Category: SEO

11Feb

How Google AI Is Changing the Way Local Businesses Win the Map Pack

For years, ranking in Google’s local map pack felt relatively straightforward. If your business was close to the searcher, had the right category selected, collected a solid number of reviews, and kept your Google Business Profile (GBP) updated, you had a strong chance of appearing in the coveted top three results.

But Google’s rapid integration of AI into search is quietly reshaping how local visibility works.

The traditional map pack isn’t disappearing overnight. Proximity still matters. Relevance still matters. Reviews still matter. But AI is changing how Google evaluates businesses — and that shift is raising the bar for everyone.

Local SEO isn’t dying. It’s evolving into a deeper game of authority, clarity, and trust.

From Keywords to Entities

Historically, local search leaned heavily on matching keywords, categories, and location signals. If someone searched “plumber near me,” Google primarily evaluated:

  • Proximity to the searcher
  • Category match
  • Basic prominence signals (reviews, citations, links)

Now, AI systems allow Google to understand businesses more like entities rather than just listings. An entity isn’t just a name and phone number — it’s a fully defined business with services, reputation, expertise, and connections across the web.

AI enables Google to analyze:

  • The specific services mentioned in reviews
  • The sentiment and context of customer feedback
  • The depth of service coverage on your website
  • Consistency of information across platforms
  • Brand mentions beyond your own website

This means local ranking is becoming less about minimal optimization and more about comprehensive digital presence.

The Rise of “Entity Confidence”

AI systems need confidence before recommending a business — especially in high-intent searches like “best divorce lawyer for high net worth cases” or “emergency roof repair after storm damage.”

Instead of simply displaying three nearby options, Google’s AI-powered experiences may highlight fewer businesses in certain contexts. When that happens, selection criteria become stricter.

What builds entity confidence?

  • Detailed service pages that clearly explain what you do
  • Reviews that mention specific services and outcomes
  • Strong engagement signals (clicks, calls, direction requests)
  • Consistent branding and information across platforms
  • Clear alignment between your website and your GBP

A thin profile with only hours, a phone number, and a handful of generic reviews is far less competitive than a fully developed digital footprint.

Reviews: From Star Ratings to Semantic Signals

Reviews have always influenced local rankings. But AI changes how reviews are interpreted.

Instead of simply counting reviews or averaging star ratings, AI can now analyze:

  • Service-specific mentions (“fixed my tankless water heater”)
  • Quality indicators (“transparent pricing,” “arrived within 30 minutes”)
  • Emotional tone and satisfaction signals
  • Patterns over time

A business with 75 detailed, service-rich reviews may outperform a competitor with 200 vague five-star reviews.

The takeaway isn’t to script keyword-heavy reviews — that’s risky and often against Google’s guidelines. Instead, it’s about delivering strong service experiences that naturally generate detailed feedback.

Review quality and consistency matter more than ever.

Website Depth Is No Longer Optional

Many local businesses have relied on minimal websites for years. A homepage, a contact page, maybe a services list — and that was enough.

AI-driven search changes this dynamic.

Google now cross-references your website with your Google Business Profile. If your GBP lists “commercial HVAC repair,” but your website barely mentions it, that inconsistency can weaken confidence.

Strong local performers increasingly have:

  • Dedicated service pages
  • Location-specific pages (when appropriate)
  • Structured data markup
  • FAQ sections addressing real customer concerns
  • Clear internal linking between related services

This doesn’t mean bloated content. It means clarity and completeness.

AI favors businesses that clearly demonstrate expertise in specific services.

Brand Signals Are Gaining Weight

Another quiet shift: Google appears to reward recognizable brands more aggressively in competitive markets.

Brand signals include:

  • Branded search volume
  • Direct website traffic
  • Mentions in local news or blogs
  • Social engagement
  • Backlinks from reputable sources

AI systems are good at recognizing real-world authority. If a business is frequently mentioned across trusted sources and generates consistent engagement, that strengthens its entity profile.

This doesn’t mean small businesses can’t compete. It means local marketing can’t rely solely on Google Maps optimization anymore. Broader visibility strengthens local rankings.

Proximity Still Matters — But It’s Not Enough

It’s important to clarify: proximity hasn’t disappeared.

If someone searches “coffee shop near me,” physical closeness still plays a major role. But in more competitive or specialized searches, proximity alone may not secure a top spot.

