I've Done SEO for a Decade.. Google's 2026 Changes Are the Biggest Shift I've Ever Seen!

Google's AI Overviews now reach 2.5 billion monthly users! Click-through rates have dropped 15-46% on queries where they appear. Forum content is outranking professional publishers on opinion, comparison, and experience-based queries. Google is explicitly rewarding user-generated content. And the SEO industry is either panicking or pretending nothing's happening. Both responses are wrong.

I've been doing SEO for over a decade. Not the "I read a blog about keywords once" kind. The kind where you're white-labelling for agencies, running campaigns worth hundreds of thousands of pounds, teaching workshops at universities, presenting SEO webinars at Startup Networks events, and then coming home to build your own site and watching everything you taught people get reshaped by a Google update you didn't see coming.

I've survived every major Google algorithm change of the last ten years. Panda hammered thin content sites in the early days and I watched agencies lose half their client portfolios overnight. Penguin destroyed dodgy link building and I watched SEOs who'd built their entire strategy on private blog networks scramble to disavow thousands of links. The Helpful Content Update wiped out sites that existed purely to rank rather than to help people. Every single one of those updates felt like the end of SEO at the time. It wasn't. SEO adapted. We adapted!

But this one feels different. I've sat in rooms with agency owners who've built their entire businesses on organic search and watched them go quiet when the conversation turns to AI Overviews. Not because they're worried. Because they genuinely don't know what to tell their clients anymore. The playbook they've been selling for a decade is being rewritten by Google in real time and the new version isn't finished yet.

That's not a reason to panic. It's a reason to pay attention. Here's what's actually happening...

Ten Years of SEO and Nothing Prepared Me for This

I need to give you some context because my perspective on this is shaped by where I've been.

I started in SEO doing white-label work for digital agencies. The kind where the agency sells the client a £7,000/month SEO retainer and then pays you £4,000 to actually do the work. You never meet the client... You never get credited... You just deliver the results and the agency takes the glory... I did that for years across campaigns worth collectively millions of pounds. Retail, SaaS, professional services, e-commerce, local businesses. Everything.

I taught SEO workshops at universities. Stood in lecture halls explaining keyword research to marketing students who'd never opened Google Search Console. Ran webinars for founders at Startup Networks breaking down how to get their first page ranking. Built SEO strategies for startups that had no budget and somehow needed to outrank companies spending £50,000 a month on content.

Through all of that, the fundamental model was the same. Understand what people search for. Create content that answers those searches better than the competition. Build authority through links and brand signals. Get Google to send you traffic. Convert that traffic into revenue.

That model worked for twenty years. It's not broken yet. But it's cracking. And the cracks are getting wider every quarter.

What Google Actually Announced at I/O 2026

On May 19, 2026, Google held its annual I/O developer conference. Sundar Pichai got on stage and said things that should have made every SEO professional sit up straight. Most of them were too busy arguing about schema markup on Twitter to notice.

Here's what actually matters from that keynote.

AI Overviews and AI Mode have merged into one seamless experience.

Google called it the biggest Search box upgrade in 25 years. That's not marketing fluff. They meant it. The distinction between "regular search" and "AI search" is gone. It's all AI search now. Every query goes through Google's AI systems and the model decides whether to show you links, an AI-generated answer, or both. You don't choose. Google chooses.

AI Mode has crossed 1 billion monthly users.

It quadrupled its user base between May and November 2025, then doubled again in the six months after that. 2.5 billion monthly users on AI Overviews globally. These aren't experimental numbers. These are "this is how search works now" numbers.

Gemini 3.5 Flash is the default model powering everything.

Not an optional feature. Not a beta. The default. Every search query in the US (and increasingly globally) runs through Google's AI before it decides what to show you.

Information Agents are launching this summer.

This is the one that changes things most. Users can tell Google what topics they care about and receive ongoing AI-generated updates automatically. No more searching. Google just tells you. Continuously. Without you ever typing a query or clicking a link.

Google added a Perspectives filter

that specifically surfaces forum discussions, social media posts, and video content. This is a dedicated search feature that explicitly favours community-generated content over polished editorial. Most founders don't even know it exists but it's sending traffic to exactly the kind of content we produce.

The combined message from Google is clear: search is becoming more conversational, more AI-driven, and more weighted toward authentic human voices. That's not me interpreting. That's Google's VP of Search, Liz Reid, saying it directly to the Wall Street Journal.

