Surviving the Search Apocalypse: How Nonprofits Can Sustain Free Website Traffic in an AI-Driven World

Slide 1

Slide 1 text

Surviving the Search Apocalypse: How Nonproﬁts Can Sustain Free Website Trafﬁc in an AI-Driven World February 26, 2024

Slide 2

Slide 2 text

Katherine Watier Ong WO Strategies LLC ● Built ﬁrst website in 1994. ● Lost sites due to too much trafﬁc (EWG.org for 4 days, but still 1.2 million visits up from 300K) ● Helped NYer’s get registered on the NY Health Exchange ● Pivoted search query volume toward electronic health records for HealthIT.gov, and ● Helped Cancer.gov grow from 1.5 million to nearly 5 million searches/mo.

Slide 3

Slide 3 text

Brands Katherine Watier Ong has worked with

Slide 4

Slide 4 text

Agenda ● Understand the Shift in Search Behavior & how much organic discovery on Google is left ● Understand the Importance of a Social Media Strategy in the decision making process ● Develop an Optimization Strategy for AI Overviews ● Know how to Rank in LLM Search ● Future-Proof Your Nonproﬁt’s Organic Trafﬁc

Slide 5

Slide 5 text

The Current Search Apocalypse

Slide 6

Slide 6 text

Almost ⅔ of the Queries Result in NO Clicks

Slide 7

Slide 7 text

44.19% of searchers already know what they want. Of the 40% that is left….

Slide 8

Slide 8 text

Of the 40% that is left… ⅓ of the search queries are navigational

Slide 9

Slide 9 text

The Leftovers/Long Tail is *Tiny* “The Long Tail (<11 searches/month) is only 3.6% of demand. Big brands and topics are winning. Google has become a place people go *after* they discover a need rather than a demand-creation or even demand-nudging platform.” Rand Fishkin, CEO Sparktoro

Slide 10

Slide 10 text

How are the AI Overviews Impacting Search?

Slide 11

Slide 11 text

Google’s AI Overviews

Slide 12

Slide 12 text

Bing’s AI Overviews

Slide 13

Slide 13 text

AI Overviews are Increasing (and take up all of the above the fold)

Slide 14

Slide 14 text

Current Organic Search Clickthrough Rate See the full study here

Slide 15

Slide 15 text

Current AI Search Impact These are visits, not searches, but still…

Slide 16

Slide 16 text

Estimated Growth of AI Search

Slide 17

Slide 17 text

People are researching with social media, not search

Slide 18

Slide 18 text

Discovery: Only 34% start with search, 32% start with social media April 2024 data

Slide 19

Slide 19 text

Gen Z and Millennials use Social for Discovery Source Gen Z Millennials

Slide 20

Slide 20 text

Social is ranking in Search “Perspectives” First person Social profiles ranking: ● Study.com ● Quora.com ● YouTube ● Facebook

Slide 21

Slide 21 text

Every platform rewards Zero Click content

Slide 22

Slide 22 text

How much of the organic trafﬁc can I track?

Slide 23

Slide 23 text

Organic Search Tracking Issues - People Issues ● People decline cookies and remove UTMs ● “Not set” are people who keep your website open in a tab. ● 67% of US block cookies or the browser does (Safari, Firefox, Brave)

Slide 24

Slide 24 text

Organic Search Tracking Issues - GA4 ● You can not track absolutes ● Impossible to know how many users you have unless you have a login ● You can’t measure the bounce rate ● AI Search, many mobile visits and others are lumped into “direct” ● You should be tracking percentages and sharing trends

Slide 25

Slide 25 text

What is At Stake for Nonproﬁts?

Slide 26

Slide 26 text

This is how many of you rely on free traffic Source

Slide 27

Slide 27 text

This is what happens if you stop/do NO SEO Will Critchlow found a company who made this decision. This is how much traffic you lose: Year 4: -13% organic traffic Year 5: -20% organic traffic Year 6: -30% organic traffic

Slide 28

Slide 28 text

This is what happens if you stop social media

Slide 29

Slide 29 text

2025 Vital Strategy Changes

Slide 30

Slide 30 text

1. Be where your audience is online

Slide 31

Slide 31 text

No content

Slide 32

Slide 32 text

Social Media Research PRIOR to Brand Search Sparktoro to ﬁnd where they are:

Slide 33

Slide 33 text

2. Coordinate search + social

Slide 34

Slide 34 text

Social Media Ranking in Search Ahref’s SERP features:

Slide 35

Slide 35 text

Steps 1. What platforms are showing for your target keywords? 2. Create content optimized for that topic, or try to get your brand mentioned by the inﬂuencer.

