Competitive Set Strategy for Hotels: Building the Right Compset for Smarter Pricing

Every pricing decision a hotel makes is, implicitly or explicitly, a competitive decision. Setting your rate $10 above the market average is a statement about your perceived value differential. Setting it $20 below signals aggressiveness or desperation, depending on context. The problem is that most hotels base these competitive positioning decisions on a compset that was assembled somewhat arbitrarily — the five nearest hotels, or the properties a general manager happened to think of, rather than the true set of alternatives a guest would genuinely consider.

A well-constructed competitive set is one of the most valuable assets in a hotel’s revenue management toolkit. It provides the market context that makes pricing intelligence meaningful, the benchmark against which performance is measured, and the signal source that drives the most critical reactive pricing decisions. Building it right — and maintaining it as the market evolves — is foundational work that pays compounding dividends.

What Is a Hotel Competitive Set?

A hotel’s competitive set (compset) is the defined group of properties that compete directly for the same guests in the same market. It is the lens through which competitive pricing intelligence is gathered, interpreted, and acted upon. When a revenue manager checks what competitors are charging for a given date, or when an AI system calculates a market rate index, the compset is the reference point that makes those comparisons meaningful.

The compset is not simply “the other hotels nearby.” A city centre market might contain dozens of hotels within a two-kilometre radius spanning budget hostels, mid-market business hotels, boutique properties, and five-star luxury resorts. A 3.5-star family hotel and a 5-star business hotel in the same building would not be in each other’s compsets — they are competing for entirely different guests with different willingness to pay and different purchase criteria.

What a Compset Is Used For

• Rate benchmarking: Measuring your ADR and rate index relative to comparable properties.
• Competitive positioning: Setting deliberate rate positioning (premium, parity, or discount) relative to the compset average.
• Demand signal detection: Identifying compset availability closures as leading indicators of strong market demand.
• Market share analysis: STR and similar services use your compset to benchmark your RevPAR performance relative to the market.
• Pricing strategy calibration: AI pricing engines like Propeter use compset data as a continuous input to demand and price elasticity models.

How to Select the Right Comp Properties

Compset selection should be a rigorous analytical exercise, not an intuition-based exercise. The core question to answer for every potential comp property is: “When a guest considers staying at my hotel and then decides not to, where do they go instead?” The properties that capture your lost bookings are your true competitors — and those are the properties that should be in your compset.

The Five Core Selection Criteria

1. Geographic proximity. Comp properties should be close enough that a guest would consider them a genuine alternative for the same trip. In a dense urban market, this might mean within 1–2 kilometres. In a resort market, it might mean within the same micro-destination — the same beach strip, the same ski resort access point, the same national park entrance. Properties far enough away that a guest would consider them a materially different experience are not true competitors.

2. Star rating and service level. The most robust guidance is to include properties within one star category of your own. A 4-star hotel’s compset should contain 3.5-star to 4.5-star properties. Guests considering a 4-star property rarely cross-shop 2-star budget options as genuine alternatives — the product is too different. Including budget properties in a mid-market compset creates a rate index that is artificially low.

3. Comparable ADR range. Even within the same star category, significant ADR variation can exist. A boutique 4-star property with a design-forward product might average $200/night in a market where other 4-star properties average $140. Including the $140 properties in the boutique hotel’s compset is misleading unless those properties genuinely compete for the same guests. ADR bands provide a useful filter: limit your compset to properties within 30–40% of your own ADR.

4. Room count and scale. A 20-room boutique and a 500-room convention hotel have fundamentally different demand dynamics. The convention hotel fills and closes to availability at times and demand levels that have no relevance to the boutique’s inventory management. Large-scale properties make poor comps for small properties, and vice versa.

5. Guest segment overlap. The most important criterion, and the hardest to quantify directly. Properties with significant overlap in their target segments — leisure, corporate, group, extended stay — are genuine competitors. Properties primarily serving segments you do not target are not. A corporate-focused business hotel and a leisure-oriented boutique in the same postcode may not be each other’s primary competitors even if they are adjacent.

Primary vs. Secondary Competitive Sets

Most sophisticated revenue managers maintain both a primary and a secondary competitive set. The primary compset — typically 4–6 properties — contains the closest, most directly comparable competitors. These are the properties monitored most intensively, tracked in daily rate shopping, and used as direct pricing benchmarks.

The secondary compset — an additional 4–8 properties — includes properties that are comparable but less directly competitive. Perhaps they are slightly further from your location, in a slightly different price band, or serving a partially overlapping but not identical segment mix. Secondary compset data provides market context without the noise that comes from including too many primary comps. It is particularly useful for detecting broader market trends that are not yet visible in the primary compset.

Common Compset Selection Mistakes

The most common compset mistakes all share a root cause: using criteria that are easy to measure (proximity, brand name, physical size) rather than criteria that reflect actual competitive dynamics (guest segment overlap, booking cross-shopping behaviour).

