How to Identify Your Anchor ICP (and Why It Matters More Than the Other Four)

Sources: HVS demand-segmentation framework (Bisema, 2009), Cornell Center for Hospitality Research on amenity ROI by guest type (Dev, Hamilton & Rust, 2017), Dolnicar's work on a priori vs. data-driven segmentation in tourism (2004; 2017), Guo, Barnes & Jia's LDA analysis of 266,544 hotel reviews (Tourism Management, 2017), Cornell Hospitality Report on competitive-set pricing across 67,008 hotel observations (Enz, Canina & Lomanno), Kalibri Labs cost-of-acquisition research, HSMAI total-revenue-management segmentation guidance. Last reviewed: 2026-04-21.

Key takeaways

Most hotels carry 3-5 identifiable ICPs but only one Anchor. The Anchor is the persona whose trip triggers, review language, and amenity expectations compound with your property's structural strengths. Pick the wrong Anchor and every downstream decision (hero photo, title copy, rate band, which compression event gets cover billing) pulls slightly off axis. Pick the right one and they all pull together.

Anchor is not "biggest volume segment." A 4-star property two blocks from a teaching hospital may book more leisure stays on a peak-weekend basis, but if medical-affiliated trips (residents, visiting attendings, patient families) are the persona where the property's location, workspace, and 2-night stay pattern structurally outperforms the comp set, medical is the Anchor. Volume fills nights. Anchor defines fit.

Anchor ICPs show three observable signatures: (1) a recurring, verifiable demand generator within roughly 3 miles, (2) review-language dominance: the persona's vocabulary dominates the 4+ star reviews of your direct competitors; (3) amenity and structural fit: what the property already has matches what the persona needs, not the other way around.

Why the Anchor matters more than the other ICPs

The HVS demand-segmentation framework, still the working reference in hotel feasibility work, groups most hotel demand into three buckets: commercial, meeting-and-group, and leisure.¹ Every hotel, even a 40-room boutique, spreads across all three. But performance is never evenly distributed. In the original Bisema framework, a hotel's viability depends on matching property configuration to the dominant demand base in its immediate catchment: commercial demand concentrates Monday-Thursday, group peaks spring and fall, and leisure fills Friday-Saturday.¹ In a downturn it is group demand that lags the decline, because the funds were committed before the market turned.¹ A hotel that tries to be equally good for all three usually ends up being structurally best for none.

The Anchor is the formal name for that dominant demand base seen from the inside of a single property. Four other ICP types sit around it. Yield is the highest-rate persona, used for ADR compression tactics during specific windows, not volume. Gap is the persona a competitor set systematically underserves; it's the strategic play, the way you win share you don't have yet. Filler keeps lights on during shoulder periods (midweek leisure in a business-heavy property, weekend business in a leisure-heavy one). Local covers F&B, event, and day-use revenue that doesn't show up in RevPAR but pays rent.

Why Anchor over the others? Two reasons. First, Cornell research on brand-amenity ROI found the amenities that attract first-time guests and the amenities that drive repeat visits are not the same set, and that ROI is guest-type specific.² You can't build a listing that's optimized for every persona at once; you will underperform everywhere. Second, Cornell's pricing research across 67,008 hotel observations showed that hotels positioned with ADRs above their direct competitive set achieved higher relative RevPAR without proportional occupancy loss, regardless of economic cycle.³ Positioning above the comp set requires a clear Anchor: a persona willing to pay the premium because the property is structurally right for their trip. Without an Anchor, the only pricing lever available is discounting; the same Cornell data found hotels that priced below their comp set earned lower RevPAR without meaningful occupancy gains.³

The Anchor is the axis the other ICPs swing around. Yield pricing makes sense only when you know which weeks matter to the Anchor (so you know when to compress). Gap strategy is legible only once the Anchor is fixed (the Gap is the persona your Anchor-aligned comp set is also ignoring). Filler and Local are residuals.

How to identify your Anchor from observable evidence

The market-research pipeline this article backs operates at a specific evidence ceiling: llm-synthesis-over-places-reviews-grounded-search. That means it can see Places-verified competitor data, mined keywords from competitor reviews, and grounded-search output about the market. It cannot see PMS booking history, rate data, booking windows, or cancellation patterns. Every identification signal below is reachable from that ceiling, or it's called out as a framework claim.

