Allen Wu
For the first time, we piece together how ChatGPT shopping works behind the user experience, piecing components together, from user query to product search to card-offer relationship, to help merchants truly begin to understand how to approach AI shopping. Here's what over 200k shopping prompts revealed.
Helen Martin and Nitish Natarajan
FactCheck catches inaccurate claims about your brand. Now integrated with Noble's Mention Refresh, getting the source material corrected is easier than ever.
Matthew Huo and Achintya Ashok
Profound is now available as an official Anthropic connector.
Helen Martin and Josh Prunty
Not sure what to build next with your AI Search data? Brainstorm Mode guides you from a broad goal to a scoped, ready-to-run Agents.
Jennifer Zou
Every AI assistant has a daily and weekly rhythm. It just isn't the same rhythm on every platform, in every region, or for every user.
Davis McCain
Gemini, AI Overviews, and AI Mode share the same underlying infrastructure, but they don't behave like the same product. Across 15,155 brand configurations tracked daily in May 2026, brands saw a median 8-point visibility gap between their best and worst-performing Google model, driven not by how many brands each model mentions (they're nearly identical, 4.4 to 5.0 per response) but by which brands they choose to feature and which sources they cite to back them up. Gemini leans on editorial and reference sources like Reddit, YouTube, and Wikipedia, while AI Overviews and AI Mode cite far more heavily from social and UGC platforms, at roughly double Gemini's citation depth per run.
Multiple authors
Profound teamed up with Kevin Indig and Clickstream Solutions on one of the first behavioral studies of how real users actually make decisions inside ChatGPT, the product of more than 150 hours of research design, participant sessions, behavioral coding, transcript analysis, statistical work, and writing. Watching 56 people compare products across 221 real shopping tasks, the study uncovered a surprisingly strong link between how often a brand shows up in ChatGPT's answers and what people ultimately choose to buy. The findings reveal something most marketers haven't caught onto yet: the moment of decision has quietly moved somewhere their analytics can't see.
Davis McCain
Google is increasingly keeping AI Mode users inside its own ecosystem, with google.com rising to the model’s second most-cited domain and Business Profiles and Product Knowledge Panels appearing directly in answers. This shift is most pronounced across local-intent industries and physical consumer products, where customers may see a Google-hosted card instead of visiting a brand’s website. For brands, accurate profile information, complete product data, and strong review management are becoming critical to winning visibility and preventing customers from switching to competitors.
Davis McCain and Kate Underwood
Travel planning is moving away from the traditional search bar and into AI answer engines like ChatGPT. For the better part of two decades, a typical traveler would type a destination-focused keyword into Google, open a dozen tabs, and piece together a trip on their own. Now, they ask a question or have an ongoing conversation, letting the answer engine synthesize recommendations or even book a trip. This shift rewrites how searchers find travel brands. Visibility no longer comes from securing top-ranked blue links in traditional search. Instead, it results from earning mentions and citations in AI search. Increasingly, the brands that appear in AI search are the ones that win the trip.
Allen Wu and Kate Underwood
Christmas shopping has shifted away from the search bar. More shoppers are turning to AI answer engines to research products, compare options, and decide which gifts to buy. For retailers, a reframe is critical. Answer engines shortlist products and guide purchase decisions long before shoppers ever reach retail or ecommerce websites. This compresses the shopping funnel and makes traditional search insufficient as a primary customer acquisition channel. Instead, retailers need to ensure answer engines recommend them throughout the shopping journey. This requires understanding how shoppers use answer engines for Christmas gifting, how to earn mentions and citations in relevant AI responses, and how to maximize visibility across AI search. This report gives you that edge. It draws on real shopper behavior from Christmas 2025, analyzed through Profound’s AI visibility lens, illustrating how people use AI to shop for the holiday. It also provides actionable takeaways, revealing how to optimize e-commerce products before the 2026 season ramps up so you can win the AI shelf this Christmas.
Jennifer Zou
Why we "only" run each prompt one time per day (and the statistics that show it’s plenty).
Nikolas Laskaris
Measuring downstream web browsing after AI brand mentions
Multiple authors
Your brand is being misrepresented in AI search right now, and you probably don't know it. Nearly half of every AI response includes unsolicited comparisons, opinions, and recommendations the user never asked for. We analyzed 50,000 prompts across seven industries to understand why. Here's what we found, and what you can do about it.
Allen Wu
The rise in importance of product feeds in ChatGPT Shopping. Are PDPs still prevalent when ChatGPT searches for product information? What does this mean for brands? Here's what over 1 million ChatGPT shopping offers revealed.