The Click-Through Rate Numbers (I Know Because I'm Living Them)

Let me show you what this looks like in practice, because I'm not reading about it in a case study. I'm watching it happen to Startup Networks in real time.

We're approaching 1 million monthly impressions. That sounds incredible. And in one sense it is. Our content is being seen by hundreds of thousands of people every month. Google is showing our pages in search results, in AI Overviews, in featured snippets, in People Also Ask boxes, in Perspectives panels.

The problem? Our click-through rate is lower than it should be given those impression numbers. We're getting massive visibility and relatively modest traffic. Millions of impressions, thousands of clicks. The ratio is off.

This isn't unique to us. The data across the industry tells the same story.

Amsive analysed 700,000 keywords across ten websites and measured an average 15.49% CTR drop on queries where an AI Overview appears. Non-branded queries fell 19.98%. Queries showing an AI Overview alongside a featured snippet dropped 37.04%.

Pew Research Center found a 46.7% relative decline in click rates on queries with AI summaries. 8% clicked when an AI Overview was present vs 15% when it wasn't.

Ahrefs measured a 34.5% CTR drop for position-1 rankings across 300,000 keywords. Position one. The spot every SEO has been fighting for since 1998.

Seer Interactive tracked 2.43 billion impressions across 53 brands over 15 months. Average CTR on AIO-present queries dropped from 1.76% to 0.61%.

HubSpot estimates 70-80% organic traffic decline. Business Insider lost 55% of its organic traffic between April 2022 and April 2025 and cut staff by 21%. Chegg's revenue fell 24% year-on-year and the company sued Google explicitly citing AI Overview impact.

But here's the nuance nobody's mentioning. Datos's Q1 2026 State of Search report found that US zero-click rates actually fell from 24.5% in December 2025 to 22.4% in March 2026. The worst of the disruption may be levelling off as AI Overview prevalence stabilises. The sky is not falling. The ground is shifting. There's a difference, and it matters for how you respond.

What Google Is Actually Pushing TOWARD (Not Just Away From)

Everyone's obsessing over what Google is taking away. Nobody's paying enough attention to what Google is actively rewarding. That's where the opportunity is.

Forum content and user-generated discussions. Liz Reid told the Wall Street Journal in October 2025 that Google has officially updated its ranking systems to show more forum discussions and UGC. Google's Danny Sullivan publicly said the rise of forum content in results is partly a response to "how much low-quality, marketing-driven content people were getting instead." Google's own spokesperson is admitting that professional content is so bad that people prefer anonymous forum discussions. That's remarkable.

Video and visual content. Short-form video is appearing in more search results than ever. The Perspectives filter specifically surfaces video alongside forum posts and social content. If you're not producing video, you're invisible in an increasing number of queries.

Experience-based content over expertise-based content. The extra "E" in EEAT stands for Experience. Google isn't just asking "does this person know about the topic?" It's asking "has this person actually done the thing they're writing about?" A blog post about public liability insurance written by a content writer who researched it is less valuable than one written by a founder who's actually bought a policy and watched a community member get sued without one.

Interactive tools and original data. Calculators, comparison tools, original surveys, proprietary data sets. These are content types that AI Overviews can't replicate because they're interactive or they contain data that doesn't exist elsewhere on the web. A startup cost calculator that gives personalised results based on your industry can't be summarised into an AI answer. It has to be used. That means a click. That means a visit. That means ad revenue.

Brand entities over generic keywords. Branded queries gained 18.68% CTR even with AI Overviews present. When someone searches specifically for your brand, Google sends them to you. AI doesn't cannibalise brand traffic. Building your brand is now an SEO strategy, not a separate activity.

Conversational long-tail queries. AI Mode is conversational. Users are typing longer, more specific questions. "Best public liability insurance for a sole trader electrician in Manchester" instead of "public liability insurance." Long-form, opinionated, experience-rich content is better suited to these conversational queries than short generic pages. That's an advantage for anyone producing the kind of deep, specific content we publish.

The Reddit Deal and Why Forum Sites Just Became the Most Valuable Properties on the Internet

In February 2024, Google paid Reddit $60 million for structured API access to its content. Sixty million dollars. For forum posts.

By early 2025, Reddit had become the second most visible site in Google US results. Behind Wikipedia. Ahead of Amazon. Ahead of every news publisher, every SaaS company, every media empire.

