Innodata: DPrice Dynamic Pricing System

The main articles are:

• Business Intelligence Definition
• Big Data

2022: Registration of the DPrice dynamic pricing system for developers

On May 11, 2022, the Russian developer Innodat announced that it had registered its own dynamic pricing product in the Unified Register of the Ministry of Telecom and Mass Communications of Russian programs for electronic computers and databases. DPrice is a system for ensuring transparency in setting prices for real estate in development. The solution has predefined templates, flexible settings and customization options.

Illustration: images-cdn.cian.site

DPrice includes a set of services for automating labor-intensive business processes, for example, indexing prices for real estate objects, determining the starting price, analyzing plan execution and sales rates.

The solution operates on the basis of a mathematical algorithm that predicts the likelihood of selling real estate. The platform offers the ability to adjust the parameters for calculating the rise in prices for them depending on the requirements of the developer and the market situation. The system is easily customizable to the customer's individual needs.

The DPrice platform was based on artificial intelligence technologies (Artificial Intelligence), including optimization and machine learning methods (Machine Learning). The solution is applicable for risk mitigation in cases where the company does not have automated market assessment tools. In addition, it helps to improve the effectiveness of decisions made.

"DPrice is a specialized tool for developers, taking into account the specifics and many factors necessary for selling and maintaining positions in the real estate market. We have been implementing projects in the field of demand forecasting and pricing management for a long time and this made it possible to include current practices in the functionality of the product. We already have several ongoing projects, and we see that the demand for DPrice is only growing. Several releases and scaling of the system to other industries are planned for the near future, taking into account the specifics of the business of each of them, "- comments by the executive director of Innodata, Alexander Sergienko.

2018

Opportunities. Principle of operation. Tasks to be solved

According to February 2018, the intelligent pricing system is designed for automated accurate forecasting and balancing of prices and tariffs.

Principle of operation

The system works as follows: three blocks of information are formed on a daily basis for system users.

The statistics block provides for an interactive report, including indicators related to sales dynamics, price level, customer activity, etc. It is envisaged to receive reports of different degrees of aggregation, ranging from the total indicators of the company to the level of a specific real estate object.

The forecast block provides for a daily updated probability of selling a property in the next month. Prediction results can be aggregated up to the level of the type of apartments and even up to the level of risers of a particular section in the project.

The recommendation block includes daily updated values for the value of changes in prices for real estate objects, types of apartments, risers. In this case, recommendations can be customized by the user, changed dynamically depending on the possibilities for changing prices that add up for a particular real estate object based on the results of analyzing the input components of the system.

Opportunities

• formation of a basic model for forecasting the dynamics of pricing, identification of the main visible and hidden factors affecting the dynamics of development;
• construction, optimization and monitoring of the business model;
• high-precision adjustment of parameters and variables affecting its operation;
• enriching the model with additional data.
• assessment of the probability of the transaction;
• calculation of daily forecast for each transaction,
• grouping of results, as well as price control based on actual demand for the object: if the forecast for actual demand exceeds the planned one, then it is possible for more frequent increases in value.

At the same time, the process of forming recommendations and comments on pricing management is automated.

Tasks to be solved

The main business tasks that the intelligent pricing system solves:

• maximizing revenue
• increase in sales without increasing costs
• prompt response to events affecting pricing in a highly competitive market
• pricing dynamics forecasting
• consideration of the number of influence factors,
• minimizing the influence of the "human factor."

Intelligent Pricing Announcement

Innodata introduced an intelligent pricing system on February 14, 2018. According to the developers, the solution will be in demand among developers in construction, retail, transport and logistics companies and large service organizations with a constant line of services.

The high probability of error due to the "human factor" in manual forecasting of supply and demand makes the pricing process more complex, implying long-term, time-consuming and expensive research. Many factors must be taken into account in order to ensure that sales are efficient when creating an optimal price. The intellectual pricing system from Innodata will help to do this relatively quickly.

Using Big data technologies and neural networks, the company's specialists have developed an approach to post-processing of the received data, which allows you to achieve high efficiency from the constructed mathematical models, reduce error and increase the interpretability of the result.

The system allows you to predict sales and the best period for price changes, reduce human labor costs for the process of value formation due to the optimization of the business process, provides support in real time.

The solution model is balanced and provides for about 200 variables, while taking into account seasonality factors, both internal and external determining factors are used, such as, for example, fluctuations in currency quotes.

The results obtained by the system are achieved using self-learning algorithms of a mathematical model (for example, using XGBoost). The analytical model is built on the basis of several developed methods. The model takes into account historical data. For the final completion of the training, the model is trained in real time. 90% of the accuracy of the transaction falls on the period that is reflected in the model. Provided that 85% or more of the data is provided, the model correctly predicts the statistics of expected transactions.

The main effect of using the intelligent pricing system in the company's business architecture is to achieve the main goal - to maximize revenue without increasing costs, - said Maxim Sytnikov, Product Owner solutions, Innodata. - It is difficult to overestimate the effect on business from using the system: first of all, it is maximizing revenue without increasing costs, increasing the level of competitiveness, stimulating demand, increasing revenue, fine-tuning value fluctuations by predicting a future transaction, checking the appropriateness of recommendations and elasticity of demand in real time, increasing additional profits through a flexible approach to data. And as a pleasant bonus - optimization of labor costs, for example, the analytical department, which supports pricing in manual mode, as well as an increase in the speed of decision-making from a day to several minutes.