Signal 1: A recurring, verifiable demand generator within ~3 miles

This is the strongest single signal because it's the one most directly observable from grounded search and map data. A demand generator is any location that predictably pulls overnight visitors: a corporate campus, a teaching hospital, a convention center, a university, an airport with meaningful connecting traffic, a sports arena with a full event calendar, a regional medical center, a state capitol. HVS's framework treats the presence and scale of these generators as the first-order feasibility question for a hotel site.¹ Leisure markets substitute attractions (beaches, national parks, historic districts, ski lifts) for business generators, but the logic is the same: something has to pull people into the catchment.

One-off generators do not anchor an ICP. A single festival week, a once-every-four-years event, a celebrity chef pop-up: these are Yield opportunities, not Anchor signals. The Anchor generator is the one that fills 40+ weeks a year. If grounded search surfaces a convention center with a published 12-month calendar, that counts. If it surfaces a hospital system with 800 beds and an active residency program, that counts. If it surfaces a single summer arts festival, that's a compression event, not an anchor.

Signal 2: Review-language dominance in the competitive set

The second signal lives in competitor review text. Text mining of hotel reviews, including the Guo, Barnes & Jia analysis of 266,544 reviews across 25,670 hotels, has repeatedly shown that review vocabulary clusters into roughly 15-20 stable dimensions (location, cleanliness, staff, breakfast, business amenities, family amenities, and so on) and that those dimensions vary systematically by hotel star rating and guest type.⁴ For Anchor identification, the question is not which dimensions appear (they mostly all do) but which dimension dominates the positive reviews of your direct competitors.

A business-hotel Anchor shows up as recurring language about workstations, quiet rooms, fast Wi-Fi, airport shuttle reliability, breakfast timing, and check-in speed. A medical-visitor Anchor looks like language about proximity to a named hospital, late check-in flexibility, extended-stay suitability, kitchenette access, and parking. A leisure-destination Anchor looks like language about views, pool, walking distance to attractions, family rooms, and staff warmth. The specific content is less important than the share: as a working rule of thumb (not a measured cutoff), if roughly a third or more of the 4+ star reviews across the comp set keep returning to one vocabulary bucket, that bucket is the competitive set's working Anchor, and it's probably yours too unless you have a structural reason to defect. You don't need a text-mining tool to read this signal by hand: pull the most recent 30 four-plus-star reviews for each of your top three competitors and look for the noun that keeps recurring, not an exact percentage. If "hospital," "appointment," or "treatment" shows up across all three, that's your answer.

Negative-review language is a separate and equally useful signal. Research on review text has found that negative reviews tend to concentrate on a narrow set of topics, while positive reviews touch on more.⁴ The concentrated complaint topics in your comp set often describe the Gap persona, not the Anchor: the persona competitors are failing, which you might be able to serve.

Signal 3: Structural fit and property amenities

The third signal is the match between what the property already is and what the persona needs. Cornell's work on amenity ROI is explicit that amenities are not fungible: internet access drives first-time bookings, complimentary bottled water drives repeat business, and fitness-center ROI is near zero for both groups over a 12-month measurement window.² The lesson for Anchor selection is the inverse: don't pick a persona whose baseline needs require amenities you don't have and can't realistically add. Pick the persona whose requirements your property already satisfies.

A 24-room property with no meeting space, no business center, and no on-site F&B after 9pm cannot Anchor around corporate groups no matter how close the corporate campus is. It can Anchor around the individual business traveler who wants a quiet room, fast Wi-Fi, and a walkable breakfast: that's what the property already delivers. A 140-room property with two ballrooms and a catering kitchen has the opposite structural bias. The Anchor you pick must be reachable from the property you actually have.

Signal 4 (framework only): Price tolerance matches your ADR band

This signal is a check, not a driver, and the market-research pipeline cannot directly observe it because rate data sits below the evidence ceiling. The framework claim is: if the persona you've hypothesized as the Anchor has a typical price tolerance of $80-$120 and your property's rack rate is $220, either the Anchor is wrong or the pricing is wrong. Either fix is expensive. A hotelier can validate this signal directly against PMS data; the pipeline can only flag when the competitor ADR band (grounded-search output) and the hypothesized persona's trip-type don't appear consistent.

Decision matrix

Signal	Weight	Where the pipeline sources it
Recurring demand generator within ~3 mi	High	Places + grounded search
30%+ share of 4+ star competitor review vocabulary	High	Review miner (linguistic segmentation)
Property amenity and structural match	High	Places amenity tags + property Ideal Listing
Price tolerance matches ADR band	Medium (framework only)	Out of pipeline scope; hotelier check
Repeat-guest mix, booking window, LOS	Medium (framework only)	PMS data, out of scope

The three High-weight signals are the ones the market-research report can produce from its evidence ceiling. The Medium-weight signals belong in the hotelier's own validation pass.