Slide 36

Slide 36 text

3. Keep your organic visitors

Slide 37

Slide 37 text

Convert the User! ● Make sure you’re capturing their email or phone # ● Free Microsoft Clarity to test your conversions ● Google is watching your users in the Chrome browser (thank you Google algo leak and DOJ trial) - so you should too!

Slide 38

Slide 38 text

Conversion Rate Optimization Tips ● Google’s UX Playbook for news and content. ● Track your key events in GA4 ● Install Microsoft Clarity (Free) ● Use LLMs to test your UX. ● Use this sheet to track your improvement ideas

Slide 39

Slide 39 text

4. Figure out how much of your audience is using LLM Search

Slide 40

Slide 40 text

Is YOUR Audience using LLM search? ● You can look by website or topic. ● For “science” - not only is OpenAI in the high-afﬁnity site list, but it’s increasing in popularity.

Slide 41

Slide 41 text

GA4 - LLM Referrals ● One off exploration report -> ● Create a custom channel report: ○ Video walk-through & explanations from Kickpoint here ● Looker Studio report lets you compare your LLM trafﬁc against organic trafﬁc.

Slide 42

Slide 42 text

5. Create an AI Summaries Strategy

Slide 43

Slide 43 text

First: NLP? LLM? RAG?

Slide 44

Slide 44 text

Natural Language Processing (NLP) ● Allows computers to understand the meaning of your queries or sentence, and/or spotting a name (entity, verb, object, subject, etc) in a chunk of text that has additional meaning. ● Allows search engines to determine the salience of a chunk of text to a subject. Google started using it in 2017. ● It can spot sentiment, categorize the purpose of the content. ● Used in: Siri, Alexa, Google Assistant, Google Translate, Gmail, Google Search (BERT, MUM language models used in the algorithm) ● Google uses Natural Language Generation to create text & summaries in search.

Slide 45

Slide 45 text

What are Large Language Models (LLMs)? ● Large Language Models (LLMs) learn from mountains of text, nearly all of the internet (MadLibs style) to get good at next-word predictions. ● With some input text (from you the user), the model is able to create a probability distribution of the most probable next set of words & randomly pick from those to come up with a conﬁdent, human-sounding response.

Slide 46

Slide 46 text

Retrieval Augmented Generation (RAG) ● It is an architecture that augments the LLM by connecting it to an information retrieval system. ● Connecting LLMs to external sources of information (Google Drive, Google search) to get up to date information - Copilot today Source Google Drive & Microsoft ﬁles being uploaded into ChatGPT 4 - soon.

Slide 47

Slide 47 text

How to Rank in AI Overview

Slide 48

Slide 48 text

1. Rank in the top 20 “Brands ranking on page 1 of Google showed a strong correlation (~0.65) with LLM mentions. Bing rankings also mattered but less so (~0.5–0.6).”

Slide 49

Slide 49 text

2. Cosine Similarity Measure the cosine similarity between a topic and passage on any URL by using AI Overviews Visualizer. A cosine similarity of 1 means identical, and zero means not related.

Slide 50

Slide 50 text

How is this measured? 1. AI converts the word and the passage to vector embeddings. 2. Then it calculates the distance between the two vectors to determine if the keyword and topic are semantically related.

Slide 51

Slide 51 text

3. Load Quickly, no JavaScript ● The URL should load quickly and not use JavaScript to render the content. ● LLM crawlers cannot render JS (except for Googlebot, Applebot & Bingbot) ● Essential text should be in HTML, and server speed needs to be under 500 ms.

Slide 52

Slide 52 text

6. Use this GPT - can you rank?

Slide 53

Slide 53 text

Create a Credible Fragment ● Your fragment is credible and has citations, quotes, and stats. ● The citations should increase the content's trustworthiness and relevance, and no more than 5-6 per fragment

Slide 54

Slide 54 text

7. Make helpful content ● People-First ● driven by content quality. Experience, Expertise, Authority and Trust. ● Demonstrate hidden gems content that show personal ﬁrst-hand experience with the content.

