Internet Search Remedies For Reuniting Local Communities

An Open Letter to Honorable District Judge Amit P. Mehta regarding the remedies for United States v. Google LLC

We propose industry-wide remedies which more closely resemble the equal-access principles which regulated AT&T and the regional bell companies. Today the commodity to be regulated is the inclusion or exclusion of a given individual or organization in search results, especially at the regional, local level. For instance, there is only one correct answer to the question "what plumbers are near me?" Yet Google, under the status quo, is hiding most local plumbers and instead showing the results of the highest bidders, which are mostly non-local plumbers and agencies which have invested heavily in search engine optimization. Google has not always presented data this way and there are simple changes which can restore the Internet, and Google's original mission, of making data more universally accessible and useful.

The competitive landscape for Internet search peaked in the late 1990s and 2000s when phone books offered comprehensive local directories and Internet search companies sought to match and exceed their functionality. Indeed, Google's mission of "organizing the world's information and making it universally accessible and useful" originally included actual directories, as shown below:

Figure 1, Google in early 2011.

In late 2011 a pivotal shift took place as Google abandoned its directory. Having already ascended to the #1 spot for Internet search, a financial incentive became clear: How large is the advertising revenue opportunity if all organizations are readily discoverable? For instance, if a search for "plumbers near me" results in an actual and correct list of nearby plumbers, how incentivized would the plumbers be to advertise? Alternatively, how large would the advertising revenue opportunity be if, instead of presenting "organized" data which can be sorted by proximity, A-Z, etc., what if search-result ranking was the only means of information access?

Of course Google was correct in choosing the latter route to maximize revenue, but at what cost to society?

Today, according to our study in South Whidbey, 94% of local organizations are undiscoverable on Google by keyword search.

Why it matters

Intra-community communication, the foundation of human culture,¹ is decades in decline.² In 2023, the U.S. Surgeon General declared an 'Epidemic of Loneliness and Isolation'³ and included technology as a contributing factor. For local communities, Google search can be a gateway or a gatekeeper, depending on the extent to which local inquiries produce local results.

Our Approach

By leveraging verified reference data, we measure Google's search inclusivity (of local organizations) and explore the limitations of Google's search-only approach to Information Retrieval (IR.) We conclude with recommendations for corrective actions as remedies in the United States v. Google LLC.

Highlights

• 94% of local organizations are undiscoverable on Google by categorical keyword-search.⁴
• Limiting IR to search, without categories or sorting, is a gatekeeping tactic. It limits information-access while increasing the demand for advertising
• According to Google, Artificial Intelligence (AI) represents the biggest breakthrough in search in 25 years.⁵ As of March, 2025, open-source AI is competitive⁶ with Google, ushering in new-options for more democratized and less commercialized solutions.

Arriving At The Reference Data

Local organizations of all types serve as productive centers for cultural and economic exchange, hence we started with inclusivity as a first-principle: All South Whidbey organizations and individuals doing business in their own name are included. Specifically, the reference data includes all of the following:

• Organizations with an active status with the Washington Secretary of State, located in South Whidbey, with an active website.
• Organizations with a non-local headquarters which have at least one full time employee in South Whidbey who primarily serve the South Whidbey area.
• Individuals doing business in their own name, who live in South Whidbey, and have an active website.
• All have been contacted by South Whidbey Online, A Social Purpose Corporation, to be made aware of this initiative to 'make our online community, our own.' All have been given the opportunity to utilize our directory at https://southwhidbey.online and/or to opt-out. Feedback from the community is available as appendix Figure 2.
• All are verified to be included in Google's network (and can be found when the exact business name and/or exact business name in combination with the city and state of their headquarters is searched.)
• Many sources were used to arrive at the reference data. A summary of the top 20 sources is available in appendix Figure 2.

621 local organizations were identified and organized according to three levels of detail for categorical search-term purposes:

1. Main categories (10)
2. Subcategories (31)
3. Detailed subcategories 387 (where applicable)

For instance, a pet-care service and graphic-design consultancy may have keyword-search phrases per the above as:

(Tricia's Pet Care)	(CK Graphic Design)

1. (Main Categories) Animal and Pet Care Services | Services and Retail
2. (Subcategories) Pet Sitting | Web-Design
3. (Subcategory Details) Overnight Pet Resort | Full Service Creative Agency

By arriving at a reference-data set, we can measure Google's inclusion and/or exclusion of local organizations.

From Gateway to Gatekeeper

Gatekeeping, in this context, refers to limiting information access in order to charge a premium for increased visibility (advertising). We identify two ways Google limits access to local organizations: first, by underrepresenting local organizations in search-results, and second, by limiting information retrieval tools to advertisement-based ranking schemes.

For the first causal mechanism, the underrepresentation of local organizations in search-results, we analyzed 930 search results from 31 carefully selected key-word searches. Each search-phrase was derived from the verified reference data to ensure at least 10 local results, on average, were possible for each query. "Near me" was appended to each search with location services activated at the approximate center of the South Whidbey area. We then classified all 930 results, by URL, as either local or non-local. 58 of the 930 results, or 6%, were local, and the rest were non-local. We believe this lack of local information is having a highly divisive impact on our local community and likely on most local communities Nationwide.

