We use cookie technology. Cookies
allow us to provide you with up-to-date information about your preferences and ensure that the site operates correctly. You can block cookies in your browser settings, but then you won't be able to interact with our site properly.
allow us to provide you with up-to-date information about your preferences and ensure that the site operates correctly. You can block cookies in your browser settings, but then you won't be able to interact with our site properly.
The path to purchase has changed: it's now tech-driven
‘Consumers are increasingly finding answers directly in AI-generated results, and there are fewer and fewer clicks through to the websites of brands, online shops and publishers.’
‘Brands may need to focus more on shaping consumer perceptions even before purchasing decisions are made. They also need to help shape first- and third-party content that large language models (LLMs) rely on, in order to position their products effectively in AI-enabled environments.”
McKinsey surveyed nearly 5,000 consumers across five countries (March–April 2026). The report is extensive, but the main (and expected) conclusion is that the moment of intent has shifted. Consumers still use search and enter search queries, but they make their decisions at a different stage, according to experts at TON OP company. Now, AI Overview responds before the click.
‘Brands may need to focus more on shaping consumer perceptions even before purchasing decisions are made. They also need to help shape first- and third-party content that large language models (LLMs) rely on, in order to position their products effectively in AI-enabled environments.”
McKinsey surveyed nearly 5,000 consumers across five countries (March–April 2026). The report is extensive, but the main (and expected) conclusion is that the moment of intent has shifted. Consumers still use search and enter search queries, but they make their decisions at a different stage, according to experts at TON OP company. Now, AI Overview responds before the click.
What the report reveals
McKinsey uses the term ‘generative engine optimisation’ - reputation in an environment where decisions are made without you. The AI’s response is compiled from dozens of sources, and only 1 per cent of these come from the brand’s own website. The rest is written on your behalf. 99 per cent comes from Reddit, YouTube, retailers and reviews. The consumer is presented with a patchwork of different texts. And if the sources contradict one another, the party for whom the answer is important (the seller, the shop, the brand) may simply be ‘left out’ of the response.
What should businesses be doing today?
Follow the exact same path that your customer takes, recommend the experts at TONOP doo ltd. Enter potential search queries into a search engine. Ask Gemini, ChatGPT, Copilot, Perplexity, Claude or Grok questions about your brand - and compare the answers with what’s on your website. The gap between the two is now more dangerous than a competitor offering a lower price.
Repeat the search in the languages available on your website’s interface (you’re most likely interested in the target audience that speaks these languages). Keep two lists: where you’re not present and where you appear weaker than your competitors. Make a separate note of the complaints that appear most frequently in negative reviews.
The more precisely the product information is structured, and the more clearly the specifications, benefits and terms of purchase are described, the easier it is to analyse - not only for humans, but also for search engines and AI tools, which are increasingly involved in the decision-making process. You can use the TON OP programme to plan and analyse your product range.
Below is a brief checklist to help you assess how well your brand is prepared for the new model of search and decision-making.
Repeat the search in the languages available on your website’s interface (you’re most likely interested in the target audience that speaks these languages). Keep two lists: where you’re not present and where you appear weaker than your competitors. Make a separate note of the complaints that appear most frequently in negative reviews.
The more precisely the product information is structured, and the more clearly the specifications, benefits and terms of purchase are described, the easier it is to analyse - not only for humans, but also for search engines and AI tools, which are increasingly involved in the decision-making process. You can use the TON OP programme to plan and analyse your product range.
Below is a brief checklist to help you assess how well your brand is prepared for the new model of search and decision-making.
Speak the customer’s language
People are increasingly entering long search queries. Therefore, the text on your website should be written in a style that mirrors everyday speech. Use the phrasing that customers actually use, as recommended by the experts at TON OP company.
Help AI understand your product
Compile a list of 10–15 questions that customers ask most frequently in search queries, via chat, over the phone or in reviews. Provide short and clear answers to them. Explanations of limitations, common mistakes when making a choice, and practical advice on using the product or service are particularly useful.
Neural networks find it easier to analyse structured information. Therefore, it is best to present specifications, deadlines, warranty terms, delivery and returns policies, compatibility and usage scenarios as clearly as possible. The easier it is to compare options, the higher the likelihood of being included in the AI’s response.
Customers are not only looking for the benefits, but also want to know about the drawbacks and any issues that might suddenly arise. Include responses to common objections, comparisons with alternatives, cost calculation examples, case studies and detailed terms of cooperation. The fewer uncertainties a customer has, the easier it is for them to make a decision, according to experts at TONOP doo ltd.
Compare, don’t just describe
Address objections in advance
Back up your claims with facts
Promises are less effective than evidence. Use figures, results, completion times, real-life examples, limitations and potential risks. Specific details help both people and AI systems to evaluate the proposal.
Make the next step easier
Once they’ve received an answer, people don’t want to waste time searching any further. A calculator, a configurator, quick product selection, transparent delivery and returns policies, and a short order form all help to ensure you don’t lose the customer at the final stage.
Offer add-ons based on the situation
Additional products and services work best when they appear to be a logical extension of the purchase. Experts at TON OP advise against pushing them on customers; instead, explain why the customer might need the next product or service.
Give customers the chance to sort things out for themselves
Users who are used to AI expect an immediate response. It is therefore important that most queries can be resolved without having to call a customer service representative: through clear instructions, a help section and transparent company policies.
FAQs are becoming an essential feature
For each category of products or services, it is useful to prepare answers to real customer queries: regarding delivery, selection, compatibility, care, warranties and returns. A brief FAQ section comprising three or four questions is sufficient for a product page. It is precisely these sections that often serve as a source of information for AI-powered search.
What should you do?
The main question today is no longer how to rank higher in search results. It is more important to understand what a person will find out about your company or what you offer if they ask an AI about it.
The moment of intent has indeed shifted. Decisions are increasingly being made before the customer even lands on the website. That’s why optimising for AI-powered search has become an essential routine. Whether you run an online shop, a corporate website, a blog, a media platform or an online service, your site must be user-friendly and understandable to the algorithms that extract information and generate ‘short’ answers. Everything else - what others write about you - you can only accept, but not control. This is precisely what reputation management is all about nowadays.
TON OP software helps analyse market trends. Specialists at TONOP doo ltd have developed a modern, intelligent application that helps automate most business processes: management, business planning, marketing, sales, accounting and monitoring.
The moment of intent has indeed shifted. Decisions are increasingly being made before the customer even lands on the website. That’s why optimising for AI-powered search has become an essential routine. Whether you run an online shop, a corporate website, a blog, a media platform or an online service, your site must be user-friendly and understandable to the algorithms that extract information and generate ‘short’ answers. Everything else - what others write about you - you can only accept, but not control. This is precisely what reputation management is all about nowadays.
TON OP software helps analyse market trends. Specialists at TONOP doo ltd have developed a modern, intelligent application that helps automate most business processes: management, business planning, marketing, sales, accounting and monitoring.
Our Contacts
Do you have any questions? Send us an request via this form!
TON OP DOO LTD
Address: 8 BORBA STR., PLOVDIV, BULGARIA, 4021
Mail: info@tonop.bg
Address: 8 BORBA STR., PLOVDIV, BULGARIA, 4021
Mail: info@tonop.bg