Reddit's readership nearly tripled between August 2023 and April 2024 according to the Columbia Journalism Review. By mid-2025, Reddit was the most cited domain in Google AI Overviews and Perplexity results according to analytics platform Profound.

Why? Because Google has an AI content problem. 74.2% of newly created web pages contain AI-generated content. The quality of what's being published online is declining because of model collapse. Google's AI Overview system depends on high-quality source content to synthesise, and that source content is getting worse because AI is generating most of it now.

Forum content is the antidote. Real people having real conversations can't be faked by AI at scale. The messy, specific, opinion-heavy, sometimes contradictory nature of forum discussions is exactly what makes them valuable to Google's systems. A blog post says "HubSpot is the best CRM for startups." A forum thread says "I tried HubSpot for six months and it was overkill for our three-person team, switched to Pipedrive, much better, but the migration was painful and here's what I'd do differently." Google knows which answer is more useful to a founder making a real decision.

Now think about what Startup Networks is. A founder forum. Real founders. Real conversations. Real problems discussed with real opinions. Every thread has multiple perspectives. Every comment adds depth. This is exactly what Google told us they want.

To be clear about where this advantage applies. Forum content outranks professional publishers specifically on opinion-based, comparison, and experience-driven queries. "Best CRM for startups." "Is public liability insurance worth it?" "What's it like to raise pre-seed funding?" These are the queries where Reddit threads with three upvotes sit above sites with thousands of backlinks because Google has decided that real human discussion answers the question better than polished marketing content.

The HouseFresh example tells you everything. HouseFresh, an independent review site, wrote an article called "How Google is killing independent sites like ours." A Reddit thread discussing that article now outranks the original article in Google. Reddit is ranking above the source it's discussing. That's the world we're in now.

Where forums DON'T outrank professional content: technical documentation, official government guidance, medical reference, legal statute. Reddit won't outrank GOV.UK for "how to register a company" or NHS for "symptoms of appendicitis." Facts are still facts and specialist sites still own them. But for anything involving opinion, experience, recommendation, or "should I?" - forums are winning.

We didn't build a forum because we predicted the AI shift. We built it because that's what a community is. But the timing has worked in our favour and the structural advantage is gangster.

The Sites That Are Getting Hammered (And Why)

I'm not going to name names because some of these are people I know. But in the UK startup content space, there are sites right now that are mass-producing AI-generated articles. Fifty a week. Seventy a week. Churning out generic "best X for startups 2026" content that all reads the same because it's all generated by the same models trained on the same data...

These sites saw initial traffic gains. Google indexed the content. Some of it ranked. For about three months, the strategy looked like it was working.

Then the December 2025 core update hit. Then the March 2026 update. Then the helpful content classifier caught up. And those sites lost 40-60% of their organic traffic in a matter of weeks. Because Google's systems got better at identifying mass-produced AI content and demoting it. Not just the AI pages. The entire domain. The programmatic SEO approach that Mark Williams-Cook warned about on LinkedIn is destroying sites from the inside out.

We publish four to six hand-written articles a week. Every one has a named author, personal anecdotes, community references, and original analysis. Every one passes AI detection. Every one gets indexed within 35 minutes of requesting indexing. Google trusts our site because the content is genuine and the publishing pattern is consistent.

That's not a scalable strategy in the way that generating fifty AI articles a week is scalable. But the AI articles are getting demoted while ours are ranking. Speed doesn't matter if the destination is a penalty.

How Our Insurance Cluster Proves the New Model Works

I want to give you a concrete example because theory is useless without proof.

We built a business insurance content cluster over the last few months. Public liability insurance. Sole trader public liability. Employers' liability. Cyber insurance. Each article is 3,000-5,000 words. Each one is written in first person with community anecdotes, personal opinions, and specific experiences. Each one links to the others. Each one was indexed within hours of publishing.

The insurance content targets commercial-intent keywords where AI Overviews appear only 4% of the time (compared to 39.4% on informational queries). When someone searches "do I need public liability insurance for my sole trader business," they're not looking for a definition. They're looking for a decision. AI can define public liability insurance. It can't tell you whether YOUR specific business needs it based on YOUR specific circumstances with YOUR specific risk profile. That requires nuance, opinion, and judgement that the user needs to evaluate for themselves.

That's why commercial-intent content still generates clicks even in the AI Overview era. The AI gives a partial answer. The user wants more depth, more opinion, more "what would you actually do?" That click goes to content that has a genuine human perspective. Our insurance articles are exactly that.

