Two decades later, that shopper may instead ask a chatbot where to buy a sofa, leaving companies built on those once-prized domain names scrambling to make sure they show up in the answer.

Furniture.com, which traces its roots back to the dot-com boom, is one such company grappling with how AI and chatbots are changing the way shoppers search for information.

The site was originally founded in 1999, when simple web addresses could help steer shoppers straight to a website. The domain eventually changed hands before being acquired by furniture retailer Rooms To Go’s venture arm, which relaunched it as a standalone business in 2023.

Today, the company operates as a market aggregation website that helps shoppers browse products from more than 70 furniture retailers in one place. But like many online businesses, it is also preparing for a future in which people increasingly ask chatbots for shopping recommendations instead of typing keywords into Google. Furniture.com is focusing on publishing more accurate, up-to-date product information that AI systems can easily interpret. The company is also redesigning its website and engaging more with users on social media sites, including Reddit.

“The shift from keyword to more semantic-based and conversational search and discoverability is a big part of how we’re thinking about Furniture.com as a website and as a platform,” said Alex Seaman, Furniture.com’s senior vice president and co-founder.

Whether shoppers will ultimately complete purchases directly through AI chatbots remains an open question. The Information reported last week that OpenAI is scaling back plans to introduce shopping directly inside ChatGPT. But there are plenty of signs consumers turn to AI assistants for shopping research. In a recent IBM survey, 41% of consumers said they use AI assistants to research products, while 33% use them to find reviews and 31% to look for deals. Another survey by eMarketer found that 53% of U.S. consumers have used AI tools to research purchases. For brands, the growing use of AI for shopping research is reason enough to try to show up in chatbot answers.

From SEO to GEO

Traditional search engine optimization, or SEO, emerged in the early days of the web as businesses competed to appear at the top of Google’s list of blue links. Early tactics often involved trying to game the system, said Mollie Ellerton, head of SEO at digital optimization agency Hookflash. In the early days, some companies stuffed pages with keywords or built networks of sites linking to each other in an attempt to boost rankings.

“People were putting loads and loads of keywords onto a page and just making it a white font to match your background,” she said.

Over time, Google refined its algorithms to reward more relevant content, pushing brands to focus less on tricks and more on creating useful information for users. But the arrival of AI-powered search tools like ChatGPT and Google’s Gemini is changing the rules once again. The new discipline emerging from this shift is often called generative engine optimization, or GEO. Instead of trying to rank on a list of links, brands are trying to ensure they are included in the answers that chatbots generate.

Furniture.com executives see GEO as an extension of traditional SEO, rather than a replacement for it entirely. “The shift from SEO to GEO is actually more an evolution than a shift,” Seaman said. “A lot of the original pillars of SEO have just changed, and they started changing really rapidly as AI and GEO became more of the landscape.”

One key difference is the way people search. Instead of typing short phrases like “modern sofa,” consumers can now ask detailed questions. For example, a shopper may ask a chatbot for a sofa and side table for their living room in avocado green tones, with a budget under $900 and delivery before Christmas.

Experts say companies need to ensure their product data and website content can be understood by AI systems, not just traditional search engines. In practice, that means feeding the chatbots a steady diet of detailed information so they have plenty of material to pull from when answering users’ questions.

“You really have to make sure that data is standardized and contextualized in such a way that it will show up in these very long-answer, semantic prompts,” Seaman said.

Furniture.com has been redesigning its website with that in mind. The company is standardizing the information attached to each product, including details such as color, style, price, materials and delivery timing, so AI systems can more easily understand and recommend items. That involves restructuring product listings so the same types of information appear in consistent formats across the site. For example, a couch’s color, dimensions and materials must be labeled the same way across thousands of listings rather than described differently by each brand. It is also adding more images, videos and detailed product descriptions to help its products surface when shoppers ask chatbots highly specific questions.

Accurate, consistently formatted product information is essential if AI systems are expected to handle shopping tasks. If the underlying data is incomplete or inconsistent, an AI system could recommend the wrong item, display an outdated price or try to purchase a product that is out of stock or unavailable in a certain size. Bringing all of that information together — and organizing it in a way machines can easily interpret — remains a challenge across the industry. The Information reported in January that OpenAI has been slow to introduce checkout features inside ChatGPT in part because retailers’ product catalogs are often inconsistent or poorly structured.

Despite the hype around GEO, many of the old SEO fundamentals still matter.

“There are many things that are exactly the same — like speed matters, performance matters,” said Dan Russotto, Furniture.com’s co-founder and general manager. “All the things you already were doing for SEO, you still do those things.”

What has changed is the range of sources that AI models pull from when generating answers. Social media is becoming an increasingly common source of information for chatbots. The share of AI citations attributed to social platforms climbed to more than 9% by January 2026, with much of that growth coming from Reddit, according to a recent report from the marketing firm Tinuiti and the AI analytics platform Profound.

In some cases, Reddit posts can play an outsized role in shaping how a brand appears in AI-generated answers. That’s because large language models often look for human opinions when summarizing products or services, said Jen Cornwell, senior director of AI SEO innovation at Tinuiti. Those discussions often skew negative, since consumers are more likely to post when they have a complaint, she added.

“When somebody goes and asks, ‘What’s the best fitness studio?’ it can’t have an opinion, so it needs to pull from somewhere,” Cornwell said. “That’s what makes Reddit such a great source.”

Furniture.com has a small marketing team that watches furniture forums and joins threads on Reddit, where it can answer questions or share useful guides, such as how to pick the right sofa. But the company says it isn’t flooding Reddit with posts, either, and wants content to stay “real” and “organic,” so AI systems are more likely to pick up helpful, accurate comments about the company, Seaman said.

Older information can also unexpectedly shape how brands appear in chatbot answers. Cornwell said one audit her team conducted for a baby gear company found that an old lawsuit from a decade earlier was frequently surfacing in AI-generated responses about the brand. Because chatbots pull from information scattered across the internet, she said, companies sometimes need to actively reshape the narrative around their reputation. “You have to own that narrative,” Cornwell said. In the case of the baby gear brand, she recommended publishing an annual safety report or similar content that highlights the company’s strengths and helps counter older negative information.

Companies therefore need to pay attention to how they appear across the internet, not just on their own websites. “It’s not necessarily just about the domain anymore,” Ellerton said. “It is also about what you’re doing off-site, too.”

Still, a domain like Furniture.com retains plenty of advantages. Long-established websites often accumulate years of credibility with search engines and other online platforms through marketing, links and brand recognition, Ellerton said. That history can make it easier for a site to be understood as an authoritative source in search and AI-generated answers. “It’s still valuable real estate,” she said.

But those advantages do not make the shift to AI search simple. Unlike traditional SEO, which largely meant optimizing for Google’s search engine, the emerging world of GEO involves a growing number of AI systems, each with its own way of retrieving and interpreting information.

“Optimizing for all of them simultaneously is no small feat,” Seaman said. “And the truth is, there’s still a lot we simply don’t know.”