151x: The Search Gap Between the Name BWH Is Retiring From the Masthead and the One It's Promoting

On October 27, BWH Hotels announced it is now WorldHotels. Four thousand hotels, eighteen brands, one masterbrand. Within a day the comparison everyone reached for was Dunkin’ Donuts dropping “Donuts” — usually as a cautionary tale.

We went looking for the data to support that story. We found something more useful: two of our own starting hypotheses collapsed under testing, and what survived is worse for WorldHotels than what we set out to prove.

Before anything else, the thing most coverage is getting wrong in the comments: the hotel brands are not being renamed. Skift, Hotel Dive, CoStar, Travel Weekly and HOTELS all report the eighteen brands keep their identities. A Best Western stays a Best Western. What changed is the parent company’s name, and in January, the loyalty program.

That second part is the one with a date on it.

The gap, measured nine ways

Twenty-nine times at the narrowest, six hundred and twenty at the widest, clustering around 100–200x.

One honest caveat we want on the record, because it’s the kind of thing that gets overstated: these are two instruments measured nine ways, not nine independent instruments. The branded keyword universe, property-level search, organic traffic and traffic value all derive from the same underlying volume tables. What matters is that the two genuinely independent ones — Google Ads reported volume and Ahrefs clickstream modelling — land on the loyalty ratio almost identically: 104x and 102x.

That loyalty number is the one to watch. A parent-company name is something hoteliers read in a trade journal. A loyalty program is something seventy million people log into. On January 5, Best Western Rewards and WorldHotels Rewards become one program, and the name on the door is the one that gets 260 searches a month.

Dunkin’ is the benchmark. Almost nobody cites it accurately.

In January 2019 the word Donuts came off the brand. Per the FY2018 10-K there were 12,871 points of distribution at the time, 9,419 of them in the US.

Here’s the part that gets left out of the deck.

Demand for the retired name went up for three years after it was retired, peaked in 2021, and has taken seven years to give up a fifth of its ground.

We’ll flag our own error here, because we published the first version of this analysis with a worse number. An earlier draft said demand was up 3.6% — that compared January 2018 to November 2026, two single months, and January 2018 happens to be a seasonal trough. On annual averages it’s −21.8%. The corrected curve is a better argument than the flattering one was: a decline that didn’t even begin for three years is exactly how slowly retired brand demand actually dies.

Three things Dunkin’ did that matter more than the chart:

• They never moved the domain. Eight years on, the website is still dunkindonuts.com. They never even owned dunkin.com — the Wayback capture from January 2019, the month the rename took effect, shows a squatter page titled “DunkinWorks NOT Dunkin Donuts.” No migration ever happened. This is the cleanest lesson available and it’s the one the hotel industry should copy.
• The store count went up, 9,419 US locations at FY2018 close to 9,630 at FY2019. Neither filing discloses any rebrand-related risk or impact.
• They said the old word out loud, on purpose. From the CMO’s announcement letter: “while donuts may no longer appear in our logo, we remain committed to serving our signature donuts… There is no Dunkin’ without donuts!”

One number we won’t use, because it looks like evidence and isn’t: Q1 2019 comparable sales grew 2.4% on ticket, “partially offset by a decrease in traffic.” FY2018 — entirely before the name took effect — used identical language. The traffic decline was a continuation, and the rename shipped alongside an espresso relaunch, mobile ordering, a store redesign and menu simplification.

Dunkin’ didn’t win because demand followed the rename. It won because it never asked demand to.

We thought the profile name was the whole story. Germany proved us wrong.

Here was the theory, and it’s a good one. The Google Business Profile name is a ranking input, not a label. Dunkin’ renamed roughly 9,400 US profiles, deleting the category keyword from the single highest-leverage field a local business controls. If that costs visibility, hotels appending a brand descriptor should worry.

It’s testable, because Dunkin’ never finished the job abroad.

That’s a null result and we’re publishing it as one. If deleting “Donuts” from 9,400 profile names had cost Dunkin’ its local pack position, the German listings should still hold a position the American ones lost. They don’t.

