- pass_by collaborates with NeuralProphet, developed by Stanford Ph.D. candidates, to build the most advanced, AI-powered foot-traffic model for retail.
- Retailers can now access AI-powered retail insights instead of relying on low coverage, inaccurate time series models.
- This new methodology is live in pass_by’s revolutionary suite of retail products, in some cases doubling correlation to ground truth data.
- This update provides our customers with the most accurate insights into retail-based consumer behaviour on the market.
In today’s fast-paced retail industry, data-driven decision making has become the norm for retailers of all sizes. The issue though, is that data signals can still be inaccurate at scale. This poses a problem to an industry with increasingly tight profit margins, and less and less room for error. Historically, retailers have relied on their own sales data to inform these decisions, but are increasingly looking for third party data solutions to help provide context and depth to their planning procedures.
pass_by offers world class foot traffic analytics that help retailers make the right call on several parts of their value chain, including supply chain management (by signaling demand months in advance), assortment planning (where our foot traffic predictions would be paired with stock management tools), and labor optimization (where store level visits predictions can improve staff scheduling). And now thanks to a recent update to the way we process our data to provide insights to retail businesses, our analysis is more accurate than it ever has been before.
We are thrilled to announce the release of an upgraded version of pass_by’s foot_traffic data methodology, collaborating with NeuralProphet to incorporate cutting-edge time series modeling. This update marks a significant milestone in enhanced retail data analysis, revolutionizing the accuracy of our visits data and helping to empower businesses with invaluable AI-powered retail insights.
What is NeuralProphet?
Developed by PhD candidates at Stanford University and many open-source contributors world-wide, NeuralProphet is an advanced time series forecasting library that harnesses the power of artificial intelligence. This library employs both traditional time-series methods and neural networks to analyze historical data to provide interpretable forecasts.
While time series data has applications in some industries, widespread usage has only recently been implemented. However, as businesses, particularly in the retail sector, have increased the amount of data they collect from customers, as well as widening the scope of data they collect, existing methods for analyzing data no longer match the amount of data that there is to be analyzed. As a result, attention has been turned on to how time series analysis in retail can be improved, ensuring retailers get the most accurate insights from what is available to them.
Since the beginning of 2023, pass_by has been working in partnership with Oskar J. Triebe, PhD student at Stanford University, in order to expand NeuralProphet’s time series capabilities to fit the needs of our methodology. As an early adopter of this advancement, pass_by is uniquely placed to benefit from the improvements offered by the NeuralProphet framework. Our application demonstrates that NeuralProphet is a capable framework for forecasting deployments in the retail industry.
It has been a joy to work with Alfonso from pass_by to extend NeuralProphet for retail forecasting use cases. It’s motivating to see them improve and simplify their large-scale forecasting, making use of modern yet interpretable methods. We are grateful for their open-source contributions, such as the addition of global-local hybrid modeling, which will benefit many other users.Oskar J. Triebe
By incorporating the NeuralProphet update into our methodology, we will now provide businesses with AI-powered retail insights, demonstrating a notable improvement from the time series models used by the majority of the analytics in the retail industry.
Improved Data Quality for Enhanced Accuracy
As pass_by has moved to the framework provided by NeuralProphet, we have seen a dramatic increase in the accuracy of the data insights that we offer our customers. With the retail data analytics upgrade, we now have a clearer picture of seasonality and trends that affect the number of visits to a store, which provides a tighter correlation to the ground truth visits that are provided to us by our partners.
The framework provided by NeuralProphet helps us to assess the number of visits a location gets more accurately. By helping us to assign a visit to a particular location at a specific time of day, helping to make sense of the vast amounts of mobility data we process, we can provide retailers with a higher level of accuracy.
As outlined throughout this article, accuracy is the most important aspect for retailers when it comes to decisions on their next data partner. This proprietary model, as well as the increased accuracy it offers, will help pass_by provide even more retailers with the accurate data they need to power their decision making processes. With tighter margins than ever before, retailers are now demanding the highest levels of accuracy, and more and more data providers are hoping to provide the retail industry with the accuracy they need.
