Posted 15 hours ago15 hr comment_13430 Google is getting worse – or at least, that’s what “we” say. The common complaint: Search is dying, SEO is trash, and Google is nearly unusable. X, LinkedIn, and Reddit are awash with complaints about deteriorating search quality. The contradicting reality: Search remains the dominant channel for web traffic. Google still holds 90% of the search market, grew 20% year over year, and processes 5 trillion searches annually. If it were truly broken, wouldn’t we have left by now? The complex truth: Poet John Keats described “negative capability” as the ability to hold conflicting ideas without demanding a simple answer. That’s where we are with Google: it may be improving in many ways while simultaneously feeling worse. The truth resists easy categorization. This article: Challenges the claim that Google is objectively worse. Explores the real ways it might have declined. Examines what the available data does and doesn’t show. Unpacks why it feels worse regardless of the facts. Explains why we still can’t say for sure if it actually is. Maintaining search quality is a deadly tightrope act Google wants to maximize profits. That means keeping us on its cash cow, search, and clicking on ads. Simple, right? Not quite. The deeper you go, the more complex the incentives become. To keep the machine running, Google has to: Keep shareholders happy. Send enough traffic to creators to keep them publishing. Keep advertisers spending. Deliver fast, relevant results to users. Every tweak to Search sets off a chain reaction across these groups, often creating tension or imbalance. Google makes more money by showing more ads, but that comes at the cost of user satisfaction. Ironically, the happier users are, the less Google earns compared to when users are just slightly less happy. Quality is a soft, slippery metric – hard to define and even harder to measure. Ask five people and you’ll get five definitions. When we talk about search quality, we mean this: Great content from reputable sources ranks well. The user finds what they need quickly and easily. That’s a solid proxy for user-centered quality, but it may not match Google’s priorities. To stay dominant and profitable, Google has to juggle the expectations of all its stakeholders. From its perspective, search quality improves if usage grows and revenue rises. What (could have) led to a decline in Google search quality? As the market leader, Google has always operated under pressure, but in the last 2–3 years, that pressure has intensified. Here are five external forces that have increasingly constrained Google’s strategy. Shopping no longer starts on Google Amazon has carved out a major share of high-value commercial queries. 59% of millennials use Amazon first for product searches. 50% of product searches start on Amazon. 61-74% of European shoppers use Amazon instead of Google. In response, Google: Changed SERP layouts (especially in the U.S.). Continued pushing the product grid SERP feature. Introduced AI-based and personalized shopping results in the Shopping tab. Social media is more engaging – and drives demand Social platforms have evolved into informal search engines, especially among younger users. 40% of Gen Z prefers TikTok or Instagram over Google for restaurant searches. Social is part of the “messy middle” of the purchase journey. TikTok excels in search, Instagram tries to model after them. Google is adapting by integrating more social-style features into search: Videos. Shorts. The Perspectives filter. AI incumbents reshaped the way we find information Despite pioneering transformer tech, Google hesitated – opening the door for competitors. Google isn’t a first mover and has to catch up. Bard (now Gemini) stumbled in early demos. “Funny” AI Overviews. While Gemini shows promise, Google is no longer the bold innovator. ChatGPT’s lead in user adoption is significant. Distrust toward Google Once guided by “don’t be evil,” Google now faces skepticism from all sides. Traffic to publishers is considered “a necessary evil.” They prioritize partners like Reddit. “They prefer themselves over everyone.” Can Google still be trusted? “Yes, as long as you aren’t a competitor, advertiser, or user.” 5. Legal pressure limits flexibility The DOJ’s antitrust trial is Google’s biggest legal challenge to date. The DOJ laid out its plan to regulate Google, which Google didn’t like (e.g., selling off Chrome, its magic powder). If Google is forced to sell Chrome, its power would decrease dramatically over time. Google’s dilemma under pressure With great power comes… fragility. Google’s dominance gives it unmatched reach – but it also makes every misstep a potential knockout. The result? A more cautious, mistake-averse strategy. 