Analysis of the Thesis: "In 2030, 80% of Companies with an Extensive Catalog of Offerings Will Have Implemented AI Search"

SWOT Analysis

Strengths

• Better user experience and higher conversion rates: AI search (e.g., conversational chatbots or NLP-based search engines) can deliver more accurate results and more personalized recommendations to users. Customers more often find what they are looking for, which translates into higher conversion rates. Statistics show that shoppers who use the on-site search are 2 3 times more likely to make a purchase and generate several times more revenue than those who do not use it. For example, on Amazon the conversion rate rises from about 2% to 12% (a sixfold increase) when a user utilizes the search function. AI search also reduces frustration from irrelevant results it minimizes "no results" situations, which are currently a cause of customer churn (according to Forrester, as many as 68% of users abandon purchases due to search frustration). Improving search quality is thus a major advantage that boosts customer satisfaction and sales. Gartner research confirms that 80% of customers prefer to shop with retailers offering personalized search experiences, and platforms with AI-based search engines achieve up to a 25% higher level of customer satisfaction and engagement.
• Ability to handle a broad product assortment and natural language queries: Companies with dozens or hundreds of products often struggle because traditional filtering or a basic search function cannot keep up with the diverse queries of users. AI search can understand customer intent expressed in natural language even if the query is colloquial or imprecise and find the appropriate products. For example, a language model can handle a query like "I need something warm for winter training" and find sports jackets, whereas a traditional search engine might not make that connection. This is a huge advantage for a large assortment because AI can handle both broad and very specific queries (e.g., "women s waterproof trekking shoes, size 38") by understanding context rather than just keywords. Thanks to this, customers can find matching offers faster, which builds their trust in the site.
• 24/7 customer service and reduced staff workload: AI-powered chatbots can serve as the first line of customer contact around the clock, answering questions about products, availability, prices, or order status at any time. This increases user convenience and allows companies to handle more inquiries without having to proportionally expand their customer service team. AI can automate up to 80% of all customer interactions (according to forecasts for the coming years), which represents a huge relief for employees. Already in 2021, AI handled 15% of interactions in customer service, and the chatbot industry was growing at around 30% annually. As a result, companies can respond to customer needs faster, improving their experience and loyalty, while simultaneously cutting operational costs. It is estimated that chatbots allow companies to save up to 11 billion USD in costs and 2.5 billion hours of work per year by replacing part of traditional customer service.
• Competitive advantage and market necessity: Implementing AI search is becoming an increasingly important competitive necessity. Companies that adopt AI search more quickly can outpace competitors by offering better service and personalization. Conversely, ignoring this trend can mean being left behind. More and more managers realize this 87% of retailers have already implemented AI in at least one area of their business, and 80% of retail executives anticipate that by 2025 their organization will be applying AI to automate processes. Similarly in the B2B sector: 81% of wholesalers and B2B companies are already experimenting with or using AI in their operations. These figures suggest that intelligent search will become a standard within a few years both in B2C e-commerce and on B2B websites because companies do not want to risk losing customers due to outdated solutions.

