Taste Boulder works. The question is where to point the same machine next, and the honest answer is not "the biggest city".

The test Boulder passed

Boulder was worth building for one reason: page one for best restaurants boulder had no dedicated restaurant guide on it. A travel blog, the tourism board, a Reddit thread and two individual restaurants. That absence was the whole opening.

So a city qualifies on three things, in this order:

1. More search demand than Boulder. Boulder's two head terms total 39,200 searches a month.
2. No harder to rank for. Boulder measures 24 and 15 on keyword difficulty.
3. A soft page one. This is the one that actually decides it, and the only one you cannot get from a difficulty score.

What the screen found

78 US cities larger than Boulder were pulled for volume and difficulty. 27 cleared the first two tests. Then page one was read for every one of them, and real backlink authority pulled for all 112 domains that rank.

Two numbers matter per city:

• real competitors: results that are not an aggregator, listicle, social post or tourism board. Yelp and Reddit ranking is the opportunity, not the obstacle.
• weakest: referring domains of the least-linked real competitor. That is the bar to clear, not the strongest site on the page.

The answer: Baton Rouge

Boulder itself has 7 real competitors on page one at a median of 342 referring domains. We ranked anyway. Every city below is easier than that.

Baton Rouge, in one line

1.85x Boulder's search demand, and exactly one real competitor on page one. Everything else ranking is TripAdvisor, Yelp, Reddit, Facebook and Time Out. The single real site is a regional magazine with 164 referring domains. Boulder had seven competitors at a median of 342 and we still ranked.

The runners up

• Tulsa: three real competitors at a median of 132, the weakest concentration of authority in the set. 64,300 searches.
• Greensboro: four real competitors at a median of 100, the weakest is on 4 referring domains. 72,600 searches.
• Lubbock: four real competitors at a median of 118. 70,400 searches.

The traps

These look good on volume and are not:

• St Louis and Pittsburgh have the biggest markets in the set, and real city magazines hold page one: St Louis Magazine, STLtoday, Pittsburgh Magazine, Discover the Burgh. Those are exactly the established local guide the Boulder test says to walk away from.
• Colorado Springs has only one real competitor too, but it is Springs Magazine on 852 referring domains, five times Baton Rouge's.
• Minneapolis has Condé Nast Traveler ranking, on 58,945 referring domains.

What this does not tell you

Search demand and page-one weakness are necessary, not sufficient. Before committing:

• Check the restaurant count. The whole product is menu prices at scale. A city with 300 restaurants supports a thinner site than one with 900.
• Check that menus are readable. Boulder needed a paid scraper to recover 115 menus that ordinary crawling could not get. A city dominated by one ordering platform could be harder or easier.
• Local knowledge still matters. Boulder's dish taxonomy had to learn that "Flatiron" is a mountain. Every city has a word like that.

The screen costs about six cents to run for a new set of cities, so it is cheap to re-check before committing to a build.

Generated 2026-08-02 from live search volume, keyword difficulty and backlink data.