The big three pricing tools largely move your rate with the broad market. That is fine until your two-bedroom beachfront is being dragged around by studios five miles inland. The difference between pricing against a market and pricing against your real competitors is the difference between a blunt instrument and an edge.
What a comp set is actually for
A comp set is not a report you glance at — it is the definition of the choice your guest is making. Every booking is won against a short list of alternatives the traveler was genuinely considering. If your comp set contains those alternatives, your price is competing where the decision happens. If it contains everything within a five-mile radius, your price is competing against noise.
The failure mode: geographic comps
Radius-based comp sets fail in two directions at once. Inferior units drag your reference price down, and you give up rate every night you match them. Superior units pull your reference price up, and you sit vacant wondering why “the market rate” does not book. Both failures look like a pricing problem; both are actually a comparison problem.
What like-kind means in practice
Real substitutes overlap on the dimensions guests actually shop by:
- Capacity and layout — a family of five is not choosing between your four-bedroom and a studio.
- Micro-location — beachfront and two blocks inland are different products, whatever the map radius says.
- Quality tier — photography, reviews, and amenities put listings in different consideration sets.
- Use case — a couples’ getaway, a family week, and a work trip shop differently for the same city.
- Event exposure — units near the venue live a different demand life during event weekends.
Read strategy, not just price
A good comp set is a panel of revealed strategies, not a row of numbers. Calendars show how early each competitor discounts, how hard they price into events, how their weekend premium behaves, and where they hold firm when the market softens. Reading that behavior tells you where demand actually clears — and where a competitor’s panic discounting is about to hand you a booking if you hold.
How to sanity-check a comp set
Three quick tests catch most bad comp sets. First: would a guest cross-shop these listings on the same search page? If not, they are not comps. Second: do their calendars respond to the same demand events yours does? Third: does their realistic price band overlap yours? A comp set that fails these tests will steer any pricing decision built on it — human or algorithmic — in the wrong direction.
Comp sets also age. Markets add supply, listings renovate, hosts change strategy. A comp set built once and never revisited quietly degrades into a geographic one. Treat it as a living input, refreshed as the market moves, and the rest of your pricing inherits its precision.
How Quibble builds the set
Quibble matches comps the way a guest would compare them: it looks within a radius starting at 10 km, takes same-bedroom listings first and backfills, keeps listings within a 75–200% rate band, and scores candidates on bedrooms, rate and distance. About 30 comps are attached to each listing, and for each night the 10 best matches that are still available go into the model. Consumers screen down to a small consideration set of about 10; they do not have the time or attention to weigh hundreds.
Changing a listing’s annual rent target triggers a recomp toward comps that produce similar rent. On the Comp Model tab a comp can be replaced only when it is flagged Needs revision: Quibble offers up to 5 candidates, and the replacement reprices on the next nightly run. Adding a listing by URL is a read-only inquiry that shows how it compares without adding it to the set. Comp Model → Simulator shows share of a night, best comps and what-ifs. Comp rates are all-in, fees included, so compare them with the guest-facing price, not the nightly rate.
Why geography alone is not enough
Quibble’s data put San Francisco’s citywide six-month ADR at $231, useful but too coarse: the Outer Sunset rents about 20% below that average and Pacific Heights about 20% above, in a city 7 miles by 7 miles. Bedroom count is a better filter, since most travelers want a bed each, but it weakens as units get larger, because search results treat a bedroom filter as a minimum. Your property can appear next to larger ones, so the real comp set is often bigger than you think. Amenities matter too: whether a 3-bedroom with a pool competes with the pool-less 3-bedroom down the street can depend on the season.
Tracking competitors without drowning
- Compare on the same metrics: occupancy, ADR and RevPAR, not rate alone.
- Keep history: past pricing, demand patterns and booking behavior make a competitor’s move readable.
- Position on what is genuinely yours (location, design, amenities) and target the segment that values it.
- Revisit the set as the market changes; a set built once decays into a geographic one.
Go deeper: Introducing search-based comp sets, How base-price models work, Forecasting demand before it happens.