340 Rockford businesses.
165 of them never showed up.

How I ran it, so you can check my work

• Who’s in it. Businesses that come up in Google Maps for their trade in the Rockford area and have a website of their own. Nobody was picked for having a bad site. I didn’t look at any of the websites before running the search.
• The search. One query per trade, the exact wording printed next to each row above, run against real Google from a Rockford location on desktop. First ten organic results.
• What counts. The business’s own domain appearing in those ten. Being named inside somebody else’s list does not count.
• When. September 14, 2026. Results move around, so this is a snapshot, not a permanent scoreboard. Re-run it yourself and you may get slightly different rows.
• The AI test. Every one of the 97 trades, same 340 businesses, nothing sampled. An AI assistant with live web access got the same buyer question and was told to answer the way it would answer a person: name a few places. It never saw my list. I then counted how many of the businesses I had checked it actually named. It answered every question, so nothing is counted as missing because the assistant refused.
• The limits of it. This is one assistant on one day, not every assistant forever. Ask the same question next month and some names will move. What does not move is the shape of it: an answer holds three or four names where a page holds ten, so most businesses are out by arithmetic before quality is even discussed.
• No names. I won’t publish a list of local companies with their names attached to a bad result. If yours is one of the 165, I’d rather just tell you.

Then I asked the AI directly

Google was the easy test. The AI is worse.

Everything above is Google, and there’s a fair objection to it: fine, but you sell AI visibility, so show me the AI. So I ran the whole thing again. All 97 trades, an AI assistant with live web access, the same buyer question, told to answer the way it would answer a real person. Name a few places. It never saw my list.

Of all 340 businesses, the assistant named 97. It never mentioned the other 243. That is 71.5% invisible, against 48.5% on Google.

That is the whole thing in one line. Page one has ten slots. An answer has three or four. Being findable quietly stopped being enough. You have to be one of the three.

Here is the part that should worry anybody who has been paying for SEO. 91 of these businesses were sitting on Google page one and still never got named. 18 of them were in the top three. Ranking and being recommended are now two different jobs.

In 30 of the 97 trades the assistant named nobody I had checked. Not one. Roofers, barbers, plumbers, lawn care, bakeries, dance studios. It had plenty to say in every single case. It just said other names.

Who is standing in your spot

About 38% of all first-page slots went to a directory rather than a business. Yelp, Angi, BBB, Expertise, a “top 10 in Rockford” roundup written by somebody who has never set foot here.

69 / 97

searches with Yelp in the top 5

68 / 97

searches with a Facebook page in the top 5

31 / 97

searches with Reddit in the top 5

39 / 97

searches where a directory held the #1 spot

Read the Facebook one again. In 68 of 97 searches a Facebook page was sitting in the top five, quite often belonging to a business whose actual website was nowhere at all. Their free Facebook page is beating the site they paid for.

In 25 trades, half or more of page one was directories. Plumbing was the worst: eight of the ten slots. A Rockford plumber with a perfectly decent website is fighting for two spaces.