The State of
Salinas Business
Websites.

That is 21.1% — a little over one in five. The remaining 470 (78.9%) have an address on their listing that answers with a real page.

21 of the 617 are absent from that figure and from every other figure on this page. Twenty of them run bot protection that turned an automated request away twice, and a website that refuses a script is not a broken website — scoring it as one would be the easiest way for a study like this to mislead. The twenty-first has an address on its listing with no top-level domain in it, which is a typing error on a directory rather than a fact about a business. That leaves 596.

A business with no website on its listing at least knows it has no website. These forty-four are the other case: they paid to be listed, they gave an address, and the address is dead. Nobody has told them, because the only people who find out are customers who do not call.

The twelve at the top of that list are the worst of it. Those domain names do not resolve anywhere — not to a parked page, not to an error page, to nothing. A customer who types one gets the browser’s own failure screen, which reads to most people as though the business has closed.

A further 4 listings send people to a www. address that does not resolve while the same address without the www. works perfectly well. We counted those 4 as having a website, because they do. What they have is a listing pointing at a hostname their own DNS does not serve, and it is the cheapest thing on this page to fix.

Everything from here on is measured against the 470 businesses whose site actually served a page, because those are the only ones with a page to read. Each line below is a plain absence — a tag that is not there, a link that is not there — not a judgement about design.

232 of the 470 — 49.4% — have no tap-to-call link anywhere on their home page. The number is usually printed there as text. On a phone, text is something a person has to read, remember, leave the page for, and type into a keypad, and a fair share of them do not come back.

This is the figure most exposed to the limitation described at the bottom of this page. We read each page as the server sends it, before any JavaScript runs, and 27 of the 470 sent almost no text that way. For those, the real answer may be better than what we recorded. It cannot be worse.

96 of the 470 live sites — 20.4% — put the word Salinas in their page title. The title is the line a search engine shows as the link, and it is one of the few things on a page that every search engine and every answer engine reads.

We are not claiming that adding a place name moves a ranking. This study did not measure rankings and says nothing about them. What it can say is that 374 of these businesses do not state, in the one line their search result shows, where they are — and that 166 of them use a title that is missing, generic, or nothing but the business name, which tells a searcher only what they already typed.

Five checks, each one either done or not done, none of them a matter of taste:

174 sites (37.0%) do all five. 54 do two or fewer. The shape of that distribution is the useful part: most of these businesses are not far off. The common case is a site that does four of the five and is missing one thing, usually the structured data or the description.

328 of the 470 live sites print a copyright year somewhere on the page. Of those, 66 (20.1%) print 2024 or earlier, and the oldest we found was 2006.

Treat this one gently. A copyright year is a weak signal: plenty of well-kept sites never print one at all, and 142 of these did not. It is evidence about a footer, not about a business. It is on this page because it was measured, and leaving out a measurement because it is inconvenient to interpret is how a study becomes an argument.

Speed, for the same reason: the median site answered in 0.69 seconds and the slowest tenth in over 1.86. That is one fetch each, from one machine, on one afternoon, and it is not a page-speed score. It is reported because we have it.

None of this needs an agency. All five are an afternoon for whoever built the site, and four of them are free.

The sample frame. Every business listed in the Salinas Valley Chamber of Commerce member directory on 23 August 2026. The directory returned 698 listings; merging entries that were plainly the same business twice — same domain, or the same name once the Inc and LLC are taken off — left 617 distinct businesses. This is a census of that directory, not a sample of it: every member was checked.

What was read from the listing. The business name, the category the chamber files it under, the address, and the web address the chamber shows. That last one is the only source for whether a business “has a website” here. A business could have a site the chamber does not list, and some surely do.

What was done with it. Each web address was fetched the way a browser would fetch it, following redirects, with the certificate checked rather than ignored. Anything that failed was tried again with a full browser header set. Anything that still failed had its domain resolved against two independent public resolvers before this study was willing to call it dead — a check added after a spot inspection caught this machine’s own resolver failing on a domain that was fine.

What was recorded. Whether the address resolved; whether it returned a page; whether the browser ends up on https; whether the page declares a mobile viewport at device width; whether it has a tap-to-call link; the page title, and whether it is more than the business name; whether there is a meta description, a main heading, and structured data; and the copyright year if the page prints one.

Who was left out, and why. Twenty businesses whose sites answered an automated request with a block page on both attempts, and one whose listed address has no top-level domain in it. Both groups are out of every fraction on this page rather than counted as failures.

Other directories, and why none of them are here. Four more were tried. YellowPages and Yelp refused automated access outright. The Better Business Bureau and Opendi both answered, and both were discarded for the same reason: neither shows a business’s own website on its listings at all, so importing them would have turned a quirk of two page layouts into several hundred businesses that “list no website”. That would have made the sample look three times larger and every headline figure worse, and it would have been an artefact.

The arithmetic. Every figure on this page was produced by a script from the collected data. None was estimated, rounded for effect, or typed from memory.

The findings are free to quote with attribution, and we would rather they were quoted accurately than favourably — if a figure here is wrong we want to know and we will correct it on this page.

The full anonymised dataset, the collection scripts and the aggregation script are available to any newsroom or researcher who asks, so the whole thing can be run again and checked. Row-level data naming individual businesses is not published and will not be.

Salinas Marketing, “The State of Salinas Business Websites”, August 2026. A census of the 617 businesses listed in the Salinas Valley Chamber of Commerce member directory, collected 23 August 2026.

Questions, corrections and data requests: the contact page.