I Asked ChatGPT to Recommend a Smart Thermostat. Only 3 Brands Showed Up.

\n\n\nLast Tuesday I typed a simple question into ChatGPT: "What's the best smart thermostat for a 2,000 square foot home?"\n\nThe answer came back in seconds. Three brands: Nest, Ecobee, and Honeywell Home. Clean, confident recommendations with pros and cons for each.\n\nThat's it. Three brands in a market with over 40 manufacturers.\n\nI tried Claude next. Same three. Gemini added Emerson Sensi to the list but still topped out at four. Perplexity pulled in some review citations but the actual recommendations? You guessed it—Nest, Ecobee, Honeywell.\n\nThe smart home market hit $100 billion in 2025. Hundreds of brands are competing for consumer attention across thermostats, security cameras, smart locks, lighting, and sensors. But when consumers ask AI for help—and 40% of Gen Z now prefer AI search over Google—the conversation is dominated by a tiny handful of names.\n\nThis is the smart home AI visibility gap. And it's worse than most IoT brands realize.\n\n## The Experiment: 50 Smart Home Prompts Across 4 AI Engines\n\nWe ran a structured test. 50 product recommendation prompts across smart home categories—thermostats, security cameras, video doorbells, smart locks, robot vacuums, smart lighting, and home hubs. Each prompt was sent to ChatGPT, Claude, Gemini, and Perplexity.\n\nWe tracked every brand mentioned and counted frequency.\n\nThe results were brutal.\n\nAcross all 200 responses (50 prompts x 4 engines), the top 5 brands in each category captured an average of 83% of all mentions. The remaining dozens of competitors split the leftover 17%—and most got zero mentions at all.\n\nHere's what the concentration looked like by category:\n\n- Smart thermostats: Nest (92% of responses), Ecobee (88%), Honeywell (76%). Everyone else below 15%.\n- Security cameras: Ring (94%), Arlo (82%), Wyze (71%). Reolink appeared in just 12% of responses.\n- Video doorbells: Ring (96%), Nest (74%), Arlo (45%). Eufy at 18%.\n- Smart locks: August (78%), Yale (72%), Schlage (68%). Level and Lockly under 10%.\n- Robot vacuums: iRobot Roomba (91%), Roborock (74%), Ecovacs (52%). Dozens of brands invisible.\n\nThese aren't obscure products getting ignored. Some of these invisible brands have 4.5-star ratings on Amazon, thousands of reviews, and competitive pricing. They just don't exist in AI conversations.\n\n## Why This Matters More Than You Think\n\nHere's a data point that should keep smart home brand managers up at night: ChatGPT mentions specific brand names in 99.3% of e-commerce product recommendation responses. Compare that to Google's AI Overviews, which only mention brands 6.2% of the time.\n\nThat's a 16x difference in brand mention density.\n\nWhen someone Googles "best smart thermostat," they see a list of links and can click through to discover various options. The discovery funnel is wide. But when they ask ChatGPT, they get 3-4 names and a confident recommendation. The funnel is incredibly narrow.\n\nAI doesn't show 10 blue links. It shows 3 names and picks a winner.\n\nThis means the stakes of AI visibility are fundamentally different from Google visibility. In Google, being on page 2 is bad but survivable—you still exist in the index. In AI, being absent means you literally don't exist in the consumer's decision process.\n\n## Why Most Smart Home Brands Are Invisible\n\nAfter digging into the patterns, we identified four primary reasons brands get locked out of AI recommendations.\n\n1. Training Data Concentration\n\nLarge language models learn from the internet. The brands with the most extensive web presence—reviews, articles, forum discussions, comparison pieces, Wikipedia entries—get the strongest signal in training data. Nest has a Wikipedia page with 4,000 words of history. That startup thermostat brand with 500 Amazon reviews? It barely registers.\n\nThis creates a reinforcing cycle. Well-known brands get written about more, which means AI learns about them more, which means AI recommends them more, which means they get written about more.\n\n2. Positioning Ambiguity\n\nI looked at the websites of 15 smart thermostat brands. Seven of them described themselves with some variation of "smart, efficient, connected home comfort." AI models can't differentiate between them.\n\nThe brands that appear in AI recommendations have razor-sharp positioning:\n\n- Nest: "The thermostat that learns your schedule"\n- Ecobee: "Smart thermostat with built-in Alexa and room sensors"\n- Honeywell Home: "Trusted climate control with professional installation network"\n\nEach has a clear, distinct identity. AI can categorize and recommend them for specific use cases. The brands with generic positioning get lost in the noise.\n\n3. Missing Third-Party Signals\n\nWe cross-referenced AI recommendations with coverage on major tech review sites—Wirecutter, CNET, Tom's Guide, The Verge, PCMag. The correlation was striking.\n\nBrands recommended by at least 3 of these 5 sites appeared in AI responses 78% of the time. Brands with no coverage on any major review site appeared in AI responses 4% of the time.\n\nAI models heavily weight authoritative third-party validation. A brand's own website saying "we're the best" carries almost no signal. Tom's Guide saying "this is the best budget option" carries enormous signal.\n\n4. Ecosystem and Compatibility Gaps\n\nSmart home is fundamentally about ecosystems. Works with Alexa? Google Home? Apple HomeKit? Matter? Thread?\n\nThe brands that AI recommends almost universally have clear, well-documented compatibility across major ecosystems. When AI encounters a brand with limited or unclear compatibility information, it defaults to safer recommendations—brands it knows work everywhere.\n\nWe found that brands explicitly documenting Matter protocol support saw a 34% higher mention rate compared to similar products without that documentation. Ecosystem compatibility isn't just a feature—it's an AI visibility signal.\n\n