The CPC on insurance keywords is £5-8. The RPM on insurance content is $15-35. Four articles generating 10,000 monthly pageviews between them produces more ad revenue than a hundred general startup articles generating 50,000 pageviews. Fewer visitors. More money. That's the new maths.

What This Means for Ad Revenue

Here's the bit that connects all of this to actual money.

Alphabet's Q1 2026 earnings showed Google Network ad revenue, the money that flows to publishers through AdSense, already under pressure. If AI Overviews reduce the number of times users click through to publisher websites, there are fewer pageviews. Fewer pageviews means fewer ad impressions. Fewer ad impressions means less revenue for everyone except Google.

This directly affects our revenue strategy. We're building toward Mediavine and Raptive qualification. Those premium ad networks pay $15-45 RPM versus Adsense's $3-8. But if AI Overviews continue eating clicks, we need more pageviews to generate the same revenue. Or we need each pageview to be worth more. Or both.

That's why the shift toward high-CPC content matters. Insurance content at $30 RPM means we need one-sixth the traffic of general startup content at $5 RPM to generate the same revenue. In an environment where clicks are harder to earn, making each click more valuable is the rational response.

Revenue per visitor matters more than visitor count. That's a sentence I never would have written three years ago. Now it's the cornerstone of our content strategy.

What To Actually Do on Monday Morning

Enough theory. Here's what I'd tell you to change this week.

Rewrite the first paragraph of every article. Lead with a direct answer. Google's AI pulls from content that answers queries immediately. If your article builds up to the answer over 2,000 words, the AI will pull someone else's direct answer instead. Our BLUF (Bottom Line Up Front) approach exists for this reason. Bold first paragraph. Direct answer. Then the personal story. Then the depth.

Check which of your queries trigger AI Overviews. Go to Search Console. Look at your queries. Search each one manually. See which ones show AI Overviews. For those queries, your goal is citation, not ranking. Structure your content so Google's AI cites you as a source, which generates 35% more clicks than not being cited.

Add FAQ schema to every article. FAQ sections with proper schema markup give Google structured Q&A content that's easy to pull into featured snippets and AI Overviews. We add 5-8 FAQs to every article we publish. Each one is a potential featured snippet.

Start one forum discussion per week. If you have a community or forum, create a genuine discussion thread about a topic your audience cares about. The discussion generates user-generated content that Google is explicitly prioritising. It costs nothing and creates indexable content.

Post on LinkedIn linking to every article you publish. Social signals drive direct traffic that's independent of Google's algorithm. A LinkedIn post about your article generates immediate visits, engagement signals, and brand searches. All of which help SEO indirectly.

Build one interactive tool this month. A calculator, a checker, a comparison widget. Something that requires user input and produces personalised output. Google can't summarise it into an AI Overview. Users have to click through and use it. That's guaranteed traffic in an era where informational clicks are declining.

Audit your content for AI patterns. If you've been using AI to draft articles, check whether they pass detection. Google's helpful content system is getting better at identifying AI-generated content at scale. We run every article through our own editorial audit before publishing, checking for the patterns that flag content as machine-generated: uniform sentence lengths, corporate language, predictable structure, lack of personal voice. The articles that rank within 35 minutes are the ones that are genuinely human.

Diversify your traffic. If 90% of your traffic comes from Google organic, you're one algorithm update from disaster. Build an email list. Grow a community. Post on social media. Start a newsletter. Every traffic source that isn't Google is insurance against Google changing the rules again.

The Metrics That Actually Matter Now

Search Console is measuring the wrong things for 2026. Not deliberately. It's just built for a world where impressions lead to clicks. That relationship is weakening.

Here's what I actually track now:

• Brand search volume over time. Are more people searching specifically for "Startup Networks" month on month? This tells me whether brand awareness is growing even if organic clicks are flat. It is. And branded queries don't get cannibalised by AI Overviews.

• Direct traffic trends. People who type our URL directly or have us bookmarked. This traffic is completely independent of Google. If direct traffic is growing, our brand is working.

• Community metrics. New member sign-ups, forum activity, event registrations. These are the outcomes that matter. A click from Google is only valuable if it converts into something. We're measuring the something, not just the click.