Nick Lafferty
Five recent studies are reshaping how marketing leaders should think about AI Search.
Brandon Punturo
On May 7, 2026, ChatGPT quietly shipped the largest single-day change in AI-driven brand traffic we've measured all year. OpenAI referrals to monitored brand sites roughly doubled overnight. ChatGPT started hyperlinking brand names directly to their homepages, turning every AI mention into a clickable referral.
Jennifer Zou
When a user types a question into an AI engine, the engine doesn't always search for exactly what was typed. It fans out. Profound tracked 10,000 prompts across ChatGPT, Copilot, and Perplexity over 14 days and found striking differences in search behavior.
Davis McCain
Profound analyzed 3.25 billion citations across 7 models and 14 countries, and the data shows that query language is the dominant force reshaping social citation rates across AI models. Google AI Overviews and ChatGPT respond to non-English prompts in fundamentally different ways, with major implications for global brand visibility in AI search.
Brandon Punturo
Walmart leads rank-1 buy links. Target leads total presence. The carousel reshuffles on every request. Here's what 22.5 million ChatGPT shopping offers actually revealed.
Brandon Punturo
ChatGPT Shopping is driven far more by product category than by purchase intent language. In our analysis of 1.18 million prompts and a 7,500-prompt labeled sample, prompts naming shippable consumer goods were much more likely to trigger Shopping cards, while software, services, travel, and financial products almost never did. Adding 1–2 concrete constraints like price, features, or use case increased trigger probability, but only within product categories. The clearest rule was simple: if the prompt’s main noun is something you could buy on Amazon, Shopping is likely to appear. Using that logic, we built a classifier that reproduced ChatGPT’s Shopping behavior with about 95–97% accuracy.
Davis McCain and Brandon Punturo
In the last three months, LinkedIn has become one of the most authoritative sources shaping AI-driven discovery. Between November 2025 and February 2026, the platform surged from outside the top 20 to rank among the most‑cited sources on ChatGPT – and is now the #1 most‑cited domain for professional queries across all AI Search platforms. That’s powerful data for both companies and individuals looking to grow their influence.
Brandon Punturo
79% of prompts never trigger shopping. Not rarely — never, across nine months of data. The ones that do are surprisingly sticky: 83% chance of triggering again the next day. But that persistence decays, and model updates can reset it overnight. We tracked 26 million prompts across 13,000 categories to map where shopping actually shows up, how consistently it fires, and what that means for brands building strategy around a surface that barely turns on.
Brandon Punturo
We ran a controlled experiment to test whether serving Markdown to AI crawlers increases bot traffic. After 381 pages and 3 weeks of data, here's what we found.
Brandon Punturo and Sartaj Rajpal
ChatGPT cites Reddit threads, YouTube channels, and LinkedIn profiles but ignores X/Twitter. Which social platforms matter for AI visibility?
Brandon Punturo
We analyzed 700K ChatGPT conversations with web citations from Q4 2025. The findings: Turn 1 captures most citations, Wikipedia is the default knowledge layer, and sources cluster by topic. What marketers need to know about the new citation economy.
Josh Blyskal
Here's a copy of my deck from Tech SEO Connect in Raleigh, NC on December 5th 2025.
Josh Blyskal and Sartaj Rajpal
Profound, in collaboration with Reddit, analyzed how conversational data informs AI-generated answers.
Multiple authors
In mid-October, ChatGPT rolled out a major response update that changed how brands appear in its answers. Using Answer Engine Insights, Profound analyzed millions of prompts across ChatGPT and other leading answer engines to understand what changed and how it affected brand visibility across categories.
Josh Blyskal and Sartaj Rajpal
New research reveals ChatGPT's dramatic pivot from Bing to Google search results, fundamentally altering how we need to think about AI visibility and optimization.
Josh Blyskal and Sartaj Rajpal
New research reveals 'citation drift' in source selection across major AI platforms, showing how AI citations can change by up to 60% in just one month.
Josh Blyskal and Sartaj Rajpal
The era of one-size-fits-all content strategy is over. Our analysis of 100,000 distinct prompts run across both ChatGPT and Perplexity revealed a surprising truth: these models are citing fundamentally different corners of the internet.
Josh Blyskal and Sartaj Rajpal
New research reveals drastically different citation patterns between major AI platforms. Understand how ChatGPT, Google AI, and Perplexity source their information differently and what it means for AI visibility.
Never miss an update
Get all the news and updates about Profound, directly to your inbox.