GPT Party 3.0. AI-powered sales: creating personalized customer experience GPT Party 3.0, the largest Russian-language networking event dedicated to artificial intelligence, took place in Silicon Valley on March 9-10. More than 450 people gathered at Plug and Play to meet leading experts, entrepreneurs and investors, discuss the latest trends in artificial intelligence and gain practical knowledge. The panel discussion “Sales with AI: creating a personalized customer experience” discussed which artificial intelligence-based tools are already used in sales and how sales AI will develop. The speakers of the panel discussions were: Andrey Nokhrin, Albert Golukhov, Slava Orekhov, Alex Zhuravlev; the moderator of the discussion and the author of questions was Alexander Volchek. As a professional investor, what companies in the sales AI space are you paying attention to? Alex Zhuravlev: “We have already invested in two companies that specialize in using artificial intelligence in sales. The first company is developing a platform that integrates with CRM, mail and other internal services, and then uses an AI recommendation engine to analyze data and select quality leads. The platform learns over time, offering increasingly relevant leads, and even conducts outreach communications, initiating meetings or calls with potential clients. The second company is digital workers. This is a large company that started its activities with sales and created a platform that really works. Their product integrates with many internal systems, including CRM and calendars, and acts as a real employee, interacting with clients. They already have more than 100 clients and a queue of 7,000 companies, so now there are difficulties connecting to their platform due to high demand It is important for us to invest in already working products, given that there are now a lot of startups in the sales industry seeking to use AI for quick profit.” You do a lot of B2B sales in the corporate sector. Have you implemented any new approaches to sales in your business or have you seen any real cases of using AI in sales? Albert Golukhov: “I am not involved in software development and implementation, but I can share my thoughts from a B2B perspective. In both of our companies, we are currently implementing tools based on artificial intelligence. This is a really exciting technology that allows us to optimize processes and work with leads, as already mentioned. We use software that already has integration with CRM and other necessary systems. I am not sure about chatbots, I have not yet seen any successful cases of their use.” How do you use AI to manage real estate sales, given the significant traffic in this industry? Andrey Nokhrin: “Our main activity is in fintech, where we develop projects based on machine learning for issuing loans. We have used AI to evaluate real estate and make decisions about issuing loans. We have products where we help clients in the US in the competition for real estate by offering financial support when buying a home. Another product we have developed gives clients the opportunity to receive additional funds to buy a new home, provided that the old one is sold within six months. We use machine learning to make decisions about providing housing and loans, as well as to assess the market value of real estate.” Has the emergence of ChatGPT or other language models impacted machine learning? What progress have you made in machine learning over the past year? Andrey Nokhrin: “It is precisely the technologies of large language models, such as ChatGPT, that have not had a significant impact on us. However, machine learning as such, thanks to the increase in data volume and long-term use of this technology, has begun to produce more accurate forecasts and reduce errors in real estate valuation. For example, if four years ago the model allowed an average error of 8-10%, now this error has decreased to 4%. This helps to make more accurate forecasts of housing prices. The technology becomes more effective over time and with an increase in the amount of data. In our previous startup, we sold solutions based on machine learning and artificial intelligence for business. We had over 2,000 clients. We used social media data, mainly Facebook, to predict who might sell their home. We analyzed data about Facebook users, examining parameters such as their age, marital status, the presence of children, and even changes in their profiles. These signals allowed us to effectively predict whether someone would sell their home. Although it sounds a little strange, we knew about the intentions to sell the house before the owners themselves realized it.” The deal cycle time is, say, one to two months. The company has 20 to 50 sales managers, each of whom is engaged in lead processing. Every day, 20 potential leads come in. What specific measures can be proposed to this company to guarantee an increase in the efficiency of sales managers? Slava Orekhov: “There is actually quite a lively discussion about what the future of artificial intelligence in sales will be. Basically, there are two directions of development. On the one hand, it is possible to improve the efficiency of current salespeople. Imagine that you interact with a friendlier person who understands your needs better. One of the key trends on which this direction is based is democratization. This means changing your buying experience, as you will have a personal sales assistant. Although the main goal of the business remains to sell you more, in the end it will be beneficial for you, as you will receive more relevant and useful information about products and services. But we are betting on a different school of thought: we focus on replacing humans with technology. This means that our technology allows for the automation of work performed by a living person, although not in all scenarios. It will take many more years for a computer to be able to completely replace a human. However, access to technology and the ability to replace a human with a computer will lead to the emergence of salespeople in areas where they did not exist before. For example, we have successfully worked with companies that had not previously used cold outreach or other methods to attract customers. We were able to help them use artificial intelligence to attract new customers, and it turned out to be effective.” How exactly can AI be implemented in sales now? What exactly can we consider and apply at the moment? Slava Orekhov: “For us, the most effective and correct option is to work with the top of the funnel, that is, lead generation. Why? Because artificial intelligence has a significant room for error. We cannot replace a person, but we can automate certain processes depending on the sales cycle. If, for example, your sales cycle is two months, as mentioned earlier, significant human intervention is usually not required. It will be enough to find a suitable client and offer him your goods or services.” What prospects do you see in the development of CRM systems for artificial intelligence? And how do you assess them from an investment perspective? Alex Zhuravlev: “At the moment, it is not advisable to consider the direction of CRM for artificial intelligence. Firstly, the CRM market is in a very saturated state with many players. This segment is more needed for sales managers, but if artificial intelligence takes over sales functions, the question arises: is a CRM system needed then? AI is capable of processing and analyzing data within itself, so CRM is more likely to become a reporting tool than for operational management. Now we do not see any point in investing in the development of CRM using AI, since the market is not ready for this yet. Another problem is that the transition from one CRM system to another is a long and complex process for both large corporations and small and medium-sized enterprises.” Albert Golukhov: “It seems strange to me that no one has mentioned this before regarding the future. It’s not just about working with leads, but about personalized work with potential clients based on their social networks and so on. Using artificial intelligence, it is possible to create sales scripts adapted to each specific person, based on their behavior, habits, social status, etc. This approach makes the presentation of products or services more personalized.”