The level of “location sensitivity” was no less important for any numbers, but also for the initial. Today this approach is no longer sufficient and is rapidly transforming. The sales potential of the chosen location is very high, and the competition is very intense. Retail is moving into a phase where intuition does not know, but rather changes its nature - it becomes supported by great data. Here geomarketing comes to the forefront. For VARUS, this transformation has already become part of a systematic approach to the development of the framework. The company is gradually collapsing into a format that is based on great data. At the center of this approach are models that allow one to assess the potential of a location even before opening a store.. Nowadays, for VARUS, modeling is no longer just a selection of ready-made market instruments, but directly, which the company develops as a power expertise. Today we are working on a new model for forecasting the potential of a location. It is formed by a large team of analysts from the financial department, the development department and the Project Office from the acquisition of external leading experts in geomarketing.. This approach makes it possible to integrate real sales data, customer behavior, strategic development of measures and the best market geoanalytics practices. These models combine dozens of parameters: from pedestrian and automobile traffic to population size, income level, type of forgetfulness, behavioral characteristics of clients and the competitive environment. The greatest benefit - analyze and factor. The location works like a system, not like a bag of showmen. This is how a whole picture of a location is created. As a result, the company takes away not abstract statements about the object, but a specific forecast that can be used on an equal basis with investment decisions. One of the key tools in this process is heat maps with overlaid data balls.. They allow you to literally “boost” the market: where is the potential, and where is the limitation. Regions, places and surrounding areas are assessed based on a set of factors and their “intensity” is determined by color – from the most promising areas to the quiet ones, which are inappropriate for the store. This approach changes the very logic of location search: instead of analyzing hundreds of options, the company focuses more on those points that have already passed the first data selection. As the process becomes more technologically advanced, the basic principles of retail become immutable. The main “currency” of a location is still traffic, not just traffic, but traffic that converts into purchases. People continue to enjoy the convenience of a store that sells handouts, often outperforming cheaper alternatives. In geomarketing models, particular respect is given to the division of flows: stationary, which is formed by shopping centers and office workers, and transit, which is caused by transport hubs or pedestrian routes. No less important and a competitive environment. It ceases to be merely a streaming factor and becomes part of the modeling of the location's potential. The presence of competitors does not always reduce the potential. She often confirms that the location has already been formed as a point. Now you can point out other details - the layout of the entrance, the visibility of the whiskey or the direction of the flow - can indicate which store itself is receiving the client. Geomarketing clearly shows the boundaries of possibilities. And from a basic understanding, without any confirmation, the store has no sense regardless of forecasts. This includes logistical accessibility, the availability of necessary infrastructure, the presence of “dead zones” or above-world competition. An important indicator of the effectiveness of this approach is the accuracy of prediction. VARUS focuses on recovery in the range of 10–15% and gradually collapses to an accuracy of 80–85%. What is important is not the indicator itself, but the process: after the results are returned to the store, they are compared with the forecast, and the model is completed. This creates a closed cycle in which the new skin moves forward with the decision to make a decision. However, most foldable models do not fit absolutely everything. There are factors that are difficult to formalize - local origins, cultural characteristics, or the level of trust in a new brand in a particular region. The remaining decision is lost to people. Data can enhance the examination, but not replace it. The next step in the development of this approach for VARUS is the integration of a new object into the CRM system. We see this as an important stage of digitalization in the process of developing a framework, which will help improve the efficiency of location searches, provide better solutions to the expansion of stores and better synchronize the work of teams,. Such integration can support the continuous development strategy of the company and create scaling measures even more systematically and efficiently. The role of Big Data and AI in retail is soon to grow less. Models become more advanced and accurate, and processes become more automated. But the key principle is unlikely to change: technology will be deprived of tools, so the responsibility for decisions lies with experts. Geomarketing in this context is becoming not just a technology, but a new kind of retail. And those companies that begin to use massive amounts of data, modeling and analytics with deep understanding of customer behavior will gain not only better locations, but also a strong competitive advantage.