AI is starting to change where purchase decisions happen. Consumers can now research a need, compare options and act on a recommendation without ever entering a brand-controlled environment.
A homeowner asks an AI assistant about a dripping faucet. The assistant explains what might be wrong and recommends a vetted professional in the homeowner’s neighborhood. The professional gets hired, and the homeowner never opens an app or visits a website. The entire transaction happens inside a conversation the brand cannot see.
Some journeys are still AI-assisted, with consumers using AI for research or recommendations. Others are becoming more agentic, with AI comparing options and acting on the consumer’s behalf.
A BCG study of more than 13,000 consumers across 12 markets found that nearly a third now use AI at some point in their purchase journey, roughly triple the number from eighteen months earlier. Among the most committed users, seven in ten end up buying products AI recommends. And in more than half of AI-assisted purchase journeys, the technology introduces consumers to brands they had never previously considered.
That last finding deserves particular attention. If a competitor appears in an AI conversation and your brand does not, the consumer may never encounter you. A consumer goods company might lose consideration because an AI assistant surfaces another brand whose information it considers more relevant or credible. A marketplace might find its professionals being hired through AI conversations it never knew were happening.
When the storefront disappears
The traditional answer to brand trust was the storefront itself. A polished website, a curated app experience, visual identity, customer reviews and carefully designed product pages all helped create the environment in which consumers decided whether a brand deserved their money.
The consumer may now encounter the brand inside a conversation with an AI assistant, stripped of many of the visual and experiential cues companies have spent decades building. What remains is the information the AI can access and interpret, from the brand’s own content and structured product data to third-party coverage, reviews, social conversations and direct platform integrations.
Brand recognition still matters when AI makes the recommendation. Research from PSE Consulting covering 4,250 consumers across the US and Europe found that 89 percent say recognizing the seller’s brand is important or very important even when AI makes the recommendation. Most consumers still apply their own judgment rather than automatically accepting AI’s top suggestion. But AI has already influenced the decision by determining which brands enter the conversation in the first place.
But being recognized inside an AI-generated answer is different from being recognized on a marketplace page. There may be no logo, product photography or carefully curated environment. The first encounter could simply be a brand name inside a recommendation.
That puts more weight on whether AI systems can find reliable information about the brand and make sense of it.
Thumbtack‘s first CMO, Llibert Argerich, has been navigating this firsthand. Rather than waiting for AI platforms to discover its marketplace of 300,000 professionals, Thumbtack built partnerships with providers including ChatGPT and Claude, making its vetted, background-checked professional network available when consumers ask AI assistants about home services.
As Llibert explains on the latest episode of In Other Words,
“The trust, the authority, the expertise really comes across the content. Not just the content on our website, but the content distributed everywhere on the internet. That’s how LLMs decide whether to quote you on a particular question that a customer asks.”
AI systems can draw on a combination of brand content, product information, news coverage, reviews, social media conversations, community forums and other sources when assembling an answer. No single piece determines whether a company appears, but together they shape how much useful and credible information the system has available about the brand.
Content increasingly influences whether a brand is discovered and how it is represented. For companies that have spent two decades investing in search engine optimization, that changes the model considerably. SEO assumed that consumers would search and brands would compete for visibility in the results. AI increasingly compresses that process. The consumer asks a question and receives an assembled answer rather than a page of links to work through.
Llibert frames the opportunity in terms of what that evolution demands.
“The advent of AEO and GEO has really given content marketing and content in general a much bigger role to play,” he says, referring to the emerging disciplines of answer engine optimization and generative engine optimization.
For brands, the commercial question is no longer simply whether customers can find their website. It is whether AI systems have enough reliable information to include them in the answer.
The quality bar for invisible storefronts
Every product description, help article, FAQ page, marketplace listing and piece of third-party coverage contributes information that AI systems may use to understand a brand. International companies have to make that information accessible and credible across multiple markets and languages.
Many global brands have invested heavily in English-language content while coverage in German, Japanese, Portuguese or Spanish remains thinner, slower to update or less consistent. In conventional digital commerce, those gaps could affect search visibility and conversion. In AI-mediated commerce, they can also affect how easily systems discover, interpret and accurately represent the brand in those markets.
Phrase CEO Georg Ell argued at SlatorCon London earlier this year “No one’s interested in a translation. They’re interested in content that sells something, motivates the reader or viewer to do a thing.”
The same content now has to work for both people and AI systems. AI systems need information that is accurate, current and sufficiently rich to answer a consumer’s question. Consumers still need content that sounds credible and appropriate for their market. Simply making an English source available in another language does not guarantee either.
Keeping that information current across markets becomes critical. Product information, pricing, customer expectations and brand terminology all change over time and across markets. If one version of the company’s information is current while another is six months out of date, an AI system may be working from an incomplete picture.
Braze, the customer engagement platform serving more than 2,000 brands in over 70 countries, illustrates the operational challenge. As the company expanded into markets including Japan, where it saw 200 percent adoption growth, it needed customer-facing content to remain consistent as both volume and geographic reach increased. By centralizing its content operations on the Phrase Language Intelligence Platform, Braze entered four new markets in 12 months while maintaining a consistent brand experience across regions.
Companies preparing for a more AI-mediated customer journey need content infrastructure capable of keeping information accurate and consistent as the number of channels, markets and machine interfaces increases.
The price of invisibility
AI-assisted purchasing is still developing, and consumers remain more cautious about delegating some decisions than others. Consumer trust in AI is projected to increase substantially by 2030, while technology companies are steadily expanding what AI assistants can do on a consumer’s behalf.
Brands will need to give AI systems credible, current and machine-readable information wherever customers are asking questions.
This changes where commercial influence sits. A growing share of purchase decisions will happen inside conversations that some brands never enter.
The question of whether your brand is among them is worth answering now, before your customers’ AI assistants answer it for them.
Watch the full conversation
Llibert Argerich, Thumbtack’s first-ever CMO, built an 11-channel growth engine for a marketplace of 300,000 home services professionals, where every zip code is effectively its own market.
He explains why Thumbtack is now embedding those professionals into AI platforms, and what influences whether a brand is recommended when the purchase decision happens inside AI conversations.