Slide 55

Slide 55 text

Helpful Content Update - Sept 2023 ● It is harder to get indexed due to content quality. ● A sitewide signal, if the machine learning ﬁnds a high % of content is “unhelpful”, the site will be impacted.

Slide 56

Slide 56 text

Making Helpful Content Guide around how to provide helpful content lists questions to ask. ● Quality content ● Written from a palace of expertise, with author info clearly displayed (bio) ● Good UX ● Content created for people with a purpose that will help them with their goal. WHY are you creating the content. ● Check with this custom GPT

Slide 57

Slide 57 text

Barnacle SEO Options If you are not ranking in top 20: Develop a strategy for “pitching” content to sites like Quora, Reddit, and Wikipedia that have answers ranking.

Slide 58

Slide 58 text

Bing AI Overview Optimization ● Create a Bing Webmaster Tools account ○ Check for content quality issues inside ○ Grab and install Microsoft Clarity ● Use IndexNow and Bing’s instructions ● Base your content on a basic HTML template (nested H1, H2 etc) ● Other Bing optimization tips

Slide 59

Slide 59 text

Create content that covers searcher’s journey Ready for when Google allows the searcher to ask the *next* question (like they did in the beta) Map your customer’s journey and create questions for the *next* question and link to it from your *ﬁrst* question. (use the “People Also Searched for”, or AlsoAsked.com)

Slide 60

Slide 60 text

Measuring your Impact

Slide 61

Slide 61 text

Google Analytics 4 ● Use Google Analytics (or whatever you use) to report on pages referred from Google containing #:~:text=. ● This format highlights speciﬁc parts of the page in SERPs for featured snippets and AIO links.

Slide 62

Slide 62 text

Google Search Console and Bing Webmaster Tools ● AI Summary visits are not segmented - instead they are rolled into the other metrics.

Slide 63

Slide 63 text

Creating a Testing and Optimization Framework

Slide 64

Slide 64 text

Building a Testing Plan ● AI Overviews checklist to track your testing plan. ● Measure impact on trafﬁc (now and based on predictions) ● A/B test your AI-optimized content ● Measuring visibility on AI platforms

Slide 65

Slide 65 text

Developing an LLM Search Strategy

Slide 66

Slide 66 text

Create an LLM Ranking Plan It’s not about “position” It’s about getting included in answers, summaries, and conversations

Slide 67

Slide 67 text

How Ranking works in an LLM Your brand mentioned in the LLM’s training data next topics associated with your brand. Key ranking factors include: ● Allow LLM crawling ● Content Quality and Authority ● Contextual Relevance - Match to Search Intent ● Concise, Structured Responses ● Internal Linking and Semantic Relationships (Topical Authority)

Slide 68

Slide 68 text

How to Structure Content to Rank in an LLM 1. Featured Snippet-Style Responses 2. Hierarchical Content Structure 3. FAQs Section 4. Semantic Optimization Add related keywords, entities, and topics SEObility.net

Slide 69

Slide 69 text

Leveraging Topic Clustering Why topic clusters matter: ● AI assesses how well your site covers a broad topic comprehensively. ● You can cluster pages around key missions or causes (e.g., climate change, hunger relief). Google algo leak & topical authority

Slide 70

Slide 70 text

Create Strategic Internal Links ● Guide AI bots to related content to signal authority. ● Link posts, case studies, donation pages, and resources around the same topic. ● AI Prompt (for a chat with RAG): Provide a list of semantically-related URLs from [domain] that are related to [provide URL] in a table format. The ﬁnal output should be a list of URLs taken using only from [domain].

Slide 71

Slide 71 text

Conversational Keyword Optimization for AI 1. Question-Based Keywords 2. Long-Tail, Intent-Focused Keywords Natural, full-length phrases that align with search intent 3. Semantic Keywords Integrate synonyms, related phrases, and entities

Slide 72

Slide 72 text

Building Content that AI Recommends ● Create evergreen content that addresses common topics. ● Optimize for context-based triggers, such as stories of impact, donor success stories, or how-tos. ● Maintain a consistent content refresh strategy to signal relevance. ● Leverage Videos, PDFs, etc.

