as valid JSON
• tsc --noEmit exit 0
• eslint exit 0
• next build compiled successfully (the post-compile typecheck step is OOM limited in this build environment, a known constraint for this repo; typecheck was validated separately via tsc above)
• Pure i18n string swap: the render path t("h1", { name: displayName }) is unchanged from the previously working string, only the value changed, {name} placeholder preserved
• No em-dashes; crowd angle kept in parentheses per copy rule

Measure after deploy: do the DE destination pages start picking up "beste Reisezeit [Ort]" impressions, and does destination CTR on the "best time" query family improve?