In the past, a nearby but mediocre listing might appear simply because it was closest. With AI-enhanced evaluation, a slightly farther but more authoritative business may outrank it.

The difference becomes more noticeable in saturated industries like law, medical services, home services, and financial services.

The End of “Set It and Forget It”

There was a time when claiming your GBP, choosing categories, uploading a few photos, and collecting some reviews could sustain rankings for years.

That era is fading.

AI-driven systems favor:

  • Ongoing review acquisition
  • Regular updates
  • Fresh engagement signals
  • Active reputation management

This doesn’t mean posting daily updates or obsessively geo-tagging photos. Many of those tactics are overstated.

It means maintaining an active, credible presence that reflects a real, operating business.

What Local Businesses Should Do Now

The shift toward AI doesn’t require gimmicks. It requires maturity.

Here’s a sustainable approach:

  1. Fully complete and regularly update your Google Business Profile.
  2. Build out detailed service pages that match your GBP offerings.
  3. Implement structured data where appropriate.
  4. Create a consistent review generation process.
  5. Respond thoughtfully to every review.
  6. Seek legitimate local press, partnerships, and mentions.
  7. Ensure information consistency across all platforms.

The goal is not to manipulate an algorithm. It’s to remove ambiguity about who you are, what you offer, and why customers trust you.

The Real Opportunity

While some see AI as a threat, it’s also an opportunity.

In the past, lower-quality competitors could rank simply because they were close to the searcher. Now, businesses that invest in depth, reputation, and authority may gain stronger positioning — and hold it longer.

AI amplifies quality differences.

Businesses that build real trust signals will benefit. Businesses relying on shortcuts may struggle.

The Game Hasn’t Changed — The Stakes Have

At its core, Google still wants to show users the most relevant, trustworthy, and useful local options.

What AI changes is how well Google can evaluate those qualities.

The fundamentals of local SEO remain intact:

  • Relevance
  • Proximity
  • Prominence

But prominence increasingly includes semantic depth, authority, and brand strength.

For local businesses, the message is clear: visibility now favors clarity, consistency, and credibility.

The map pack isn’t disappearing. It’s becoming more selective.

And for businesses willing to build real authority, that selectivity can be a competitive advantage rather than a threat.

07Jul

AI and Your Website: Six llms.txt Plugins Reviewed

As I said in my previous post (Is AI Killing SEO), traditional search is not going away any time soon. However, search traffic is plateauing, and click-through rates for informational searches are falling, as AI systems like ChatGPT and Gemini deliver direct answers to users’ questions. For content websites that depend on traffic, it’s important to be cited, and linked to, and AI search results.

Fortunately, as I’ve previously explained, a good SEO strategy will also help get you listed in those AI snippets.

A New Standard: the llms.txt File

There’s also a new and important tool in the AI landscape: the llms.txt file. This emerging standard gives website owners more control over how AI systems interact with their content: what they can access, what they can’t, and how they should interpret it. While not yet universally adopted, major AI platforms are beginning to support it, and adoption is expected to grow.

If you’re running a WordPress website, there are already several plugins to automatically generate an llms.txt file. I’ve reviewed the current batch of plugins, and found a winner.

Llms.txt Plugins Compared

Website LLMs.txt / version 7.1.4

Author: Ryan Howard

This is a minimal plugin. The Settings allow the user to select the type of content that appears in the llms.txt file (pages / posts), and also specify the frequency of updating the file (daily / weekly / immediate).

In addition to page URLs, the llms.txt file generated includes publication data and text. There is no way to control the file content.

LLMs.txt Generator / version 1.0.2

Author: AEOmatic (Pedro Ladeira)

I tested the free version of this plugin. It indexes pages (not posts or other content), and it has no other settings. The llms.txt file generated includes URLs, and also default access rules:

# Custom Rules
Training: allowed
Scraping: limited
APIAccess: restricted

There is a paid version of the plugin, for $49, which I did not test. It claims to add the following features:

  • Include Posts in your llms.txt file
  • Daily and Weekly automatic updates
  • Custom taxonomies support
  • Custom generation rules
  • Full Multisite support

LLMs Text Generator for WordPress / version 1.0.5

Author: Fulgid

A new version of this plugin was released the day after my initial tests, so I retested it. The description below is based on version 1.0.5.