Weaknesses

• Immaturity or technological limitations in specific applications: Despite rapid progress, current AI solutions are not flawless. A chatbot or NLP search engine may not always correctly understand an unusual query or might return an answer with an error (a so-called AI "hallucination"). In industries requiring precision (e.g., medical online pharmacies, or financial), an incorrect answer can mislead users. Training an AI model on a specific domain (for example, mechanical parts in a B2B wholesale context) can be difficult it requires feeding it appropriate data and continuous refinement. Simple chatbots with a limited conversation script have limited usefulness, while modern LLM-based chatbots (large language model-based) can sometimes give overly "gimmicky" or imprecise answers if they are not well tailored. This means not every company can immediately implement AI search that truly delivers value it requires data maturity and constant oversight of result quality.
• Implementation costs and uncertain ROI for smaller companies: Although more off-the-shelf solutions are appearing on the market, building and integrating an intelligent search engine can be costly especially when a company has unique requirements. According to analyses, the cost of developing a chatbot can range from 5 thousand to 500 thousand USD, depending on the scale and complexity of the solution. For small and medium-sized enterprises (even if they have an extensive product range), the high initial cost and the need to maintain the AI system can be a barrier. If the implemented solution does not quickly deliver a noticeable increase in sales or savings, the ROI (return on investment) will be called into question. This could discourage some companies especially those that are more conservative or operate on thin margins from being among the first 80% to implement AI search.
• Need for proper data and integration: The effectiveness of AI search depends on the quality of product data (descriptions, attributes, photos) and on user behavior history. Companies that do not have well-structured data about their offerings or do not track user behavior may not harness AI s full potential an NLP search engine without data will function like a regular one, and a chatbot without integration to a knowledge base may give only generic responses. This highlights the issue: many companies may find that before implementing advanced search, they have to invest in cleaning up their data, integrating product databases, CRM systems, etc. Lack of back-end readiness is a weakness it creates an additional entry barrier. In fact, 32% of retail employees cite lack of appropriate infrastructure as a barrier to implementing AI. If a company doesn t resolve these technological shortcomings, then the chatbot alone won t work miracles.
• Possible mixed customer reactions: Although overall acceptance of AI is growing, still around half of people have concerns about AI in customer service. Some users may prefer traditional methods (e.g., manually filtering products or talking to a human) especially if they previously associated chatbots with limited, frustrating interactions. There are still complaints about a "lack of human touch" for example, a bot won t sense a customer s emotions or handle an unusual request. If AI search is not well designed, it can irritate users (e.g., when it persistently offers to chat and the customer would rather browse on their own). These factors mean that not every interaction will be positive. A company must monitor feedback negative experiences with AI can scare customers away. In studies, 34% of consumers rated the impact of AI on service as somewhat negative (and 37% were unsure), which shows there is still room to improve the quality of these solutions to win over skeptics.

Opportunities

• Rapid development of AI technology and lowering of implementation barriers: The coming years will likely bring another leap in AI development language models are getting better and better (e.g., GPT-4 and subsequent generations, as well as competing models from Google, Anthropic, Meta, etc.), and at the same time many specialized tools for e-commerce are emerging. The range of ready-made AI search as a service offerings and plugins for popular e-commerce platforms is growing. Major providers (e.g., Shopify, Salesforce, Microsoft, Google) are already integrating AI into their products for online stores. This means it will become increasingly easy to implement intelligent search even without one s own data science team the technology and cost barriers will decrease thanks to standardization of solutions. For example, there are already tools that allow you to deploy a sales chatbot powered by your own database in just a few clicks, and e-commerce engines are beginning to offer built-in semantic search with NLP. This availability creates an opportunity for even medium and smaller companies to roll out AI search en masse within 5 years, because it will become as fundamental as having a website was in the past.
• Shift in consumer preferences toward AI interactions: Society is becoming accustomed to AI in daily life. In 2025, 58% of consumers report a preference for using AI tools instead of a traditional search engine when shopping online (up from 25% in 2023). Furthermore, the majority of people who have interacted with chatbots rate the experience positively 69% of customers report good experiences with chatbots, and 68% appreciate the speed of service via AI. Younger generations entering the market as consumers are used to conversational interfaces (Siri, Alexa, ChatGPT, etc.) and expect natural, dialogue-based interaction on shopping or service websites as well. Moreover, customers want an increasingly personalized approach 71% of consumers expect personalization of offerings and communication and AI search can provide this personalization by analyzing behavior history and preferences. All this creates a favorable climate: companies have the chance to meet these needs and stand out, and customers are increasingly willing to use such solutions, which increases the economic rationale for implementing them.
• Omnichannel applications and expanded functionality: AI search on a website is not just a product search engine it is often a multi-functional assistant whose capabilities can be expanded. There is an opportunity in integrating a chatbot with other channels (messengers, mobile apps, voice assistants) so that a customer can naturally search for and purchase products anywhere. Forecasts say that by 2026 over 80% of companies will be using AI in their mobile applications, which suggests that shopping assistants will also make their way onto smartphones. In addition, AI can merge functions for example, simultaneously acting as a search engine, an advisor (recommending products that match queries or purchase history), and even finalizing a transaction within a chat window. A further development could be voice commerce voice assistants helping to search for and order a product (already today 60% of consumers say they have used voice commands for online shopping). Companies have an opportunity to use AI to create a cohesive omnichannel experience whoever is first to do this well will earn the loyalty of convenience-seeking customers. In short, AI search is a starting point for a broader digital transformation of customer interaction.
• Operational efficiency and data utilization: Implementing AI search is also an opportunity to better utilize the vast amounts of data that companies with extensive offerings already have (product data, customer queries, transactions). Intelligent algorithms can discover hidden demand patterns for example, which product features are most often searched for together which can influence assortment or marketing decisions. Furthermore, companies that introduce AI more often see increases in revenue and decreases in costs. As many as 69% of retailers report revenue growth attributed to AI, and 72% see a reduction in costs thanks to AI. Improving search can, for example, decrease the number of returns (since the customer finds exactly what they need), or reduce call center load thanks to automated bot responses all of which represent opportunities for cost savings. On a macro scale, companies that adopt AI faster can build a reputation as innovators, attracting customers who value modernity. The "AI" trend also helps in marketing it can be a PR asset to show that you have an intelligent assistant on your site. In summary, the growth opportunities thanks to AI search are multi-dimensional: from customer satisfaction, to sales, to internal efficiency.