• Including brand-name hotels purely for prestige: Being in the same market as a major chain brand does not make that brand a relevant comp if it serves a materially different segment.
• Using proximity as the only criterion: Hotels 500 metres apart with completely different product offerings and guest bases are not competitors — the guest who books a $400 luxury suite is not cross-shopping with a $90 budget room down the road.
• Never updating the compset: Markets change. Hotels reposition, renovate, close, or change ownership. New properties open. A compset built three years ago may contain properties that no longer compete with you and miss newer properties that do.
• Making the compset too large: Including 12–15 properties in a compset creates noise and dilutes the signal from the properties that actually matter. The competitive average of 15 dissimilar properties tells you very little about what any individual competitor is doing.
• Not accounting for distribution channel differences: A property that sells primarily through one OTA may price very differently on other channels. Monitoring their rates only on Booking.com when your guests primarily use Expedia creates a distorted picture.

Monitoring Tools: Lighthouse and Beyond

Once a compset is built, it needs to be monitored continuously to generate actionable intelligence. The monitoring infrastructure is as important as the compset selection — a perfectly constructed compset monitored once per day via manual rate checking is less valuable than a good compset monitored every four hours through an automated intelligence platform.

Lighthouse (OTA Insight) is the industry-standard platform for compset monitoring. It aggregates rate data from all major OTAs across your defined compset properties, stores historical rate data, and presents forward rate calendars showing competitive positioning 365 days out. Key Lighthouse metrics for compset monitoring include rate index (your ADR as a percentage of compset average), rate rank (your position in the compset by rate on any given date), and availability signals (when compset properties are showing sold-out or restricted availability).

What Lighthouse Data Tells You

• Real-time and historical rate positioning relative to your defined compset
• Availability status across all compset properties — critical for detecting demand surges
• Rate change frequency and patterns for each competitor
• Parity monitoring across your own channels and compset channels
• Market demand index derived from search and booking data

Propeter integrates directly with Lighthouse, pulling compset intelligence automatically into the AI pricing pipeline. This eliminates the need for manual data extraction from the Lighthouse dashboard and ensures that competitive data is actively used in pricing decisions rather than sitting in a reporting interface waiting to be reviewed.

Beyond Lighthouse, Propeter’s proprietary web scraping layer monitors competitor direct booking sites, metasearch results, and promotional pages that may not surface in standard OTA monitoring. This provides a complete competitive picture that catches direct-only rates, exclusive promotional offers, and rates on booking channels outside the major OTAs.

Using Compset Data in Pricing Decisions

Compset data becomes valuable only when it drives concrete pricing actions. The four most important uses of compset data in pricing decisions are: rate positioning calibration, demand signal detection, competitive response triggers, and market share benchmarking.

Rate Positioning Calibration

Your target rate position relative to the compset should be a deliberate strategic choice, revisited periodically and adjusted based on your product’s competitive standing. A hotel that outscores its compset on TripAdvisor by two full points can justifiably maintain a 15–20% rate premium. A hotel that scores at or below the compset average should position at parity or below until product investment improves the quality differential. Compset data is the measurement tool that tells you whether your actual rate position matches your intended position.

Demand Signal Detection Through Compset Availability

One of the most valuable uses of compset monitoring has nothing to do with rates — it is watching when compset properties sell out or close availability for future dates. A competitor closing to new bookings for a date 30 days out signals that demand for that date is strong enough to fill them before yours. This is a leading indicator that your own rates for that date should be higher, because the pool of available alternatives for arriving guests is shrinking.

Propeter’s Competitive Intelligence Feature

Propeter’s Market Intelligence Agent sits at the second stage of the six-agent AutoGen AI pipeline: Data Ingestion → Market Intelligence → Demand Forecast → Price Elasticity → RevPAR Optimisation → Strategy Agent. Its role is to transform raw competitive data — compset rates, availability, trends, positioning — into structured market intelligence that downstream agents use to make pricing decisions.

The agent ingests compset data from two sources: the Lighthouse integration (structured OTA rate data with full historical depth) and Propeter’s proprietary web scraping layer (direct rates, metasearch pricing, and promotional offers outside OTA channels). Every four hours, it generates a market intelligence report for each date in the 365-day forward horizon, covering your current rate position relative to the compset, availability gaps that signal demand pressure, and detected rate change patterns that may indicate competitor strategic moves.

This structured intelligence is passed to the Demand Forecast Agent, where it calibrates the XGBoost and LSTM demand models. When compset availability is tightening on a future date, the demand forecast for that date is adjusted upward — not purely based on historical patterns, but based on real-time competitive market signals. The Price Elasticity Agent then uses this updated demand forecast to calculate the optimal rate point, and the RevPAR Optimisation Agent makes the final rate recommendation that flows through the 13-stage engine to all distribution channels.