Hotels.com Data Scraping

Measure the overlap before commissioning both

Expedia Group runs several consumer brands over largely shared supply. Our Expedia page makes the case that brand is a dimension; the practical question here is whether collecting this brand alongside another in the group buys anything.

• Property overlap is high on most markets.
• Rates are frequently but not always identical for the same property and plan.
• Promotions and loyalty mechanics differ at brand level.
• So the value is in the difference, not in a second full panel.

We report overlap_with_group_brand in the pilot, measured on your own markets. Where overlap is very high we recommend collecting divergence rather than the whole catalogue — the same position our Postmates page takes on shared infrastructure.

Reward nights are a different loyalty shape, and we do not convert them

Most travel loyalty accrues in points proportional to spend. This brand's mechanic has historically worked on stay count — nights earned toward a free night, with the free night's value tied to an average of prior stays rather than to a points balance.

That changes the economics in a specific way:

• A traveller taking many cheap stays earns faster relative to spend than one taking few expensive ones.
• The reward's value depends on the stay pattern, not on a fixed points rate.
• So there is no conversion rate that holds across travellers.

We capture reward_night_progress and any displayed terms as published, with reward_cash_equivalent a constant not_computed.

That is stricter than the points position on our hotel chain page, and for a stronger reason: with points, a conversion rests on a member-specific valuation. Here it rests on a member's entire stay history, which is further from observable still.

Where it is publicly visible

Programme terms and progress indicators are captured where shown to an anonymous visitor. Where they require a signed-in session, the field is null with a reason and gated_share is reported. No accounts created.

Standard OTA mechanics, applied

Everything our Agoda and Expedia pages set out applies here unchanged:

• The record is a rate plan, not a hotel. Refundable and non-refundable are different products.
• Both dates travel with every record — stay date and observation date, with lead time derived.
• Member and app-gated rates null with a reason, with the share reported. The public rate is never substituted.
• Struck-through reference prices captured as displayed and never treated as prior prices.
• Currency depends on the market observed from, with FX stamped at the observation.

What we do not collect

Loyalty account data, reward balances, occupancy, bookings, guest or reviewer identity. Review counts and ratings only.