Threats

• Technical barriers and risk of implementation failure: From a market environment perspective, there is a risk that the technology will not live up to expectations in the planned timeframe. Despite the hype, some companies may encounter technical difficulties when attempting to implement AI search for example, integrating with existing IT infrastructure might prove complicated or time-consuming. If many companies run into failed attempts (e.g., a chatbot giving poor answers and causing a drop in customer satisfaction), a cooling of enthusiasm and a slowdown in adoption could occur. It must also be remembered that AI needs data companies operating in regions with strict regulations (e.g., GDPR in Europe) may face legal restrictions on collecting user data or sending data to AI providers in the cloud. AI-related regulations (e.g., the proposed EU AI Act) may impose additional compliance requirements, which could discourage some companies that fear penalties or liability for AI actions. Security and privacy are real threats: 53% of retail managers point to data security issues as the main obstacle to implementing AI. A data leak incident via a chatbot or a piece of bad advice (for example, health-related advice given by an online pharmacy s bot) could not only harm customers but also generate negative publicity, undermining trust in such solutions across the industry.
• High costs for advanced solutions and uncertain returns: Although a decline in barriers was mentioned, there remains a financial risk not every company will immediately find the budget for AI search. Particularly globally, in less developed markets, investing in AI may take a back seat to more urgent business needs. In studies, 39% of retail company executives cited financial risk as an obstacle to implementing AI. If an economic slowdown occurs or other spending priorities arise, the wave of AI investments may be smaller than forecasted. Moreover, companies that do not see quick successes from the pioneers may wait longer to implement it themselves (preferring, for example, to invest in cheaper improvements). The cost-related threat is also tied to maintaining the solution AI models need to be updated, trained, and one must pay for infrastructure (e.g., fees for using models in the cloud). Smaller companies might worry about dependency on technology providers (e.g., AI service prices increasing in the future). These uncertainties could cause some companies to take a wait-and-see stance, as a result of which the overall adoption rate in 5 years might be lower than the optimistic 80%.
• Lack of competencies and internal resistance: Implementing AI search is not just the purchase of a tool, but also a change in processes and a need for new competencies (data analysis, model training, AI monitoring). Lack of the right talent or knowledge is cited as a barrier 43% of retail employees believe that a lack of awareness/expertise is a reason for failures in implementing AI. Companies may fear that their IT departments will not be up to the new technology. Additionally, customer service employees might resist, fearing for their jobs (task automation). Company culture and readiness for change is a "soft" factor, but it can significantly affect the pace of adoption. If management or staff do not believe in AI (e.g., due to prior bad experiences and there have been such cases; company CEOs are more likely than employees to cite negative AI experiences as the cause of an implementation failure), the project can get stuck in the testing phase. Thus, the threat is that human and organizational factors will slow down the spread of AI search despite the availability of the technology.
• Alternative solutions or changing consumer behaviors: Finally, one should consider whether in five years customers will still be searching for offers on company websites. The emergence of powerful models like ChatGPT or conversational search engines (Bing AI, Google Assistant, etc.) means that a user might begin searching for products globally through an AI assistant instead of visiting many separate sites. If, for instance, ChatGPT with a shopping feature becomes common and effective, some companies may choose to integrate with such platforms rather than develop their own chatbots. This is a scenario where the role of AI in shopping grows, but it may somewhat reduce the motivation of some entities to invest in their own AI search (especially smaller ones that can "plug into" large marketplaces or big tech companies bots). Nevertheless, most large retailers will still want to maintain control over the customer experience on their own site however, the threat remains that consumers will change their habits and companies will have to chase after them into new channels instead of focusing on their own website.