4 common arguments for why Google got worse (plus my two cents) 1. Google favors itself This is the most popular argument – and it’s not baseless. Louise Linehan analyzed the data and found: Google’s own store appears more often in AI Overviews than competitors. Google Flights is gaining visibility. In 50 of 57 cases, Google-owned properties saw increases in organic traffic. More recently, self-linking has expanded: People Also Ask now includes AI Overviews. AI Overviews link back to more Google search results. Things to know also loops users back into search. This argument has merit, but its weight depends on how the average user perceives these shifts. If users don’t mind being looped back into Google, it may not register as a problem. Google does earn more from ad clicks, not just pageviews. So, if users stay trapped in search without heading to monetizable queries, that’s not automatically a win for Google. The flywheel only works if they can direct that loop toward ads. 2. Google intentionally degrades search quality Some argue Google is deliberately making search worse to boost revenue (we’ll look at the data later). Revenue = Ad price x Query count. Google increased CPC over time and is incentivized to increase query volume. They’ve experimented with degrading quality before. But given today’s scrutiny and pressure, I doubt they can afford to run tests like that now. Their priorities – and risks – are different. 3. Google serves too much Reddit It’s true: Reddit’s presence in search results has exploded. Forum content is everywhere. Google partnered with Reddit for faster indexing. Reddit Answers is now powered by Gemini. But Reddit isn’t the only forum seeing gains. Reddit’s rise isn’t just about a deal, it reflects user preference. During Reddit’s blackouts, Google visibly panicked, signaling how much it relies on the platform. In 2023, 1% of all queries included Reddit. On Kagi, a customizable search engine, users can block, lower, raise, or pin domains. Reddit ranks low on the “blocked” list and high on the “raised” and “pinned” lists. 4. Google is good but not great AJ Kohn’s piece “It’s Goog Enough” makes a compelling case. He covers more than what’s listed here, so give the full article a read. Highlights include: Google pushed organic results 64% deeper down the SERP (2013 vs. 2020). Google increased the size of shopping ads by 34%. Google started placing ads between organic results. Google relies too much on brands (in my opinion, as a plaster for other problems like generative AI content floods). Google clutters SERPs with images and SERP features. AI Overviews compound the issue. They occupy space like any other SERP feature, but without user interaction. Sometimes, interacting with them can even hide organic results altogether. This argument lands: Google seems focused on preserving the status quo, not innovating. Still, users don’t seem too bothered. Plus, Bing is loading up on ads, and Google labels their ads more clearly. Most of these arguments are rooted in subjective impressions. So, what does the data actually say? Get the newsletter search marketers rely on. Business email address Sign me up! Processing... See terms. Is there scientific data to prove Google got worse? Google’s internal search quality study (2020) There was an internal search quality study at Google in 2020. For three months, Google degraded search by 1 IS point (equal to losing twice of Wikipedia’s information). Revenue losses were minuscule. Does this prove Google can make search worse on purpose? I don’t think so, at least not long-term. Three months is a short time. The minuscule revenue losses were first-order consequences. With measures like that, there are usually second- and third-order consequences. Recent survey data With any survey, tread cautiously. There are often problems with: Sample size. Samples not reflecting the general population. Suggestive questions. Additionally, what we say, what we do, and what we feel often differ quite a bit. Let’s look at three recent surveys: U.S. customers were more satisfied in 2023 than in the five years prior, per a Statista survey. About 42% find search engines like Google less useful, per The Verge’s “Remodeling the Internet” study. This means more than half find them more useful. 63% of people think that Google search results were better last year, per WalletHub. But there are caveats: Sample sizes were small, for instance: Statista: <1,000Verge: ~2,000WalletHub: ~9,000 Even combined, this isn’t enough to generalize broadly – especially globally. U.S.