A comp set gives your observations context. Where you sit on rate, whether your review score is a strength or a liability, whether a soft month was your problem or the market's, all of it is read against the same handful of properties. Choose the wrong ones and the lens distorts everything behind it. Your rate looks defendable when it is not, an occupancy dip looks like weather when it is a competitor, and the room nights you are losing go to hotels you never thought to watch.
What a comp set is actually for.
The industry reference point is STR, now part of CoStar, whose benchmarking methodology defines a competitive set as the group of hotels a property competes against for business, typically in the same area and offering similar services, according to the CoStar STR Benchmark glossary. For hotels that report data, CoStar's Competitive Set Guidelines require a minimum of four participating properties besides your own, at least three of them not sharing your parent company, operator or owner. The rules exist to keep the comparison honest and to stop any single hotel's data being exposed.
Many independent hotels never see a STAR report, and in small markets some neighboring properties do not report at all. That does not remove the need for a set. It changes what you track. Everything a guest can see, you can see too: the rates a competitor publishes on its own site and on the OTAs, where it sits in review rankings, how visible it is in the searches your guest actually runs. A comp set built on marketing-visible data gives an owner a fixed reference, so that when your numbers move you know whether the market moved with you. Without one, a good month has no denominator.
Three ways a comp set loses value.
The aspirational set. Owners often pick the properties they would like to be compared with rather than the ones guests compare them with. The result reads well and means nothing. If your guest is not shortlisting you alongside those hotels, the rate gap you see is ambition, not evidence, and it tells you nothing about the market context missing from a comparison that does not reflect genuine alternatives.
The geography-only set. Nearness is where a set starts, not where it ends. A honeymoon boutique and a highway stopover can sit half a mile apart and share a postcode but not a single guest. Benchmark against the nearest five properties regardless of who stays there and you will be pricing against hotels that never held your guest in the first place.
The set built to flatter. The most expensive mistake is choosing weaker properties so the weekly numbers feel good. You hold first place on rate and review score against a field you selected to lose, while the bookings you never saw go to competitors who are not on the list. A comparison designed to flatter leaves important market context out.
A comp set is not the hotels you admire, and not the hotels next door. It is the hotels your guest still has open in another tab while they look at yours.
Four criteria for choosing the five.
- Guest overlap. The strongest signal. Which properties appear in the same searches your guest runs, on the same OTA results pages, in the same "travelers also viewed" lists? Which hotels do your own reviews mention by way of comparison? A hotel your guest never shortlists cannot cost you a booking. A hotel they always shortlist cannot safely be left off the set.
- Rate band. Published rates close enough to yours that a guest reads the two properties as one decision. A hotel publishing at several times your rate is answering a different budget, and a hotel far below yours is answering a different guest. Where you sit on rate against your comp set only means something when the band is real.
- Trip purpose. A dive lodge and a conference hotel can share a beach and share no guests. Properties compete when they serve the same reason to travel: the same wedding, the same fishing season, the same long weekend. Purpose sorts real competitors faster than star ratings do.
- Channel mix. Where a hotel sells is visible from the outside: which OTAs it publishes rates on, how prominent it is there, how hard it pushes direct booking on its own site. A property fighting for the same click on the same channel is competing for your room night that day, whatever its brochure says.
Why five? CoStar's guidelines set the sufficiency floor at four competitors, and a fifth gives the comparison margin if one property goes quiet. Five is also few enough to watch properly. Spread your attention across a dozen hotels and the one competitor whose rate moves actually cost you bookings gets averaged into the noise.
Sanity-check the set twice a year.
A comp set is a standing decision, and standing decisions drift. Twice a year, put the five through the same questions that built the set. Has anyone renovated, rebranded or moved its published rates out of your band? Has a new property opened that your guests are now shortlisting? Do your lost bookings still go where they used to go, as far as reviews and shared search results can show? Has a member of the set closed, changed hands or stopped competing for your guest's trip purpose?
Then change the set with discipline, not mood. Under CoStar's rules, a formal competitive set change must involve a minimum of two properties, a constraint that protects data confidentiality but carries a lesson for every owner: sets are not meant to be tinkered with weekly. Each change resets your trend line, and a comparison you rewrite every month can never tell you whether you are gaining or losing ground. Record the date and the reason every time the set changes, so that next year you can tell a real repositioning from a mood.
How HTB reviews a working comp set.
Local comp set tracking is part of HTB's standing work for client hotels, and after decades of hotel marketing the pattern we see most often is a set chosen once and never questioned. HTB identifies where your property sits on published rate against its set, monitors the set's review scores, review positions and search visibility week by week, reports the movement in plain terms, and recommends when the evidence says the set itself has drifted. All of it comes from what a guest can see: published rates, review standing, visibility. Where the finding points at your own website, a rate that reads as expensive because the site undersells what it buys, HTB makes those content changes directly. What you charge, and who you finally choose to measure yourself against, stays your decision.