The benefits of this approach are:
- Improved daily and monthly visits correlation in respect to ground truth data
- Improved modeling around seasonality/seasonal events
- Increased ability to extract meaningful patterns from noisy or incomplete raw data
- Faster modeling and access to predictive data
- Vastly improved coverage of areas outside of cities where raw data counts are lower
Retail Data Optimization: Why this matters
Analytics in retail is exploding. Big data analytics in retail generated $4.85b in 2020 worldwide, and is forecast to have a 23.1% CAGR from 2023 to 2028. As greater numbers of data providers look to serve the retail industries desire for data, it stands to reason that these data providers need to be able to provide best in class data solutions.
A member of pass_by’s advisory board, Eric Sprunk, is a previous COO at Nike. In an interview, Sprunk has stressed the importance of demand planning, stating “We have to anticipate demand. We don’t have six months to do it. We have 30 minutes”. Representing just a part of the value chain, demand forecasting is one of many stages that can be optimized by accurate data. As pass_by adopts the NeuralProphet framework, the increased accuracy available will be able to help retailers with this, and many other parts of their operations procedures.
Impact on the Retail Industry: Unlocking the Power of Data Analytics
The integration of NeuralProphet features into pass_by’s foot traffic data methodology brings a multitude of benefits to the retail industry. When looking at foot traffic data, retailers face numerous challenges in data analysis, including missing data, outliers, and noisy signals. NeuralProphet tackles these challenges head-on, leveraging advanced time series analysis techniques to handle irregularities and seasonality in a much more effective way than was previously possible. Here are a few ways in which the increase in data accuracy is helping retailers.
Improved Store Visits Accuracy
NeuralProphet’s cutting-edge technology significantly enhances accuracy. By reducing errors, retailers can minimize the risks associated with overstocking or understocking, resulting in improved operational efficiency and cost savings. According to an NRF study, overstocking and understocking represents as much as $1.75 Trillion in lost revenue every year, which is why accurate data is needed to facilitate informed decision making and provide a solid foundation for strategic planning.
Enhanced Retail Trend Analysis
NeuralProphet’s powerful algorithms capture complex patterns enabling retailers to identify and analyze seasonality and trends in real-time. Retailers can proactively respond to changing consumer preferences, adapt their marketing strategies, and gain a competitive edge. The ability to uncover trends quickly and accurately empowers retailers to make data-driven decisions and stay ahead of the curve.
Data-Driven Decision Making
With the integration of NeuralProphet, pass_by’s methodology enables retailers to make informed decisions across various aspects of their business. From pricing and promotions to supply chain management and resource allocation, predictive analytics optimizes retail operations and aligns them with consumer demand. By leveraging NeuralProphet’s forecasting models, retailers can capitalize on data-driven insights to boost efficiency and drive success.
The Future of Predictive Analytics for Retail
The retail industry is experiencing a profound transformation, and data analytics and forecasting are at the heart of this evolution. pass_by’s collaboration with NeuralProphet to integrate features developed together into our retail data methodology demonstrates a seismic shift in that evolution.
By empowering retail businesses with the most accurate data available, we can help them to optimize operations, identify emerging trends, and make informed decisions. By harnessing the full potential of enhanced retail data analysis, retailers can enhance customer experiences, drive profitability, and stay ahead in the competitive landscape.
Ready to unlock the power of data-driven retail insights? Get in touch today to explore how our advanced retail data analytics and NeuralProphet methodology can propel your business forward.
Founded in 2023, pass_by is a cutting-edge geospatial intelligence company dedicated to revolutionizing the industry with AI-driven solutions. Recognizing the challenges and gaps in conventional geospatial tools, pass_by has embarked on a mission to provide businesses with accurate, real-time, and actionable products that reflect genuine ground realities. pass_by aims to be the gold standard in geospatial intelligence, serving diverse sectors including retail, real estate, and finance.