Real Examples: Visible vs. Invisible\n\nVisible: Ecobee\n\nEcobee appears in 88% of smart thermostat AI recommendations. Here's what they do right:\n\n- Wikipedia page with detailed product history and positioning\n- Reviewed on every major tech publication\n- Clear differentiation: "the one with room sensors and built-in voice"\n- Extensive compatibility documentation (HomeKit, Alexa, Google, Matter)\n- Active presence in smart home subreddits and forums\n- Structured product data on their website\n\nInvisible: Mysa (smart thermostat for electric heating)\n\nMysa makes an excellent product for a specific use case—electric baseboard heating. Over 10,000 five-star reviews. But AI almost never recommends it.\n\nWhy? Limited review coverage on major tech sites (it's a niche product). Wikipedia page doesn't exist. Forum discussions are concentrated in a few HVAC communities. Their positioning, while clear to existing customers, doesn't reach the broader information ecosystem that AI models draw from.\n\nThe irony is that when someone asks "what's the best smart thermostat for baseboard heaters," Mysa should be the obvious answer. But AI defaults to Nest and Ecobee even for this specific query because their overall signal is so much stronger.\n\nVisible: Ring (video doorbells)\n\nRing appeared in 96% of video doorbell AI recommendations—the highest single-brand dominance we measured across any category. Ring's visibility isn't an accident. Amazon's acquisition gave them a massive web footprint. Every Amazon product page, every Alexa integration mention, every Ring subreddit post adds to the signal. Their brand name has become essentially synonymous with the product category, which is the ultimate AI visibility position.\n\nInvisible: Reolink (security cameras)\n\nReolink makes genuinely competitive security cameras with local storage options—a feature many privacy-conscious buyers specifically want. They have strong Amazon reviews and a loyal community. But they appeared in only 12% of our AI test responses. Their web presence is concentrated on their own site and Amazon. Minimal coverage on Wirecutter (they appear in a few roundups but aren't top picks). No Wikipedia page. Limited presence in mainstream tech publications. The AI simply doesn't have enough authoritative third-party signal to confidently recommend them.\n\n## The 99.3% Problem: Why AI Brand Mentions Are a Different Game\n\nLet's zoom out on a critical data point. Research from Profound shows that ChatGPT includes specific brand names in 99.3% of e-commerce product recommendation responses. Google's AI Overviews, by comparison, mention brands only 6.2% of the time.\n\nThis is a 16x difference, and it fundamentally changes the competitive dynamics.\n\nIn traditional Google search, a consumer sees a list of ten results. They might click on three or four. Even brands on page one but not in the top three get some visibility. The discovery surface is broad.\n\nIn AI conversations, the model names 3-5 brands and actively recommends one or two. There's no "page 2." There's no scrolling. The AI made a decision, and either your brand is in that decision or it's not.\n\nFor smart home products, this is especially impactful because purchase decisions are often triggered by a specific need—"my thermostat broke," "I want to add security cameras"—and the buyer wants a quick, trusted recommendation. AI delivers exactly that. Fast, confident, specific. And if you're not one of the 3-5 names, you've lost that buyer before they even knew you existed.\n\nThe math gets worse when you factor in conversation context. Once a buyer starts discussing options with AI and your brand isn't mentioned, they're unlikely to ask about you specifically. They'll refine among the brands AI suggested. "Tell me more about Ecobee vs Nest" doesn't help the brand that never entered the conversation.\n\n## A Framework for Diagnosing Your Smart Home Brand's AI Visibility\n\nIf you're running marketing for a smart home or IoT brand, here's how to assess where you stand.\n\nStep 1: Run the Basic Query Test\n\nAsk all four major AI engines the most obvious product recommendation question for your category. Do it in 5 variations:\n\n- "What's the best [product category]?"\n- "Recommend a [product category] for [common use case]"\n- "[Your brand] vs [top competitor]"\n- "Best [product category] under $[price point]"\n- "Best [product category] for [specific ecosystem]"\n\nTrack every brand mentioned. If you don't appear in any of these, you have a significant visibility gap. You can run these checks quickly at geobuddy.co/check to see your visibility score across all four engines.\n\nStep 2: Audit Your Third-Party Presence\n\nSearch for your brand on the five sites that matter most for consumer tech: Wirecutter, CNET, Tom's Guide, The Verge, and PCMag. Are you reviewed? Are you in comparison roundups? If not, that's priority one.\n\nStep 3: Check Your Positioning Clarity\n\nRead your homepage headline. Could an AI model understand exactly what you do and who you're for in one sentence? If your positioning sounds like every other brand in your category, AI has no reason to differentiate you.\n\nStep 4: Map Your Ecosystem Documentation\n\nIs your compatibility information clear, structured, and consistent? Does your product page explicitly state which ecosystems you support? Is this information on your review profiles and comparison listings too?\n\nStep 5: Assess Your Information Footprint\n\nBeyond your own website, where does information about your brand exist on the internet? Wikipedia? Reddit? Industry forums? Trade publications? The broader and more authoritative this footprint, the stronger your AI signal.\n\n## The Window Is Open—But Closing\n\nHere's the thing about the smart home AI visibility gap: it's not permanent. AI models update. Training data refreshes. Brands that take action now can shift their position.\n\nBut the window won't stay open forever. As more brands wake up to this problem, the competition for AI visibility will intensify. The brands that build their AI presence today—while most competitors are still focused exclusively on Google Ads and Amazon SEO—will have a compounding advantage.\n\nThe smart home market is massive and growing. But size doesn't matter if consumers can't find you. And increasingly, "finding you" means hearing your name from an AI assistant.\n\nThree brands showed up when I asked about thermostats. The question for every smart home brand is simple: are you one of them?\n