What changes downstream when the Anchor shifts

Once the Anchor classification changes (whether because the generator landscape changed or because the original Anchor was picked wrong) four things should re-sort.

Photos. The hero image and the first 5-7 slots in the photo sequence should be legible to the Anchor in under two seconds. A medical-visitor Anchor needs the exterior and the king-bed room up front, with parking and the kitchenette visible in the first scroll. A business Anchor needs the desk, the view from the desk, and the meeting space. A leisure Anchor needs the pool, the view, and the lobby. The shot menu doesn't change; the priority order does.

Title and copy. The top-line description should name the Anchor's trip trigger directly. "5 minutes from Memorial Hospital" beats "convenient location downtown" for a medical Anchor. "Walk to Convention Center" beats "centrally located" for a group Anchor. This is also observable in competitor titles and short descriptions via grounded search; competitors clustered around one Anchor tend to converge on similar trigger language.

Rate band. Cornell's pricing research found hotels that positioned ADRs above their direct competitive set outperformed on RevPAR without giving up meaningful occupancy, across all chain scales and in multiple economic regimes.³ The Anchor is the persona whose willingness-to-pay supports the premium. If the Anchor shifts to a lower price-tolerance persona, holding the old rate structure usually produces falling occupancy without offsetting ADR gains. If the Anchor shifts up, rates should move with it; within the same study, hotels that underpriced their comp set captured lower RevPAR and didn't meaningfully boost occupancy.³

Which compression events get cover-page billing. Compression events (periods when the market runs at constrained inventory) are where Yield pricing lives.⁵ The Anchor decides which compression events belong on the property's marketing surface. For a medical Anchor, the relevant compression is graduation week for the residency program or a regional medical conference, not the citywide marathon. For a group Anchor, it's the convention center's biggest trade-show week. For a leisure Anchor, it's the peak season and the named events that drive it. Compression events matter to every hotel; which ones go in the hero slot matters to the Anchor.

Direct-booking infrastructure (loyalty, brand site) typically pays back fastest when it's built around the Anchor, because the Anchor is the repeat-visit engine, and direct bookings from repeat guests carry materially lower acquisition costs than OTA bookings; Kalibri Labs research puts OTA acquisition costs at 13-17% of room rate versus 3-8% for direct, with overall customer-acquisition costs averaging 15-25% of guest-paid revenue.⁶ And total-revenue management, as HSMAI has framed it, works cleanly only when the segmentation it's optimizing against is stable; an unstable or misidentified Anchor breaks the optimization loop.⁷

Common failure modes

Picking the highest-volume segment as the Anchor. Volume fills nights; Anchor defines fit. A property booking 60% leisure on weekends in a business corridor may still have a business-traveler Anchor if that persona drives above-comp-set ADR.

Ignoring structural mismatch. A property with no meeting space and no F&B after 9pm cannot anchor corporate groups regardless of generator proximity. Choose the persona your existing property already satisfies.

Treating the Anchor as permanent. Generator landscape shifts: a new hospital wing opens, a major employer relocates, a convention center expands. Revisit the Anchor classification annually.

Conflating the Gap persona with the Anchor. The Gap is the persona your comp set is failing. The Anchor is who you already serve best. They are rarely the same, and optimizing for the Gap before solidifying the Anchor usually produces a confused listing.

Using review language as the only signal. Review-language dominance, demand-generator geography, and structural fit all need to agree. One signal pointing to a persona is a hypothesis; three signals pointing the same way is a classification.

Self-audit checklist

• I can name a recurring demand generator within 3 miles with a 40+ week annual cadence
• I have read 4+ star reviews of my top 3 competitors and can name which vocabulary bucket dominates
• My property's existing amenities satisfy my claimed Anchor persona's baseline needs without capital investment
• My rack ADR is consistent with the price tolerance I attribute to the Anchor
• I have checked whether the Anchor classification still holds within the last 12 months

How OTALift surfaces this

The market-research report derives the Anchor ICP from three pipeline-visible signals: a recurring demand generator (Places + grounded search within 3 miles), review-language dominance in competitor reviews (ReviewMiner bucket counts), and structural fit (property amenity tags from the Ideal Listing). When the Anchor is identified or reclassifies, three recommendations fire: "Align top-of-funnel marketing to the Anchor ICP," "Re-photograph primary rooms with Anchor-persona staging," and "Name the demand generator in the property title."