It also kills the inference that looked most like a mechanism. Yes, 62% of top-ten donut results have “donut” in the business name — but that’s because independent donut shops name themselves after donuts. The name is a symptom of being an independent donut shop, not the cause of the ranking.

What the failed test exposed instead

Chasing the dead theory turned up something considerably more useful.

The largest donut chain in America does not rank organically for the biggest query in its own category — in either naming configuration, in either country. Three hundred thousand searches a month, and nothing.

And that generalises to hotels in a way the name theory never did:

Nobody wins category local packs on brand strength. Your profile name’s only real job is winning your own name.

And WorldHotels has decided to spend that field on something else.

The finding nobody has published

This needs no forecasting. The naming convention is already live, and we measured it in characters on the real profile strings.

For the Inn at Mendenhall, the program descriptor is longer than the hotel’s own name. Google’s business-name policy asks for the real-world name of the business, which makes a program tier appended across four thousand profiles a compliance question as well as a marketing one.

Compare what Marriott does with Autograph Collection, which is the endorsed-brand model done properly. When Marriott took on Hotel Park City, the Park Record reported that no change to the hotel’s name was anticipated. The property name survived; the collection rode behind it. The endorsement is a suffix, never the subject.

The measurement we couldn’t make

The obvious study: take properties that joined WorldHotels, measure branded demand before and after. We tried it on five. It doesn’t work, and the reason is worth more than the study would have been.

Look at what Google Ads returns:

Three distinct strings returning exactly 5,400 is not a coincidence, and a nine-word string cannot out-search the four-word hotel name inside it. These are one close-variant cluster being reported several times. On three of five properties the appended name scored higher than the core name, which is impossible as standalone demand. Any before/after built on this would be measuring Google’s keyword grouping, not guest behaviour.

There’s a second, fatal problem even with clean data: the appended string couldn’t have been typed before the tier existed, so its “before” is undefined rather than zero, and any growth rate is trivially infinite.

So we’ll pose it as a question and leave it there. Nobody is searching for the new name, and the old name is doing all the work. Whether re-flagging caused anything, we can’t say — and with five hotels we couldn’t say it even if the tools cooperated. At least five other explanations fit as well or better:

1. The pandemic, on five different curves. Washington DC, Miami, rural Pennsylvania, Vancouver and Buenos Aires recovered on wildly different timelines. Pooling them would be meaningless.
2. Discovery moved off Google web search. Hotel-name lookups increasingly happen inside Booking, Expedia, Google Hotels, Maps and now AI assistants — suppressing measured query volume without touching demand.
3. Market-specific macro distortion. Argentina ran 100–300% inflation. Hotel Madero’s real demand is Spanish-language and regional, so a US-only series measures American curiosity and nothing else.
4. Tool quantisation. Google Ads reports on a fixed ladder — 1,300, 1,600, 4,400, 5,400, 6,600. One rung is an 18–50% step, so a “decline” may be no decline at all.
5. There was no “independent” before. The Inn at Mendenhall was a Choice Hotels Ascend Collection property from January 2009 until spring 2019, confirmed from Choice’s own branded property URL across roughly 45 archived captures. It didn’t go independent → WorldHotels. It went soft brand → soft brand, from a direct competitor.

A defensible version of this study needs forty-plus treated properties against matched controls in the same submarkets, event-time aligned to verified flag dates. On public keyword data, for a collection this size, that study probably can’t be built — which is itself a finding about the limits of what any of us can claim from a keyword tool.

Soft-brand membership is musical chairs

We found this one by accident, checking profile names.

The Beacon Hotel & Corporate Quarters in Washington DC appears in WorldHotels’ own ranking keyword set as beacon hotel & corporate quarters member of worldhotels distinctive, reported at 4,400 searches a month. Its Google Business Profile now reads:

The Beacon at Embassy Row Washington DC, Tapestry by Hilton

One review on that profile — which is what a freshly created listing looks like. The hotel renamed, re-flagged to Hilton’s soft brand, and restarted its local presence from zero. WorldHotels’ keyword footprint still carries it as a ghost.

Then the archive turned up the same pattern running the other way, in the Mendenhall history above. And its sibling property down the road in Concordville had been a Best Western Plus since at least January 2010 — so one owner ran a Choice hotel and a Best Western hotel side by side for nine years, then consolidated onto a single parent.