• Citation frequency. How often is our content cited in AI Overviews, featured snippets, and People Also Ask boxes? If we're generating impressions at position zero but not getting clicks, we're being cited. That's brand exposure even without traffic.

• Revenue per visitor. If our insurance content generates $30 RPM and our general content generates $5 RPM, we need fewer visitors to the right pages rather than more visitors to any pages. Quality of traffic over quantity of traffic.

The SEO industry is addicted to traffic as the primary metric. Traffic is becoming a lagging indicator. The leading indicators are brand strength, community engagement, and revenue per visitor.

What I'm Telling Clients Now (After a Decade of Telling Them the Opposite)

I spent years telling agency clients to "create more content, target more keywords, build more backlinks." That advice generated millions in revenue for the agencies I white-labelled for and it worked. Consistently. For years.

I can't give that advice anymore with a straight face. Not because it's wrong. Because it's incomplete.

Stop producing content for content's sake. The "publish 30 blog posts a month" era is over. Google's helpful content system actively penalises sites with high volumes of low-quality content. I'd rather publish four genuinely excellent articles a month that get cited in AI Overviews than thirty mediocre ones that generate fleeting impressions. Quality over quantity has always been true in theory. In 2026, it's true in practice because Google's AI systems can now distinguish between content that says something genuinely useful and content that exists purely to rank.

Build something Google can't replace. If an AI can answer the query as well as your article can, your article is going to lose traffic. Build content that can't be summarised into an AI answer: interactive tools, original research, community discussions, first-person experiences, opinionated comparisons. The EMI scheme guide we wrote works because it contains personal mistakes, community anecdotes, and opinionated recommendations that an AI Overview can't replicate without losing the voice that makes it useful.

Accept that some informational traffic is gone. "What is public liability insurance?" will be answered by Google's AI and the user may never visit your site. That traffic is declining. Focus on the queries AI can't answer fully: "which public liability insurance should I buy for my three-person construction startup in Birmingham?" That requires nuance, opinion, and context that AI Overviews can't provide. Your content needs to live in the gap between what AI can answer and what it can't.

Treat SEO as one channel, not the channel. The agencies I worked with built entire businesses on Google organic. Twelve people, no product, no community, no brand - just client SEO campaigns generating traffic. Those agencies are in trouble because their value proposition depends on a distribution channel that's getting less predictable every quarter. The smart ones are diversifying into paid, social, email, community. The others are watching their margins compress while pretending the landscape isn't shifting underneath them.

What's Coming Next (And Why Panicking Is the Wrong Response)

Information Agents launch this summer. More publisher lawsuits are coming. AI Mode will become the default globally. Ad revenue for content publishers will face more pressure.

None of that means SEO is dead. It means SEO is changing faster than it has since Google launched. The fundamentals haven't changed: people need information, they need to make decisions, and they need recommendations from sources they trust. What's changing is the delivery mechanism.

The sites that will struggle are the ones built entirely on Google organic traffic with no community, no brand, no direct audience. If Google stops sending visitors, what do they actually have? A domain name and a hosting bill.

The sites that will thrive are the ones with direct relationships with their audience. Email lists, communities, memberships, brand recognition strong enough that people search for you by name. For those sites, Google's changes are a headwind, not a death sentence. The wind is real but the ship is sturdy.

Why I'm Genuinely Optimistic

I know this article has covered a lot of challenging data. But I want to end on something honest rather than forced optimism or fake doom.

Startup Networks is better positioned for this shift than I have any right to claim credit for. We're a forum in an era when Google is explicitly rewarding forums. We produce hand-written, experience-based content in an era when AI-generated content is being demoted.

We have a community of 3,000+ real founders in an era when authenticity is the scarcest commodity on the internet.

Our articles are indexed within 35 minutes of requesting it. Google trusts our site. Our content gets cited in AI Overviews and featured snippets. Our brand searches are growing. Our community is growing. The old metrics (clicks, CTR) look concerning. The new metrics (brand awareness, community engagement, citation frequency, revenue per visitor) look strong.

The golden age of "anyone can rank with enough blog posts and backlinks" is over. The new era rewards the things that are hardest to fake: genuine experience, real community, specific expertise, and the willingness to have an opinion and put your name on it.

For content farms, affiliate sites, and AI wrapper companies producing thousands of generic pages? This is the worst time in a decade.

For founders who've actually built something, who've actually made mistakes and learned from them, who've got genuine opinions worth reading and a community that trusts them? This might be the best time since Google started.