Customer Needs and AI Search Functions

Modern users of sites with a large number of offerings have specific needs, and intelligent search can address them. Below are the main customer needs and how AI search fulfills them:

Quickly finding the right product or offer

Users expect that in a sea of products on a site, they will instantly find the ones that match their criteria. Traditional search engines often return too many results or none at all if the query is unusual which is frustrating and leads to page abandonment (12% of users immediately go to a competitor when they don t find what they want). AI search improves this aspect by understanding natural language and the context of queries. It can recognize the user s intent even with a descriptive question, and account for synonyms, typos, or colloquial terms. As a result, the number of zero results and irrelevant suggestions decreases, and the customer saves time. This translates into a better experience as mentioned, consumers using an effective search engine are much more likely to make a purchase. In short, AI search fulfills the basic need of efficiently reaching sought-after products even in a very large catalog.

Ability to ask questions in a natural way (ease of use)

Many users don t want to think about choosing the right keywords or filters it s more convenient for them to type or speak a question in their own words. Up until now, site search engines often required entering specific phrases, which can be a barrier (for example, a customer looking for a "green recliner" had to manually click through the furniture category, color filter, and features instead of simply asking). AI search lets users search as if they were talking to a salesperson. They can enter a query in a full sentence or ask a follow-up question (in the case of a conversational chatbot) the system will understand the dialogue flow. This meets the need for an intuitive experience: even less tech-savvy or older users can manage because the interaction resembles a conversation. For many people this is a big convenience they don t need to know specialized terms or the site s structure to find something. As a result, the user experience on the site is better and the entry barrier is lower, which expands the pool of satisfied customers.

Personalized results and recommendations

Today s customer expects that the offering will be tailored to their needs no one likes to dig through hundreds of items; they prefer to immediately get a few relevant suggestions. AI search can combine the search function with a recommendation engine, learning the user s preferences. Based on browsing history, purchases, or even the tone of the query, an intelligent system can prioritize results that are most likely to meet that person s needs. For example, a chatbot can remember that a user previously searched for vegan products and, on the next query about cosmetics, display those with appropriate ingredients first. This kind of personalization makes the customer feel served individually, like in a brick-and-mortar store with a helpful salesperson. According to Gartner, 80% of shoppers prefer retailers that offer a personalized search experience as cited earlier. AI search is exactly what enables this: it satisfies the need to be quickly understood by the system and to receive "tailor-made" suggestions. This is not only convenient but also a factor that increases loyalty a customer is more likely to return to where the offering seems exactly tailored to them.

Comprehensive help in decision-making (guidance)

For more expensive or complex products (electronics, cars, real estate, financial services), customers often need additional information and advice before they decide. Traditionally, this role was filled by a human a consultant on chat or a call center agent. However, an AI chatbot is increasingly filling this gap, conducting a conversation with the user like an advisor. It can answer technical questions (since it has access to a product knowledge base), compare several products based on given criteria, and even suggest the best choice based on the needs expressed by the customer. For example, in real estate a chatbot can filter listings according to very detailed buyer preferences and present only those properties that meet the conditions, while also explaining the differences. This consultative AI feature addresses a need that many customers signal 95% of online shoppers admitted that their pre-purchase online experience would improve if they had human help available. LLM-powered chatbots in particular are striving to fill this gap, offering human-like help at scale. From the customer s point of view, they receive lightning-fast answers to their questions (without waiting for a human agent) and support in making a choice which builds confidence before purchase. This AI search feature can significantly raise the conversion rate for products that require deliberation, because it reduces the customer s doubts (all questions can be immediately asked and resolved in a dialogue with the AI).

Immediacy and 24/7 availability

The modern user is impatient accustomed to everything happening immediately on the internet. The need for an immediate answer (whether it s a product inquiry or a problem) is therefore very strong. AI search especially in the form of a chatbot addresses this perfectly, because it responds in a fraction of a second at any time of day or night. When a customer at 1 AM has a question about the availability of a shoe size or wants to check order status, a bot can provide the information immediately, whereas they d have to wait until morning for an email response from support. This continuous availability significantly improves the customer experience, giving users the feeling that the company is always at their disposal. Research shows that 68% of customers appreciate the speed of service via chatbots, and 62% actually prefer to talk to a chatbot over a human if it means getting an immediate answer. This is a surprising shift in preferences, showing how important response time is. AI search meets this need for speed and constant readiness which translates to higher satisfaction and a lower chance that the customer will give up (because they didn t get an answer in time). For the user of sites with large offerings, this means greater convenience: they can make a purchase or get help exactly when they want, without time restrictions.