Two framework signals (price tolerance matching ADR band and repeat-guest mix) sit below the evidence ceiling. They are flagged as hotelier-validation steps, not pipeline outputs; PMS data and booking history are required to confirm them.

Related articles

• Linguistic Segmentation from Review Keywords: how the review-mining layer that produces Signal 2 actually works
• Compression Events: Detection and Response: what to do with the Yield windows once the Anchor is fixed
• Pillar: The Hotel Revenue Flywheel: how Anchor selection feeds photos, pricing, direct-booking strategy, and review response

Sources and methodology

---

Authored by Anya Cortez · Reviewed by Anya Cortez · Last reviewed: 2026-04-21

Footnotes

1. Bisema, Brian F. (HVS Boston). "Hotel Demand Segmentation 101: Identifying Demand Segments and Understanding How They Relate to Property Performance." December 17, 2009. https://www.hvs.com/article/4150-hotel-demand-segmentation-101-identifying-demand-segments-and-understanding-how-they-relate-to-property-performance. Source for: the three-segment commercial/group/leisure framework, day-of-week and season-of-year demand patterns, and the role of demand generators in hotel feasibility. ↩ ↩² ↩³ ↩⁴

2. Dev, Chekitan S., Rebecca Hamilton, and Roland Rust. "Hotel Brand Standards: How to Pick the Right Amenities for Your Property." Cornell Center for Hospitality Research / Cornell Hospitality Report, 2017. https://ecommons.cornell.edu/items/31aefb29-d708-4565-ab2d-e47ba07f9066. Source for: amenity ROI varies by guest type; internet access drives first-time visits, complimentary bottled water drives repeat visits, fitness-center ROI is near zero over 12 months; guests overestimate their own amenity use. ↩ ↩²

3. Enz, Cathy A., Linda Canina, and Mark Lomanno. Cornell Center for Hospitality Research. Competitive pricing research across 67,008 hotel observations showing that hotels priced above their competitive set achieved higher RevPAR without proportional occupancy loss, and hotels priced below their comp set earned lower RevPAR without meaningful occupancy gains. Extended to European markets in "Competitive Hotel Pricing in Europe: An Exploration of Strategic Positioning" across 4,000+ hotels, 2004-2013. https://ecommons.cornell.edu/server/api/core/bitstreams/8870ad6b-0e73-4e9c-a4f9-9026fdb1e09a/content ↩ ↩² ↩³ ↩⁴

4. Guo, Yue, Stuart J. Barnes, and Qiong Jia. "Mining meaning from online ratings and reviews: Tourist satisfaction analysis using latent dirichlet allocation." Tourism Management 59 (2017): 467-483. https://www.sciencedirect.com/science/article/abs/pii/S0261517716301698. Source for: 266,544 online reviews across 25,670 hotels in 16 countries analyzed by LDA; 19 controllable dimensions identified; dimensional differences by demographic segment and star-rating; concentration pattern in negative reviews versus broader coverage in positive reviews. ↩ ↩²

5. CoStar / STR reporting on event-driven compression. "Hotel Performance Is Becoming Increasingly Dependent on Event-Driven Demand." 2025. https://www.hotelnewsresource.com/article140916.html. Source for: ADR and RevPAR reaching new highs during concentrated-demand event periods in markets including Milan, São Paulo, and Paris; event-driven demand as a widening share of RevPAR growth. ↩

6. Kalibri Labs research on hotel guest-acquisition costs. OTA acquisition costs average 13-17% of room rate versus 3-8% for brand.com/direct; overall customer-acquisition costs average 15-25% of guest-paid revenue, with some properties reaching 35%. Aggregator citation via Hotel Management magazine because Kalibri Labs' primary reports are not publicly hosted. https://www.hotelmanagement.net/operate/consumer-acquisition-costs-too-high-and-growing-says-kalibri-labs-ceo and https://www.hotelmanagement.net/operate/new-study-from-kalibri-labs-shows-direct-bookings-push-working ↩

7. HSMAI (Hospitality Sales & Marketing Association International). "Essential Market Segmentation for Total Hotel Revenue Management." HSMAI Academy. https://academy.hsmai.org/market-segmentation-essential-for-total-revenue-management/. Source for: the total-revenue-management framework and its dependency on stable segmentation as the substrate for pricing, promotion, and distribution optimization. ↩