The setting allow you to specify the type of content to index (posts / pages / media), and also provide a header block to describe your site in the llms.txt file.

In addition to URLs, the file contains publication data and and a text excerpt.

The plugin can also generate another file: llms-full.txt, which contains the full text of all pages. Although this is a part of the evolving standard, I don’t believe it adds value in this case, since AI engines can simple scan the pages themselves.

LLMs.txt for WP / version 1.0

Author: Balázs Piller

The plugin setting allow the user to specify the type of content to index (posts / pages).

There is also a setting to generate “markdown” (.md) files corresponding to each page on the site. This is part of the proposed llms.txt standard, but I do not expect this to be widely used, since AI engines are perfectly capable of understanding standard web pages.

After choosing the settings, I was unable to find a way to actually generate the files! In its current form – in my opinion – this plugin is useless.

LLMS Full TXT Generator / version 2.0

Author: rankth

The plugin has extension settings. The user can specify the type of content to index (posts / pages / media), and the file update frequency (daily / weekly / manual). The plugin can generate llms.txt and llms-full.txt files.

It’s also possible to use the robots.txt file to restrict site crawling, or to specific a list of URLs to include or exclude. Unfortunately, the URL exclusion feature did not work; URLs I listed still appeared in the llms.txt file.

The llms.txt file also includes default access rules (which cannot be changed in the settings):

# Custom Rules
Training: allowed
Scraping: limited
APIAccess: restricted

Despite the fact that the exclusion feature did not work, this is a powerful plugin with many options.

Recommended with reservations.

Advanced LLMs.txt Generator / version 1.7

Author: Taha Büyüktaş

This plugin has a very clean and attractive user interface. Unfortunately, all of the text on the Settings page is in Turkish! However, I was able to use Google Translate (built in to the Chrome browser) to understand the settings.

Like LLMS Full TXT Generator, this plugin allows the user to The user can specify the type of content to index (posts / pages / media), and the file update frequency (daily / weekly / manual). The plugin generates llms.txt, but not llms-full.txt files.

The plugin allows the user to specify access rules. This is important if you do not want AI bots to index your site.

Like LLMS Full TXT Generator, this plugin allows the user to specific a list of URLs to include or exclude, but as with the previous plugin, the exclusion feature does not work.

Recommended with reservations.

Summary of llms.txt Generators

Plugin Version Author llms.txt llms-full.txt Frequency Content Types Crawl Permissions Disable URLs
Website LLMs.txt 7.1.4 Ryan Howard X X X
LLMs.txt Generator / AEOmatic 1.0.2 Pedro Ladeira X
LLMs Text Generator for WordPress 1.0.4 Fulgid X
LLMs.txt for WP 1 Balázs Piller X X X
LLMS Full TXT Generator 2 rankth X X X X X X
Advanced LLMs.txt Generator 1.7 Taha Büyüktaş X X X X X X

Conclusions

The llms.txt standard and the plugins that support it are still evolving.

At this point, I think it is worthwhile to have an llm.txt file on your site, and I recommend both Advanced LLMs.txt Generator and LLMS Full TXT Generator for WordPress users.

15Apr

Is AI Killing SEO?

Is AI Killing SEO? Image of a robot "fist-bumping" a man wearing a tie.

In digital marketing circles, there’s a growing buzz – some might call it panic – that AI engines are replacing traditional search engines. Why click through websites when AI can give you a direct answer?

But let’s pump the brakes.

Yes, AI tools like ChatGPT and Google’s Gemini are increasingly being used for quick answers, especially for informational or how-to queries. But here’s the kicker: organic search traffic is still growing. People are still turning to search engines—especially when looking for products, services, or local businesses.

So no, SEO isn’t dying. But it is evolving.

AI is becoming a major information gateway, and it’s smart to start preparing your content for this shift. If you want your content to be cited or surfaced by AI tools, there are a few key strategies that make all the difference:

(1) Go Long with Content

AI models thrive on depth. They favor long-form, comprehensive content – ideally 1,000 to 2,000 words. (Beyond 2,000 word, returns diminish.)

(2) Make It Clear and Readable

Structure matters. Use headers, bullets, and short paragraphs. Tools like Hemingway and Grammarly, or Yoast SEO, can help you hit the right readability level.