In summary, the features offered by AI search relevance, natural communication, personalization, guidance, and immediacy align with the main expectations of modern consumers using services with a large number of products. AI-based solutions are designed precisely to eliminate the pains of the past (such as irrelevant search results or long response times) and to create an experience close to ideal: fast, simple, and with a sense that the offering is tailored to the user. That s why we see growing acceptance customers increasingly want to use such tools because they see real benefits (e.g., time savings and better-matched offers). By implementing AI search, companies address the key needs of their clientele, which usually translates into mutual gains: customer satisfaction and better business results.

Threats and Challenges in Implementing AI Search

Despite the many advantages, there are specific threats and challenges that could hinder the widespread implementation of AI search in the market. It is worth analyzing them, because overcoming these will determine whether the thesis of 80% adoption in 5 years comes true:

Technological challenges (accuracy, data, and oversight)

Ensuring high-quality AI responses requires the right technology and continuous improvement. Language models are prone to errors: they might misinterpret a question or generate an inappropriate answer. In the case of product search, a mistake (e.g., recommending an item the company doesn t sell, or providing outdated information) can hurt the customer experience. Companies thus need to invest in training models on their own data and in mechanisms to verify AI responses, which is a technological challenge. Moreover, AI search works well only when it has access to complete and up-to-date data gaps in product descriptions or delays in updating stock levels can cause even an intelligent system to provide incorrect information. Implementing AI also entails the need for monitoring and moderation (e.g., to ensure a chatbot doesn t give inappropriate responses). Not every company has the know-how to meet these challenges right away. A lack of knowledge or AI experts is a barrier pointed out by many business owners half of surveyed companies feel they lack the proper skills and fear AI mistakes. There is a risk that some entities might get burned on their first implementations (e.g., they ll deploy a bot that starts "hallucinating" responses, which alienates customers and management). Therefore, working out best practices is technologically crucial otherwise the pace of adoption may slow down.

Cost and integration with existing systems

Implementing AI search often requires integration with multiple systems (product database, transaction system, CRM, etc.), which can be complicated and costly. The financial investment is a significant barrier, especially for companies outside the top tier of the largest players. While global retail chains have budgets for innovation, a mid-sized retailer or wholesaler might delay implementation until the technology becomes cheaper. There is a risk that due to costs and integration difficulties, initially only the largest companies will implement AI search, and smaller ones will wait, which would undermine reaching 80% of all firms in such a short time. The aforementioned cost threshold (up to a few hundred thousand USD) for tailor-made solutions shows that wealthier players will have an advantage in early adoption. Moreover, integration carries the risk of disruptions for instance, errors in linking AI with the order database could lead to bad experiences (the bot says a product is available when in reality it s out of stock). Without solid integration and testing, such situations can occur. This means the implementation timeline might be longer than optimistic assumptions companies will cautiously connect AI to critical systems in stages. In summary, high upfront costs and complex integration are threats that could slow down or limit the scale of implementations in the short term.

Privacy concerns and regulatory compliance

AI search typically relies on analyzing user behavior and personalization, which involves collecting customer data. In an era of growing privacy awareness and tightening regulations (GDPR, ePrivacy, the proposed AI Act), companies may fear the legal consequences of improper use of AI. For example, a chatbot that has access to personal data is subject to strict requirements for protecting that data. In studies, over half of retail managers pointed to data security and privacy issues as the main problem in AI implementations. AI-related incidents can have serious consequences: if an algorithm unjustifiably discloses customer information or gets hacked (e.g., someone introduces malicious training data), it can lead to not only financial penalties but also loss of reputation. This threat can especially affect industries such as health (pharmacies), finance, or automotive, where customer data is sensitive. Furthermore, if governments introduce additional licenses/certifications for using AI on websites (which is not unlikely, given regulatory discussions), bureaucracy may slow down implementations companies will wait for clear legal guidelines. In the extreme case, some companies might decide they prefer not to risk it and stick with the traditional search until AI is fully "safe" from a legal standpoint.