Slide 73

Slide 73 text

The Role of Backlinks and Mentions in LLM Ranking Authority through mentions and citations plays a signiﬁcant role. ● Publish high-value resources (reports, studies) that others will reference and link to. ● Ensure citations in industry databases, research hubs, or news outlets. ● Collaborate with partners for mentions.

Slide 74

Slide 74 text

Sites to Target 1. Database-style websites 2. Large editorial sites 3. Sources for the Google Knowledge Graph 4. Or you can prompt the AI chat to surface sites that make the most sense for your topic.

Slide 75

Slide 75 text

AI-Specific Technical Optimization Limited JavaScript ● Except for Google, LLMs are not rendering JavaScript. Page Load Speed: ● AI models penalize slow-loading sites during ranking selection. ● Many have tight timeouts (1-5 seconds). Long content might be truncated after the timeout. AI-Crawling Considerations: ● Optimize (don’t block in) robots.txt, XML sitemaps and interlinking for better AI bot discovery of new content.

Slide 76

Slide 76 text

AI Search Engine Test - will you appear? https://andisearch.com/

Slide 77

Slide 77 text

Your new Robots.txt # Allow AI search and agent use User-agent: OAI-SearchBot User-agent: ChatGPT-User User-agent: PerplexityBot User-agent: FirecrawlAgent User-agent: AndiBot User-agent: ExaBot User-agent: PhindBot User-agent: YouBot Allow: / # Disallow AI training data collection User-agent: GPTBot User-agent: CCBot User-agent: Google-Extended Disallow: / # Allow traditional search indexing User-agent: Googlebot User-agent: Bingbot Allow: / # Disallow access to admin areas for all bots User-agent: * Disallow: /admin/ Disallow: /internal/ Sitemap: https://www.example.com/sitemap.xml For an updated list of AI agents, check Dark Visitors.

Slide 78

Slide 78 text

Brand Optimization ● Claim your Knowledge Panel and update it ● If not panel, put in place sameas markup on your about page and link to your other off site mentions ● Manage your Wikidata and Wikipedia mentions ● Interlink your onsite entities in a semantic way.

Slide 79

Slide 79 text

Your brand + the LLM You need to check your brand entities across all of the LLM chats. Prompts to try: ● What is X brand known for? ● What topics is [brand] authoritative on? ● What brand is high authority in the [topic] or [service]? ● Who is known for X topic targeted at X people? A more elaborate process to ﬁnd the questions that people ask that are associated with your brand.

Slide 80

Slide 80 text

Analyze the Results and Look at Competitors ● Competitors Appearance ○ How are they are mentioned and what URLs? ○ Can you be mentioned there for those topics? ● How well does the LLM understand your brand?

Slide 81

Slide 81 text

Not all LLMs are the same - some have RAG Generative AI Platforms LLM Knowledge Cut-off Date ChatGPT ChatGPT-4o October 2023 ChatGPT-4 April 2023 Gemini Gemini Pro Nov 2023 Perplexity AI Perplexity April 2023 Claude Claude 3 April 2024

Slide 82

Slide 82 text

Strategy for LLM without Rag & no brand mentions 1. Place your brand (or website) with your target keywords/topics on authoritative sites. 2. Provide the LLM feedback on its responses (including thumbs up and thumbs down) 3. You will have to wait until they update their training dataset to update that LLM’s information about your brand.

Slide 83

Slide 83 text

LLM with RAG and Brand Mention Goal: Inject Information into the training set Training Data Sources that we know about: YouTube (subtitles)

Slide 84

Slide 84 text

Steps to an LLM Ranking Strategy 1. Find where the LLM search is missing information about your brand. 2. Add that information to a URL that is used by ChatGPT for citations. 3. Include reviews from real users and quote experts in your copy. 4. Easy to understand writing that uses simple language. 5. Format into lists for key aspects of the topic you are covering

Slide 85

Slide 85 text

Steps to an LLM ranking strategy 6. Increase the number of semantically related words. 7. Summarize your user-generated content. 8. Wait for re-indexing and monitor improvements. 9. Keep testing. 10. Keep doing classic SEO since AI Chatbots still lean heavily on their results to ground answers.