(3) Focus on Niche Topics

“Long-tail keywords” aren’t just an SEO hack; they’re AI gold. The more specific your content, the better your chances of being cited by AI for answering nuanced or lesser-known queries.

One of my websites, the California Mission Guide, was cited by Google's AI Overview,
One of my websites, the California Mission Guide, was cited by Google’s AI Overview for a niche question.

(4) Establish Authority

Experience, Expertise, Authority, and Trustworthiness (EEAT). These aren’t just Google buzzwords! They’re becoming essential for AI recognition. Back your claims with credible sources, include author bios, and highlight real-world experience.

Bottom Line:

AI isn’t killing SEO, but it’s changing the game. The core principles of great SEO content still apply, but now there’s a new player on the field. Marketers who adapt early, by crafting high-quality, niche, and authoritative content, will be the ones who thrive in both worlds.

16Nov

The Value of Evergreen Content

telemarks-guide-to-dogsContent-based websites and products are rarely glamorous, and are often disparaged by investors and developers. But there is a huge, overlooked advantage of content-based products…  they retain their value for a long, long time.

I was a co-developer of Telemark’s Guide to Dogs, which was released on CD-ROM in 1995. The product was originally self-published.

The disk contained all sorts of information about dog activities, profiles and photos of 168 AKC-recognized dog breeds, and a “dog breed selector” that helped users to find the ideal dog breeds based on their lifestyles and preferences.  It also included short videos of several dog activities, including police dog training and a Guide Dogs for the Blind “graduation” ceremony.

Telemark's Guide to Dogs - Version 2.0The content was later licensed to The Learning Company, where it sold over 45,000 copies and won a National Educational Media Network Silver Apple Award in 1997.

When the CD-ROM market began to die, the dog breed selector and profiles were licensed to the IAMs pet food company for use on their website, iams.com. In addition to paying license fees for the product, IAMs paid for regular updates, as the AKC added new breeds and modified breed standards. (In 2017, IAMs changed their branding strategy, and did not renew the license.)

In 2022, the Dog Breed Selector, Dog Breed Comparer, and over 200 dog profiles and photos were sold to Sugar Sweet Georgia Puppies, and are now hosted on their website, georgiapuppiesfromheaven.com. The extensive content continues to have high SEO value.

In short, an information content product has generated revenue for almost 30 years, and still has value. There are very, very few technology products and businesses that can make the same claim.

25Jun

The Web is No Longer a Level Playing Field

see-sawIn the early days of the World Wide Web, it provided a level playing field for anyone with something to say. If a website published good content, it could rise quickly through the ranks of search engine results, and get placement in major Internet directories. In short order, a small website could become a widely-recognized resource.

Now, in 2018. this is no longer the case.

I am currently experiencing this firsthand with a project I am working on.

I have a new website, EDtreatment.info, which provides information, resources, help and support for erectile dysfunction sufferers and their partners. There is already a great deal of information about ED on the web, but we are trying to do two things that other sites are not:

  1. We tailor our content specifically to the needs of the ED community. We participate in many online forums, where we see the questions that people are asking. We sponsor ongoing survey research to understand how ED affects men and their partners. We use this information to prioritize our content development.
  2. We provide links to clinical research to support and justify the information we provide.

Despite the fact that my site provides very useful, relevant, and accurate content, I am struggling to gain visibility and traffic.

The purpose of this article is to explain why.

The Rise of Megasites

Search engines give better placement to sites with a high Internet presence – that is, with many high-quality inbound links. (In the past, we could estimate this Internet presence with a measure called Google PageRank. Google has stopped making this metric public, so we now use metrics such as the Moz Domain Authority, or AHREFS Domain Rating.)

Large websites have lots of inbound links, and therefore a high rank.  The sites pass the power of their rank on to every page on the site… so as sites become larger, the number of inbound links increases, and each page on the site becomes harder to compete with.

In the case of my own site, EDtreatment.info, I have many inbound links… but I can’t begin to compete with the number of links to megasites such as WebMD.com. And therefore, it is very difficult for pages on my site to compete with pages on the WebMD site.

Authoritative Links

Search engines give preference to websites with “authoritative” links pointing to them. These authoritative links typically come from news sites, government sites, and educational sites.

Large sites tend to accumulate these high authority links. If someone at a major publication is writing an article on erectile dysfunction, they are likely to include a link to information on WebMD… and much less likely to include a link to a small website such as mine.