-centric: Search satisfaction in the U.S. doesn’t reflect global experience. The Verge survey used suggestive framing: Respondents were asked to agree/disagree with:“I find search engines like Google becoming less useful.” That phrasing nudges people toward a negative view. Still, there’s value in the expert commentary at the end of the WalletHub piece. Two responses in particular stood out. Rand Fishkin said (emphasis mine): “I think personal opinions about Google’s “quality” are not very meaningful. The data shows that Google has as many searches per searcher as ever and that their market dominance by share has never been stronger. To me, those are indications that, regardless of complaints by many tech-savvy folks, Google is doing as good a job as ever at keeping users and convincing them to come back.” Additionally, Michael King stated the following (emphasis mine): “SEOs and publishers are not the primary audience for Google, so if Google makes changes that generate better performance based on their evaluation measures, I am inclined to trust that because they have a top-level view of performance that someone from the outside can never have.” There are no good surveys to make the claim that Google got worse. ‘Is Google Getting Worse?’ study If you didn’t read this study yourself, you probably heard about it at least 386 times, as it “proves that Google got worse.” I refute. The study lasted from 2022 to 2023 and included 7,400 keywords with a pattern like “best [product category]”. A niche riddled by affiliate spam for decades. Even if the study concluded Google worsened in the tested field, that’s just one of many fields it plays in. Additionally, 2022-2023 might as well be the Paleolithic era in tech time – that’s how fast things move. How did the researchers measure quality? Type-token ratio (= vocabulary richness and diversity). Readability. Affiliate link presence (in the content). SEO proficiency. HTML page structure (emulating Google guidelines). From my SEO experience, this seems like a fraction of what Google measures to determine quality. But that doesn’t matter if you read the researchers’ conclusions (emphasis mine): “We find that search engines measurably target SEO and affiliate spam with their ranking updates. Google’s updates in particular are having a noticeable, yet mostly short-lived, effect. In fact, the Google results seem to have improved to some extent since the start of our experiment in terms of the amount of affiliate spam.” Basically: Google demonstrated improvements. Google outperforms Bing and DuckDuckGo at filtering spam. Content quality decreased across the board (it’s SEO goblins at work). Search quality can decrease without any changes from Google. The issue with all the data we have We don’t have enough of it, both in numbers and replication. The process for something to be scientifically proven is: The initial publication of a study undergoes peer reviews, which alone is not sufficient. The study undergoes replication attempts to verify the results, assume generalizability, and identify errors. One study, even if it was methodologically sound, isn’t enough to prove either side right. Why we think Google is worse (even if it isn’t) We mistake ourselves as the (only) target audience Tech-savvy people, especially SEOs (myself included), operate in a bubble. Our search experience isn’t comparable to a normal person. We “cry” about stuff the vast majority of people don’t care about. We are out of touch with reality. Also, there’s the Facebook effect. The idea is that we complain a lot, yet engage with the product as often or more frequently. Here are a few more examples: Ryanair is driving record numbers, yet public opinion suggests no one wants to fly with them. People say “Amazon is criminal,” yet the company makes more money than ever. I’ll add a personal anecdote. For a long time, a smart group of pro wrestling fans (the Internet Wrestling Community, called IWC) complained about WWE, the biggest wrestling company. Special forums, Reddit, and social media were plagued with negativity. Even in arenas, those “smart fans” were audible on camera. At the same time, WWE was doing well. A vocal minority that loves to complain, lament and moan about everything, yet, watches every show. Understand: We’re not the target audience. We’ve built a self-amplifying narrative History shows that if you tell a big enough lie and repeat it, people will eventually believe it, known as the illusory truth. I’m not saying “Google getting worse” is a lie. I don’t have the data to refute or verify my opinion, and neither does anyone else (besides Google). As market leader, Google is under close supervision. If they make a mistake, it rarely goes unnoticed and people talk about it, publishers write about it, and content creators document it. Once you “really see things for the first time,” they seem everywhere. This is the Baader-Meinhof phenomenon. When you spot flaws in Google search (or their AI products), you’ll perceive them more frequently, especially if you’re looking for them. Taking shots at Google is mainstream in SEO, so better join the party: “The mentality of a herd makes it easy to manage. Simply get