Another quality update, and your city pages

Google started another product reviews update on 27 July and finished rolling it out at the start of this month. Every local owner we spoke to last week skipped that headline, and fairly: dental practices and HVAC companies do not publish product round-ups.

Skip the headline, but not the pattern. That is the fourth product reviews update since the spring of last year, sitting alongside a core update in May, and they all aim at the same thing. Pages built to fill a slot in the results rather than to be read by anybody. The local version of that problem is sitting on most local business sites right now, and it is called the city page.

What these updates keep aiming at

Product reviews updates go after the round-up written by somebody who never touched the product. Specs copied from a manufacturer page, a paragraph of filler either side, a buy button. Google’s stated test is whether the page shows evidence of real use and real expertise, or whether it is words arranged around a keyword.

That is not a new idea. Google’s guidance has said to write for readers rather than for the search box for as long as the guidelines have existed, and the doorway pages policy has named the exact local case for years: sets of pages targeting specific cities or regions that funnel everyone into the same place.

What has changed is enforcement. Google now ships these updates several times a year and they land hard enough that people notice. If your site has twelve pages that differ only by a city name, the policy that covers them is not hypothetical any more.

The local version: the stack of city pages

Almost every local site we take over has it. Twelve, twenty, sometimes sixty pages, one template, the city name swapped, occasionally a sentence about the town’s population pulled from somewhere.

Owners keep them because somebody built them in 2016 and they seemed to work. Two things are worth being blunt about.

First, the map results were never coming from those pages. The pack is driven by the profile, the categories, proximity to the searcher and prominence. A page titled “Emergency Dentist in Springfield” does not put you in the Springfield map pack. Nothing on your site does that on its own.

Second, below the pack, that page is competing in the ordinary organic results against pages that genuinely have something to say about Springfield. A swapped variable does not win that, and twenty of them do not win it twenty times. What they do is spread the same thin content across the site and make the whole domain look like a template.

What a city page has to do to earn its place

The test we use is simple. Could somebody who lives there read this page and tell it was written about their town? If not, it should not exist.

A page that passes usually has most of this:

• A real reason to exist. You serve the place, you have customers there, somebody on staff can talk about it.
• Specifics a local recognises. Neighbourhoods, the road people come in on, where to park, how long the drive takes at five o’clock.
• Evidence of work done there. Photos of a job, a short description of it, a review from somebody in that town.
• The details that actually apply. Which hours, which team, which phone number, and where you will and will not drive.
• Copy, title and heading written from scratch. Not one paragraph with a variable in it.
• A next step above the fold on a phone, because most of this traffic is on a phone.

If you cannot do that for a city, do not make the page. One honest service area page that names the towns you cover beats twelve thin ones, and it is a great deal less work to keep true.

Service pages have the same problem in reverse

The other half of this is the site that lists eleven services in one paragraph each on a single page, then wonders why it never ranks for any of them.

If somebody searches for a service by name, buys it separately, and would compare providers on it, it needs its own page. Not a stub. A page that says what the treatment or the job is, who it suits, what happens step by step, how pricing works, what the common questions are, and what to do next. Photographs of your own work rather than stock. The name of the person who does it.

That is more writing than most owners want to do, which is why the shortcut exists. It is also the only version that holds up when an update lands, because it is the only version that a reader would choose to finish.

While you are in there, two things that quietly matter as much as the words: how fast the page loads on a phone on a weak signal, and whether the form or the phone number is reachable without scrolling. A page that reads well and takes six seconds to appear is still losing.

What to do this month

• List every page on your site and mark the ones that exist only to hold a city or a keyword. Be honest about the count.
• Pick the two or three towns you genuinely serve and rewrite those pages properly. Real photos, real details, real next step.
• Merge the rest into one service area page and redirect the old URLs to it rather than deleting them.
• Split any service you actually sell separately onto its own page, with the step by step and the pricing explanation.
• Check the top ten pages on a phone for load time and whether the call button is visible without scrolling.
• Leave your profile alone while you do this. The pack and the website are separate jobs and changing both at once makes the results unreadable.

None of this is a reaction to one update. It is the direction every update has pushed for years, and local sites have been slower to feel it than publishers. This is the bulk of what our local SEO work actually consists of, it is why we rebuild the website before adding pages to it, and it is worth remembering that the Google Business Profile carries the map results no matter how good the page is.