Hotel rate shopping: competitor rates per market, date and length of stay

Property and channel

Your own ID for the hotel plus the channel shopped: brand site, an OTA or a metasearch result.

Point of sale

The exit country (and city where it matters) of the request. Two shops of the same date from two markets are two observations, not a duplicate.

Check-in date and length of stay

The key of the matrix. Store LOS explicitly; a three-night total divided by three is not the one-night rate.

Occupancy

Adults and child ages. Default searches are usually two adults, and single-occupancy rates differ.

Room type, raw and mapped

Keep the channel's room name verbatim and your mapped category beside it, so the mapping can be audited later.

Rate plan conditions

Refundable or non-refundable, breakfast included, pay now or pay at hotel. The cheapest rate is often the most restricted one.

Price components

Base price, taxes, mandatory fees and currency as shown, plus a computed comparable total for the whole stay.

Availability status

Available, sold out, closed by restriction, or not retrieved. Four values, never one null.

Shop timestamp

Rates move during the day; a shop without a timestamp cannot be compared with yesterday's.

1. 01

   Define the shop matrix

   List the comp set, the stay-date horizon, the lengths of stay you sell (usually 1, 2, 3 and 7), the default occupancy, the channels and the markets. Multiply it out; that number drives cadence and cost.

2. 02

   One sticky session per shop

   Hold one exit for search, property page and room list, so the flow completes on one IP and in one market. Ten minutes is plenty; a new session handle per shop spreads the load across the pool.

   One shop, one exit, one market18 linescopy

   import secrets, requestsBASE = "USERNAME_pool-arcfast_country-de".split("_country-")[0]   # USERNAME_pool-arcfastdef proxy_for(country: str, city: str | None = None) -> dict:    user = f"{BASE}_country-{country}"    if city:        user += f"_city-{city}"    user += f"_session-{secrets.token_hex(4)}_sesstime-10"    url = f"http://{user}:PASSWORD@proxy.arcproxies.io:8000"    return {"http": url, "https": url}def shop(property_url: str, checkin: str, nights: int, country: str):    with requests.Session() as s:        s.proxies = proxy_for(country)        r = s.get(property_url, params={"checkin": checkin, "nights": nights, "adults": 2}, timeout=30)        r.raise_for_status()        return r.text   # parse rooms, rate plans and price components downstream

3. 03

   Parse into explicit statuses

   Every cell of the matrix ends as available (with a price), sold out, restricted, or not retrieved. Retries only apply to the last one, and only with a fresh session.

4. 04

   Normalise to a comparable total

   Convert to one currency at a stated rate, add mandatory fees, decide whether your comparison is tax-inclusive, and divide by nights only after summing. Keep the raw values beside the normalised one.

5. 05

   Feed the RMS, not a spreadsheet

   Push the cleaned rows to your revenue system or warehouse, and alert on moves that matter: a competitor dropping below your rate for a compressed date, or opening a restriction you had not seen.

Prices are list prices in EUR, excl. VAT. Volume discounts apply automatically at checkout — full ladder on the pricing page.