I know which side of that divide I'm on.

FAQs

How are AI Overviews affecting SEO in 2026?

AI Overviews reduce organic click-through rates by 15-46% depending on query type. Informational queries are hardest hit with 39.4% triggering AI Overviews. However, brands cited inside AI Overviews see 35% more clicks than non-cited competitors, making citation the new position one. The worst of the disruption may be levelling off - Datos data shows zero-click rates actually fell slightly between December 2025 and March 2026.

Is SEO dead in 2026?

No. The old version of it is being rewritten. Keyword-stuffed blog posts and mass content production are declining in effectiveness. What works now: building genuine communities, creating content with first-person experience, targeting commercial-intent queries that AI can't fully answer, investing in brand awareness, and producing interactive tools that require a click to use.

Why are forums outranking professional websites?

Google VP Liz Reid confirmed in October 2025 that ranking systems have been updated to favour forums and UGC because user behaviour is shifting toward authentic voices. Reddit became the second most visible site in US Google results in 2025. Google paid Reddit $60 million for API access. Danny Sullivan tied forum ranking growth to the decline in quality of professional marketing content.

How do I get cited in Google AI Overviews?

Lead with a direct answer in your first paragraph. Include specific data, dates, and facts that AI can extract. Build topical authority through content clusters. Create content with genuine EEAT signals that adds perspective beyond what AI could generate. About 76% of AI Overview citations come from pages ranked in the top ten, so traditional ranking still matters as the foundation.

What is the Google Perspectives filter?

A dedicated search filter that surfaces forum discussions, social media posts, and video content alongside traditional results. Launched as part of Google's push toward authentic user-generated content. Particularly relevant for community-driven sites and forums where multiple perspectives on a topic exist.

Should startups still invest in SEO?

Yes, but differently. Focus on community content that generates authentic discussion. Create experience-based articles with strong EEAT signals. Target commercial-intent keywords where decisions are needed, not just answers. Build interactive tools that require visits. Diversify traffic sources. And measure brand strength and revenue per visitor, not just traffic volume.

What are Google Information Agents?

Launching summer 2026. Users set topic preferences and receive ongoing AI-generated updates without searching. Google monitors the web on the user's behalf and delivers synthesised updates. This could further reduce direct website visits. The counterplay is building direct audience relationships (email, community, social) that don't depend on Google sending you traffic.

Written by James Beresford-Morgan. Co-founder of Startup Networks. Ten years in SEO across white-label agency work, million-pound campaigns, university workshops, and founder webinars. Survived Panda, Penguin, Helpful Content, and whatever Google decides to call the next one. Currently watching my own site's metrics do things I've never seen before and trying to figure it out in real time alongside 3,000+ founders who are doing the same thing.

If you're navigating this shift, join our community. We're sharing what works as we discover it. No guru nonsense. Just founders comparing notes while Google rewrites the rules.

Where I got my data (because I'm not about to write an article criticising AI-generated content and then not source my own claims): The I/O 2026 stuff comes directly from Google's blog post on May 19, 2026 and Pichai's keynote. The 2.5B and 1B user numbers are Google's own. CTR declines come from four separate studies and they all point in the same direction: Amsive (700K keywords, 15.49% average drop), Pew Research Center (68K queries, 46.7% relative decline), Ahrefs (300K keywords, 34.5% at position 1), and Seer Interactive who tracked 2.43 billion impressions across 53 brands for 15 months and watched CTR drop from 1.76% to 0.61%. Publisher losses come from their own public statements and filings: HubSpot (70-80%), Business Insider (55% traffic loss, 21% staff cuts), Chegg (24% revenue decline, sued Google in Feb 2025). The zero-click counterpoint is Datos's Q1 2026 State of Search Report. Reddit's 1,328% visibility explosion is SISTRIX data. Reddit hitting #2 most visible US site is Saastorm. Readership tripling is Columbia Journalism Review. The $60M Google-Reddit deal was February 2024 and widely reported. Liz Reid's comments about forums and UGC are from her Wall Street Journal interview, October 2025. Danny Sullivan's forum ranking comments are from public Google communications. The 35% citation click boost is Digital Applied, March 2026. The HouseFresh outranking example was covered by Search Engine Land. AI Overview appearance rates (39.4% informational, 4% e-commerce) are from Stackmatix. Everything verifiable. Nothing invented. Unlike some AI-generated content I could mention.