A legally clean alternative to platform review data: every metric here comes from pages destinations publish about themselves. 43 sites resolved to 33 domains, measured live and reconstructed year‑by‑year from web archives, 2022‑2026.
Borobudur/Prambanan, Besakih, Lake Toba, Komodo and Raja Ampat had no official site in 2022. All five are corroborated by domain registration dates, so this is not an archive-coverage artefact. Montenegro started saturated at 94% and added none.
Fisher exact · p = 0.018 · survives FDRThe single-year audit put this at 47% vs 19% (p=0.087) — same direction, too little power. Pooling five years of captures resolves it.
Fisher exact · p = 0.026 · survives FDRMarginal, but Montenegro is ahead in every single year of the panel (56–69% vs 22–38%). The gap is stable, not a one-year artefact.
Fisher exact · p = 0.063 · not significantThe divide is two-directional, not hierarchical. Indonesia leads on transactional capability and on presence growth; Montenegro leads on discoverability infrastructure from an already-mature base. A single-year cross-section of the same 43 sites found no significant difference on any technical feature — that null was an artefact of measuring one point in time.
Three pre-specified hypotheses, corrected as a family. Everything else run in this project is exploratory and carries no claim.
| Hypothesis | Indonesia | Montenegro | Raw p | Bonferroni | BH (FDR) | Overturned by |
|---|---|---|---|---|---|---|
| H1 · new web presence | 5/16 | 0/17 | 0.018 | fails | survives | 1 coding change → 0.085 |
| H2 · booking / e-ticket | 10/14 | 4/16 | 0.026 | fails | survives | 1 coding change → 0.063 |
| H3 · schema.org | 5/14 | 12/16 | 0.063 | fails | fails | — already n.s. |
No finding survives Bonferroni. The pre-registered rule is Benjamini–Hochberg, appropriate for a small family of related hypotheses, and H1 and H2 survive it. Both are reported here with the Bonferroni result attached rather than without it.
The binding constraint is fragility, not sample size. A single reclassification — one domain's booking signal judged differently, one presence date read differently — pushes either surviving finding back above 0.05. That makes independent second-coder validation mandatory before submission, and it is what now justifies expanding the frame toward ~30 domains per country.
The archive-coverage confound was tested, not assumed. Montenegro is archived more densely than Indonesia (median 46 vs 32 snapshots), so an absent capture could have meant an unarchived site rather than no site. Domain registration dates settle it from outside the archive: all four Indonesian domains the archive dates to 2023–24 register in the same window, and Raja Ampat’s kkprajaampat.id — live, serving HTTP 200, registered 2026-05-16, not yet crawled — is a fifth case the archive alone would have missed.
Share of each country's domains with any archived capture that year.
Montenegro's single decline is kotor.travel — the only domain in either country to lose presence. It is still live; it migrated to a client-rendered single-page app and became invisible to archival crawlers. Destination digital heritage is being lost silently.
Percentage of that year's valid captures carrying each feature. Denominators shift as sites appear and as captures fail — the n is labelled on hover.
HTTPS is shown for completeness but Montenegro's apparent decline is an artefact: the measure is the URL scheme the crawler happened to archive, and a site reachable over both schemes can be archived under either. Do not report it as a finding.
Not every archived capture is a usable measurement. Screening happens before analysis, never after.
Six captures returned exactly 73 characters under the title “One moment, please…”. Without this screen they would have entered the content series as near-zero, and a site that started blocking crawlers would read as a site that lost its content. 18 of 142 captures are interstitials or empty shells, and the share is rising in both countries.
The live audit, 4 September 2026. All 43 sites in the sampling frame, including those with no official website at all.
| ID | Site | Position | Publisher | No-JS delivery | Schema | Booking | Analytics | Langs |
|---|