User acceptance and trust

Although growing acceptance of AI among customers was noted earlier, there is still a segment of users who are distrustful or unhappy with such solutions. The human factor on the customer side can also be a threat: if in the early phase of deployments there are a few high-profile blunders (e.g., a chatbot giving absurd shopping advice that becomes viral on social media), it could lower the overall level of consumer trust in such tools. Even now, 34% of customers rate AI in service negatively their opinions can influence others. Older customers in particular like direct human contact and may boycott chatbots, feeling that the company is "fobbing them off with a robot." If a large group of customers avoids or disables AI features (e.g., closes the chat window), then from the company s perspective the point of implementing them diminishes. Furthermore, malfunctioning AI can undermine trust even among enthusiasts for example, if a chatbot suggests a product completely unsuited to the needs, the customer might deem the AI search useless and not use it next time. Thus, companies must ensure a positive first impression and continuous improvement, otherwise the threat is a loss of user trust, which on a macro scale could stall the trend of AI search proliferation (customers might force a return to human service or simpler interfaces).

In summary, while the technological, cost, integration, regulatory, and user acceptance challenges are real, they do not appear to block the trend, only moderate it. Most of these challenges are solvable over time (e.g., costs come down, competencies grow, regulations become clearer), but they are reasons why the adoption rate could be lower than the theoretically possible maximum. Therefore, when analyzing the thesis of 80% of firms implementing AI search, one must keep in mind that the optimistic scenario assumes overcoming the above barriers which is possible, but not automatic.

Summary and Probability Assessment

Basing on the above analysis, we can synthesize that the trend of adopting AI in search and customer service on sites with extensive offerings is clearly upward. However, achieving as much as 80% penetration within 5 years depends on maintaining favorable conditions and overcoming certain barriers. On one hand, market data and company statements indicate a very high likelihood of this thesis being realized. Already today, about 40 60% of companies (depending on the B2B/B2C segment) use chatbots or similar solutions, and the vast majority of the rest have them in their plans. For example, in retail 42% of retailers already use conversational AI in customer service, and another 48% plan to implement it by 2026. If these plans come to fruition, in about two years nearly 90% of retailers will be using AI in this area which makes the 80% in 5 years assumption very likely. Also globally, over 75% of enterprises plan to implement AI solutions in the next 5 years (World Economic Forum data), which confirms the overall direction. The B2B/wholesale sector, where product catalogs are enormous, also shows a strong desire for automation 81% of such firms are already experimenting with AI. On the consumer side, a shift has occurred: most customers show readiness to use AI in shopping (preference for AI-powered search, positive opinions of chatbots), so user-driven demand will push companies to implement it. In short, businesses want it and customers want it, and the technology is maturing these are ideal conditions for rapid adoption.

On the other hand, one cannot ignore the risk factors. The degree of fulfillment of this forecast may vary by company size and region. Market giants will almost certainly implement AI search (many already have); however, the question remains whether smaller entities with large catalogs (e.g., a local car dealer with a hundred cars, a mid-sized wholesaler) will also adopt AI search en masse in such a short time. If costs and lack of expertise continue to be seen as significant obstacles, some companies may push their investments beyond the 5-year horizon. Additionally, any negative incidents or disappointments (e.g., unsuccessful implementations at competitors) could slow others enthusiasm the technology must maintain the market s trust. Finally, there is a segment of companies that may feel their current search is good enough and do not feel customer pressure to change these might account for the remaining ~15 20% that do not adopt.

Probability Assessment: Considering the above arguments, I estimate the likelihood of the thesis being realized at around 75 85%. In other words, there is a very high probability (approximately 80%) that in five years the majority (~80%) of companies with extensive catalogs of offerings will have implemented AI-powered intelligent search on their websites. This forecast is supported by both current trends and company declarations (even aiming above 80%), as well as strong economic (benefits of AI) and social (customer expectations) factors. About 20% uncertainty remains due to the aforementioned threats mainly concerning smaller players and potential implementation hurdles. Nonetheless, the direction of change is clear: AI is becoming a standard in the search experience, so the thesis appears well-founded. There are concrete industry examples confirming this trend (from e-commerce to real estate to B2B), and the pace of innovation suggests that in 5 years an intelligent chatbot or semantic search engine will be as ubiquitous as today s ordinary search field if not more so. Therefore, we can be fairly confident that roughly 80% of companies meeting the thesis criteria will indeed achieve this level of transformation in the given timeframe. (Probability estimate: ~80%.)