That last detail is a confounder and we’re treating it as one: the owner had a pre-existing commercial reason to move that had nothing to do with what WorldHotels offered. But it’s also the point. In two of the handful of properties we looked at closely, the flag changed within living memory — one into WorldHotels from a direct competitor, one out of it to Hilton. Each move rewrites the hotel’s name in the one field that decides whether it wins its own name.

What would prove us wrong

A hit piece that can’t be wrong isn’t analysis. Here’s what would make us retract:

• The property names genuinely stay first. If “Best Western” leads on every sign, profile and title tag and WorldHotels stays a footer line, the search downside is close to nil and this is a B2B identity exercise.
• bestwestern.com is never retired. Do what Dunkin’ did — keep the domain, keep the entity, never ask demand to move — and the 151x gap stops mattering, because nobody is asked to search the new name. As of our pull, both domains still resolve independently with no redirect between them.
• The loyalty transition keeps the old query alive for years. If best western rewards resolves to a working login indefinitely, the 104x gap is a non-event.
• They outspend it. Marriott Bonvoy didn’t exist before 2019 and now does 1,500,000 searches a month. It’s purchasable. It cost years and nine figures — and the company has said owners won’t be charged for the rebrand, so the open question is who funds the demand generation.

Three of those four are decisions, not outcomes. They’re all still available, which is the reason to publish this now rather than in eighteen months.

So, can it work?

Yes. And the honest version of that yes is more interesting than a no would have been.

Dunkin’ renamed a 9,400-store chain, grew the store count, and the sky did not fall — because it never moved its domain, never fought the retired word, and told everyone there is no Dunkin’ without donuts. Eight years on, its own website still sits at the old name’s address, and demand for that old name has only given up a fifth of its ground.

WorldHotels is starting at 6,600 a month against a million, merging the loyalty program in January, putting its program tier in the one profile field that exists to win a hotel’s own name, and has already run this play once — bwh hotels returns 4,400 searches a month after years as the corporate identity.

Two of our hypotheses died proving this. The 151x survived all of it.

I guess we’ll see.

Method

All figures pulled 2026-11-05/06, US market unless stated. Google Ads volumes from DataForSEO keywords_data/google_ads/search_volume (exact, close variants grouped, reported on a fixed ladder). Ahrefs figures from API v3 Keywords Explorer, clickstream-modelled, variants separate. The two providers disagree by an order of magnitude on “near me” absolutes — Google Ads reports dunkin donuts near me at 4,090,000 against Ahrefs’ 347,000 — so every figure names its provider and the two are never mixed inside one comparison. The Dunkin’ time series is Ahrefs throughout.

Branded keyword universes and own-name positions from DataForSEO Labs ranked_keywords, top 6,000 keywords per domain, filtered to the brand token; property-level defined as three or more words. Local packs from the DataForSEO Maps SERP API. Domain traffic and traffic value from Ahrefs Site Explorer, subdomains mode. Google Business Profile name strings observed directly. Redirect status by direct HTTP request.

Primary sources: Dunkin’ Brands FY2018 and FY2019 Forms 10-K and the Q1 2019 8-K via SEC EDGAR; the Dunkin’ CMO’s announcement letter of 2018-09-25; Wayback captures of dunkin.com from January 2019 and of Choice Hotels’ Ascend Collection property URL from 2009 onward.

Limits. No before/after organic traffic curve for dunkindonuts.com exists here — Ahrefs historical site metrics aren’t available on our plan. Reddit and Moz couldn’t be searched, so practitioner discussion of the Dunkin’ local impact is unverified rather than cleared; no case study, trade-press analysis or franchisee complaint about Dunkin’ local visibility after the rename was found in roughly 60 queries. A hotel-side local pack test was attempted and discarded after bad geo targeting left one usable cell. Whether Choice Hotels still publishes a live listing for the Inn at Mendenhall could not be established — a possible duplicate listing under two rival soft brands is unverified, not ruled out. Hotel Madero’s tier-assignment date could not be established from any source.