Comp Sets: Benchmark by Positioning Tier, Not Postcode
A competitive set (comp set) is the group of listings your rates are benchmarked against. Every pricing tool builds one by default — usually the nearest 5–10 properties. A positioning-tier comp set replaces proximity with what actually drives guest choice: product tier, guest segment and amenity bundle.
When the comp set is wrong, every number downstream of it is wrong. Your "competitive" rate anchors to properties your guests never actually compare you against — every week, every season, compounding quietly.
The proximity trap
Distance is easy to compute, so distance is the default. But a heritage 3BR terrace in Paddington competes with heritage terraces across Sydney's inner-east — not with the budget apartment two streets over. A harbour-view 2BR competes with harbour-view product, regardless of postcode. Guests shop by product tier and trip purpose; only algorithms shop by radius.
The result of a proximity set is systematic mispricing in both directions: premium units anchored down toward cheaper neighbours (leaving rate on the table), and standard units anchored up toward product they can't match (losing conversion and falling behind pace).
In a single building where we manage 24+ units across eight floors, rebuilding comp sets per floor, view and finish tier produced RevPAR Index results from 41–67 on ground-floor 2BDs to 133–172 on 7th-floor 2BDs — inside the same building. A ZIP-code comp set averages that spread away; a positioning-tier set prices each unit to its actual market. The same rebuild in Sydney lifted ADR Index ~12 points in 60 days.
The four questions every comp set must pass
- Positioning: does every comp match my tier — build quality, design level, brand feel — not just my bedroom count?
- Segment: are these listings selling to my guest (leisure vs business vs event vs family group)?
- Amenity bundle: do the elasticity-driving amenities match — pool, view, parking, workspace?
- Freshness: was this set reviewed in the last 90 days, or is it drifting?
How to rebuild yours in an afternoon
- Pick your 10 highest-revenue units. The lift concentrates where the revenue does.
- Write each unit's positioning sentence. "Heritage 3BR terrace for families and small groups, premium finish, walkable dining." If you can't write it, guests can't perceive it — that's a different problem.
- Find 5–8 listings a guest would genuinely cross-shop. Search like a guest: same trip, same filters. Distance is allowed to be a tiebreaker, never the criterion.
- Replace the tool's auto-set (PriceLabs, Wheelhouse and Beyond all allow manual comp sets), and diarize a quarterly review.
Then let the new reference reprice the unit. If the set changed materially, expect base-price moves of 10%+ in one direction or the other — that delta is the mispricing you've been carrying.
Comp sets feed everything else
Positioning-tier sets are upstream of nearly every other play: pace comparisons only mean something against true comparables, elasticity decisions depend on knowing your genuine reference price, and RevPAR Index is literally computed against the set. Fix the reference, and every downstream number starts telling the truth.
RevBnB builds living comp sets
The agent clusters comparables by positioning tier, guest segment and amenity signature — and re-evaluates them continuously instead of quarterly. Your rates anchor to the market you're actually in.
Frequently asked questions
How many properties should a comp set contain?
Five to eight genuinely comparable listings beats twenty approximate ones. Consistency matters more than count — track the same set over time so pace and rate comparisons stay meaningful, and change members deliberately, not constantly.
What if there aren't enough true comparables nearby?
Widen geography before widening tier. A heritage terrace across town is a better comp than a mismatched unit next door. In thin markets, borrow reference from the nearest comparable market and adjust for the local demand delta.
Should different units in one building share a comp set?
Only if guests perceive them as the same product. Our Chicago data says they usually don't: floor, view and finish tier created a 130-point RevPAR Index spread inside one building. Segment units by what the guest experiences, not by the address they share.
How often should comp sets be rebuilt?
Review quarterly; rebuild when the market shifts — new supply entering your tier, a renovation changing your positioning, or a seasonal switch in dominant guest segment. A set older than a year is almost certainly mispricing something.