Slide 86

Slide 86 text

An SEO’s Chat with LLM for Ranking Ideas ● Offer Unique, Proprietary Insights ● Specialize in Niche Queries: ● Build Interactive Tools ● Inﬂuencers or Community Partnerships ● Multimedia Content ● Case Studies and Stories ● Deep and Evergreen Content ● Build a Recognizable Brand ● Get Content Syndicated to get more mentions ● Use AI APIs and ask AI. ● Create Content within the AI Chat (Custom Branded GPTs) ● Partner with LLMs Source

Slide 87

Slide 87 text

Measuring your mentions in LLM search

Slide 88

Slide 88 text

Slide 89

Slide 89 text

Measure LLM Rankings ● Knowatoa ● OtterlyAI ● BrandRank.ai ● SpyGPT ● RankScale.ai ● Profound (enterprise)

Slide 90

Slide 90 text

Organization Adaptability ● Join masterminds ● Share updates via brown bag sessions ● Take advantage of free AI training

Slide 91

Slide 91 text

AI Chat caveats for your use of AI

Slide 92

Slide 92 text

Hallucinations ● Making stuff up = “hallucinations.” ● 3-27% of the time it’s hallucinating. ● Treat AI like you would a new-to-you intern and check every step, and you should be OK. ● You can ask your AI: ○ Do you understand? ○ Please provide references. ○ If you don’t know, say you don’t know Image generated by Microsoft Designer

Slide 93

Slide 93 text

Mirrors Your Intelligence ● ChatGPT mirrors your intelligence and education level ● You must train AI to write for your target audience’s reading level. Image created by Microsoft Designer

Slide 94

Slide 94 text

Proprietary and PII Information If you answer yes, Don’t use the data in AI chat ● Is the information conﬁdential by your business? ● Is the information you are using not freely available on the Internet? ● Are you using personal information? ● Is it ﬁnancial information? ● It is login details? ● Is it something that is secret?

Slide 95

Slide 95 text

Slide 96

Slide 96 text

Bribing/Tipping Improves Response

Slide 97

Slide 97 text

Human Review is still needed ● Ideally, AI should only be used to partially generate content — a human’s expertise is still needed. ● Review the outputs — ○ The answer ○ The references ○ The logic

Slide 98

Slide 98 text

Employee policies for AI chat use?

Slide 99

Slide 99 text

Other recommended AI use cases

Slide 100

Slide 100 text

Brand Promise, Positioning & Voice ○ Brand promise - the high-level promise we make to our customers, developers and partners. It’s the reason our brand exists. ○ Brand positioning - our clear-cut stake in the ground. It’s the space we want to occupy in people’s minds. ○ Personality/Voice - assigning human characteristics and personality traits to bring a brand to life. 100 To bring inspiration and innovation to every athlete in the world. For serious athletes, Nike gives conﬁdence that provides the perfect shoe for every sport. Rugged. Exciting. Cool. Innovative.

Slide 101

Slide 101 text

Using AI to codify your brand voice? ● Unique Brand Value (UBV) Helps cement brand + topic. ● Increase brand mentions & searches. ● Google measures your brand searches and uses it in rankings. Lower brand value sites are not able to appear for various search features. ● Consistency in your customer’s experience regardless of the touchpoint. ● Your employees and customers can articulate and effectively share with others.

Slide 102

Slide 102 text

Additional Resources

Slide 103

Slide 103 text

Other Resources Our guide on how to rank for AI Overviews Creating a social media strategy that supports Google discovery. This custom GPT: Can it rank for an AI Overview? CoPilot prompt toolkit

Slide 104

Slide 104 text

Bonus Items ● Keyword Content Mapping Sheet ● SEO Writing Guide/Content Brief per main topic ● SEO Writing Checklist ● Searcher’s Intent tutorial and checklist ● Featured Snippets writing guide and checklist ● Image Optimization Checklist ● Fraggles Writing Checklist ● Guide for improving CTR ● Optimizing the SERP Snippet Generated with Gemini Advanced

Slide 105

Slide 105 text

A SEO writer’s checklist ● Does this match intent (content type, format & copy)? ● If an AI Overview appears, does your passage match the topic? ● Have I explained why my content matters to the topic? ● Did I back my claims up? ● Is every sentence adding value? ● Have I given a next step? ● Did I review the copy based on EEAT? ● Am I encouraging the click from search?

Slide 106

Slide 106 text

KATHERINE WATIER ONG 202-930-1744 [email protected] www.linkedin.com/in/katherinewatier www.WOStrategies.com Daily SEO tips via Alexa