more members moving in the desired direction and the others – responding not so much to the lead animal as to those immediately surrounding them – will peacefully and mechanically go along.” – Robert Cialdini, psychologist and author We love to blame others It’s easier to point fingers at others than at ourselves. We like to think we’re the rational ones – “I’m not irrational, narcissistic, or greedy. It’s everyone else.” Google says: “Our systems find 40 billion spammy pages every day.” That’s 14,600,000,000,000 pages a year, an unimaginable number. Why do search engines (not just Google) struggle to combat spam and find quality content? It’s buried in the trash we put out there. More trash (for example, due to AI), the harder it is to surface great results. Even if you argue that the average content quality is higher, the bell curve changes: More average stuff = less bad and exceptional stuff. The center becomes easier to hit, the ends harder to notice. This leads to content entropy, where the perceived density of good content is shrinking. Search feels worse, but not due to any Google changes. Lastly, we produce more mediocre content and love to manipulate. There’s classic SEO manipulation like: Buying links. Faking E-E-A-T. Exploiting Reddit. Infesting high authority pages as parasite. With AI and LLMs reshaping the web, it’s starting to feel like the Wild West again. And when rules blur, some resort to tactics like: Strategic text sequencing. Hidden content (this still works). Cloaking. Bribes. SEO is weaponized for business and political impacts. How ironic, we have less fun and destroy the system’s usefulness. What really reveals how good or bad Google actually is Talk is cheap; behavior isn’t. Surveys and opinions can be misleading. What people do matters more than what they say. When someone claims, “No one Googles anymore,” chances are they’re either pushing an agenda or unaware of their own habits. In reality, 99.8% of LLM users still rely on Google. We don’t always say what we think or do what we say. But we reveal our true preferences through our actions, and Google still dominates. We can’t confidently say ‘Google got worse’ We don’t have enough data to make the claim. What we know: Google earns more money than ever. Google has more users than ever. Google’s stock price levels are better than ever (barring recent political influence). What we don’t know with certainty: If Google provides enough traffic or value for content creators to continue in the long run (currently, it looks like they do). If Google serves the best results in a fraction of a second. Google does a good job overall. Good enough to retain users and better than competitors. Google doesn’t need to provide the best product. It just needs to provide the least inconvenient option that’s easiest to justify. When I wrote this piece, two questions kept coming back to me: If Google search really got worse, when was it actually better? In 2013, I searched for a washing machine after moving out. The results were flooded with affiliate links from people who hadn’t tested the product. It’s not clear things were ever significantly better – just different. If Google got worse, wouldn’t that mean other search engines and LLMs got worse too? The study cited notes that Google struggles with spam – but so do Bing, DuckDuckGo, and others. LLMs aren’t immune either. ChatGPT leans on Bing. Google uses its own index. Perplexity claims independence but often sources from Google. Yes, search could be better. But this isn’t a Google problem – it’s a web problem. Search quality is a function of both the index and the content it pulls from. Google is the gardener, but it can only work with the plants we grow. What now? So, where do we go from here? Diversify your traffic portfolio: Google owes us nothing. We’re too dependent on them and they will do whatever satisfies most stakeholders well enough. Entertain negative capability: Resist forming a clean, emotionally driven narrative. Keep two opposing ideas in your head simultaneously and don’t jump to conclusions too quickly. Put biases aside and look for the truth: The truth of my analysis is that I don’t know. I can’t prove either side, and that’s OK. Sometimes, we have to live with ambiguity. Focus on genuinely valuable content: Manipulation is always short-sighted, short-term, and short-lived. To escape mediocrity, you need to be excellent. Ignore what people say, watch what they do: We rarely do as we say and that includes all of us. The truth needs to be seen, not heard. Consider your role in the search ecosystem: It’s cheaper and more effective to work on yourself than to criticize others or to blame circumstances outside of your control. We’re all part of this living, breathing web. If we want better search, we need to contribute to it. View the full article
