Nobody Searches Your City Page

01Fifty three cities, one question

A legal information site I run has a page for 53 cities, each written for that city with its courts, its roads and its deadlines. They were the most expensive pages on the site to research. In the 90 days to 28 October, all 53 together were shown in Google 11,670 times and clicked 110 times. The median city page was shown 110 times in three months. In the same window a single article answering one plain question, how much of a settlement a person actually keeps, was shown 141,574 times and clicked 1,362 times. A page can rank in the top ten for a search almost nobody makes.

They rank. That is the strange part.

The city pages are not buried. The largest city page sits at an average position of 12.8, the next at 11.2, a big Texas city at 8.9. Those are positions that usually mean traffic. Here they mean a handful of impressions, because the searches they answer are rare. People in a city with a legal problem mostly do not type the city into the question. They ask the question, and the search engine and the assistants work out the place from where they are.

02The place is implied, the question is typed

Search already knows where you are

When a person searches for what a settlement is worth, the results are already tuned to their location, and the assistant they ask can see roughly where they are too. Adding the city to the search is redundant, so most people do not. The city page is written for a query that the searcher does not need to make. Local intent is real; local wording is rare. State pages do better than city pages, about eight times the impressions per page, because state law genuinely changes the answer and people know it, so they include the state.

Pages that compute beat pages that describe

The calculators average 2,101 impressions per page against 220 for a city page, and earned as many clicks as all the state pages together. That repeats what I found across 41 sites in calculators get quoted: a page that does something for the reader's own numbers is searched and clicked far more than a page that describes. And it repeats the concentration in one article earns everything. A single page answering the question people actually type outdrew 53 carefully researched pages combined by twelve to one.

03Before you build city pages

Measure the city searches first

Before writing a page per city, check how often the city appears in real searches for your topic, in Search Console if you already rank for the question, or in a keyword tool if you do not. If the question with the city attached is searched a few times a month, a city page will earn a few impressions a month however good it is. Research effort is not demand, and the search engine only rewards demand. Put the local facts that genuinely change the answer, a deadline, a fault rule, a court, into the state page or the question page, where people already arrive.

When city pages are worth it

They are worth it when the place is part of the question: a crash report from a specific police department, a permit from a specific office, a service area where the provider itself differs by city. They can also earn links from local organisations, which a general article rarely does. And they are not worthless if they are already built. Ranking pages cost nothing to keep, and the occasional person who does search for the city finds a good answer. The mistake is building more of them before the first ones show demand, the same lesson as the thin pages that earn, read in the other direction. Let the first ten pages of a type prove demand before you build the next fifty.

Tomorrow: a title suffix that pushed 207 titles past the limit.