Thus, sites that already have a strong web presence grow stronger and stronger, and it is virtually impossible for small sites to catch up.

The Consequences

If the web is no longer a level playing field, a small number of megasites will come to dominate broad categories of web searches… and it will become virtually impossible for new sites to establish themselves by providing better content or services.

09May

Google “Mobile-Geddon” Impact Appears to be Small

mobile-friendlyRecently I’ve gotten a lot of questions about the Google mobile algorithm change, which was rolled out on April 21.

The change – which the press dubbed “mobile-geddon” – affects sites that are not mobile-friendly. If a web page is not designed to display properly on a mobile phone, it will be penalized and will show up lower in search results on mobile phones. The change will not affect search results on desktop computers, laptops, or tablets.

(You can find out if a page is mobile-friendly using the free Mobile Test Tool provided by Google.)

I wanted to assess the actual impact of the April 21 changes on real sites. Rather than looking at placement in search results, I decided to focus on traffic.  I made a few assumptions:

  • If a site is not mobile-friendly, it should now appear lower in search results on mobile devices, which should result in fewer clicks, and less traffic.
  • If a site is mobile-friendly, it should now appear higher in search results on mobile devices (because “unfriendly” sites have slipped lower), which should result in more clicks, and more traffic.
  • Placement in search results (and hence traffic) on non-mobile devices should not be affected.

Given these assumptions, we can assess the results of the algorithm changes by measuring the percentage of traffic that comes from mobile searches, both before and after the algorithm change.  Mobile-friendly sites should see more mobile traffic, and unfriendly sites should see less.  I used percentage rather than absolute numbers, because traffic can vary for many reasons unrelated to the Google change.

Using Google Analytics, I examined eight sites that are mobile-friendly, and nine that are not mobile-friendly.  I gathered data for the period from April 11-17, before the change, and for May 2-8, after the change.  Both periods begin on a Saturday and run through Friday; there are no significant US holidays contained in either range.

The aggregate results are shown below.

Non-Friendly

Prior to the change, 30.6% of site traffic came from mobile devices.  After the change, 28.8% of site traffic came from mobile devices.  As expected, the percentage of traffic from mobile devices fell.  The decrease amounted to 1.3% of overall site traffic.

Mobile-Friendly

Prior to the change, 26.7% of site traffic came from mobile devices. After the change, 26.3% of site traffic came from mobile devices. Surprisingly, the percentage of traffic from mobile devices fell!  The decrease amounted to 0.4% of overall site traffic.

Conclusions

The fact that the percentage of traffic from mobile devices fell, even on mobile-friendly sites, calls into question the validity of the results – despite the fact that mobile traffic on “unfriendly” sites fell more (as expected).

It may be that my sample size was too small, or that other factors I have not considered affected the results.  But one thing is clear: the affect of the April 21 algorithm change is small.  It was certainly not “mobile-geddon.”

Nevertheless, it’s important to remember that mobile use is increasing. Google and other search engines will probably favor mobile-friendly sites more over time.

Plus, once you make a site mobile-friendly, it takes some time for Google to re-crawl the site and recognize the changes (see Were You Ready for Mobile-geddon?).

If you haven’t already made your site mobile-friendly, it’s time.

 

05May

Were You Ready for Mobile-geddon?

shutterstock_114417286On April 21, Google made sweeping changes to their search engine algorithm for mobile devices. Sites that are not “mobile friendly” will appear lower in search results on mobile devices.

Google announced the change months in advance, to give site operators time to make changes. They even provided a Mobile Test Tool to determine if a page is mobile friendly.

Many sites made great efforts to be ready for the April 21 deadline… but neglected to take into account Google’s indexing cycle, which may take weeks or months to revisit a site.

One client’s site was ready for the change by April 18.  However, Google last crawled the site on April 14.  Now, two weeks after the algorithm change, Google Webmaster Tools still reports 56 pages with mobile access errors (despite that fact that all pages have been fixed).

I am gathering data on how the changes impact a number of sites, and I’ll be reporting the results once the dust settles.  However, many sites that made an honest attempt to meet Google’s widely trumpeted deadline are still going to be penalized until Google catches up.

Despite the delays, the recent changes are a clear indication of the direction search companies are going.  Sites that have made changes will benefit in the coming months. Businesses that have not upgraded their sites should do so soon.