For the second causal mechanism, we zoom-out to examine Google's approach to Information Retrieval (IR). Studies in IR demonstrate how search, when combined with categories and sorting, represent the most effective means of information retrieval.⁷ Indeed, we only need to look to the public library's catalog search, ebay.com or amazon.com to see all three of these IR tools being effectively utilized. Yet Google relies on search exclusively. We submit search alone limits information access and effectively heightens the importance of result-ranking based advertising. Consider the following example to illustrate:

South Whidbey is home to 110 General Contractors. Imagine all 110 increased their Google advertising budget to $5,000 per month. Yet Google only presents data in a "top 10" type ranking format, making it impossible for 110 General Contractors to be presented on a level playing field, such as could be readily accomplished with A-Z sorting or proximity sorting.

Proposed Remedies

When Google began, it had to compete with phone books which, at the time, provided near 100% local, accurate, and inclusive content. Not surprisingly, Google was competitive at the time and even offered categories and location-specific data as can be seen via their search page from 2011 (Figure 1.) Now is the time to reinstate these fundamental community resources by putting local communities back into focus:

For any Nationwide web-search portal (NAICS 519130) we propose:

1. Location-based searches must prioritize local-entities on the first-page of the search results. This can be accomplished: a. Internally i. By correcting the algorithm and/or AI to prioritize local results for local searches, and, ii. By allowing organizations to update their data such as is available today via Google Business Profiles for Google Maps. b. Externally i. The algorithm/AI changes can be implemented by a third party. ii. The opt-in, opt-out (ie., maintenance) can be delegated to local (city or county) Chambers of Commerce, and/or to local 501(c)3 organizations which include community-building in their mission.

2. Category-based searches which include locally-available keywords (ie., "plumber," "veterinarian," "yoga studio," as applicable,) shall include a checkbox-option to view locally available organizations as described above.

3. When locality is triggered by either of the two above, sorting by proximity or A-Z must also be available.

Conclusion

While the Internet and AI dwarf the telephony era by many measures, the empowerment of local communities continues to have a legacy "land-line" resonance. Unlike geographically anonymous URLs, phone numbers and street addresses were imbued with the local information which is lost in today's FAANG dominated Internet. We propose remedies to restore this community fabric.

Google's version of the Internet is less like the public network it was built upon and more like a pay-to-play gatekeeping service in which much of the data is hidden in order to increase the demand for advertising. Google is the only winner in this scheme.

Appendix:

Figure 1. Feedback from community members.

Figure 2. Sources for the reference data. The largest source is the Washington Secretary of State, which listed 2,009 active organizations, 580 of which were found to have active websites on Google. The total organizations found across all sources amounted to 621.

Source (A-Z)	No. of Entities supplemental to the WA SOS
Chamber of Commerce - Clinton	1 net-new
Chamber of Commerce - Freeland	zero net-new
Chamber of Commerce - Greenbank	zero net-new
Chamber of Commerce - Langley	3 net-new
Drewslist (notable mention)	Copyrighted newsletter, content excluded
Facebook search by business categories	zero net-new
Google search by business name	2,009 searches using WA SOS data, 580 verified
Google search by categories	Zero net-new
Nextdoor	zero net-new
Physical bulletin boards at Grocery stores	5 net-new
Port Of South Whidbey	zero net-new
Tiktok, Instagram, X, Reddit	zero net-new
Washington Secretary of State (WA SOS)	580 net new
Whidbey and Camano Islands Tourism	3 net-new
Whidbey Community Foundation	5 net-new
Whidbey Island Arts Council	7 net-new
Whidbey Island Grown Cooperative	5 net-new
Whidbey Telecom Phone Book	13 net-new

Figure 3: Thirty one Searches. The first 30 results of each (930 total) were analyzed for inclusivity. 94% of results were non-local. Complete details are available via footnote no. 4.

"Veterinarian near me", "dog training near me", "pet care near me", "fine art studio near me", "church near me", "general contractor near me", "plumber near me", "electrician near me", "remodel near me", "homeschooling near me", "gym near me", "martial arts near me", "farm near me", "naturopathic care near me", "counseling near me", "coaching near me", "dentist near me", "hair salon near me", "massage near me", "pilates near me", "marketing services near me", "web design near me", "jewelry near me", "tax services near me", "B2B consulting near me", "landscaping services near me", "event spaces near me", "lodging near me", "house painting near me", "septic services near me", "home construction near me"

Footnotes

1. Pagel, M., 2017. What is human language, when did it evolve and why should we care?. BMC Biol 15, 64. Link: https://bmcbiol.biomedcentral.com/articles/10.1186/s12915-017-0405-3 ↩

2. Putnam, R. D., 2000. Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster. Link: https://muse.jhu.edu/article/16643 ↩

3. Julianne Holt-Lunstad, 2023. Our Epidemic of Loneliness and Isolation: The U.S. Surgeon General's Advisory on the Healing Effects of Social Connection and Community, Office Of The Surgeon General. Link: https://www.hhs.gov/sites/default/files/surgeon-general-social-connection-advisory.pdf ↩

4. Pitney, F., 2025. Search results compared to reference data https://southwhidbey.online/search_comparison_results.json, Search results summary https://southwhidbey.online/search_comparison_summary.json ↩

5. Ried, E.,2023. Supercharging Search With Generative AI. Link: https://blog.google/products/search/generative-ai-search/ ↩

6. Wei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas Angelopoulos, Tianle Li, Dacheng Li, Hao Zhang, Banghua Zhu, Michael Jordan, Joseph E. Gonzalez, Ion Stoica, 2024. Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference. arXiv.Link: https://lmarena.ai/?leaderboard ↩

7. Ben-Yitzhak, O., Golani, M., Ravid, G., & Shmueli, G.,2008. An exploratory approach for analyzing faceted search. ↩