Bruno Ferreira brings a highly practical perspective to one of the most underestimated yet critical areas in eCommerce operations: demand forecasting. With over 16 years of experience working with Amazon Vendor Central across global markets, he has seen firsthand how even well-established brands struggle to manage data, predict demand, and adapt to Amazon’s often unpredictable ordering behaviour.
During the Balkan eCommerce Summit 2026, Bruno shared practical insights into the fundamentals behind effective forecasting, explained why access to data alone is not enough, and highlighted the operational and financial risks companies face when they get forecasting wrong.
In this interview, he offers a clear, experience-driven perspective on how brands can move from reactive decision-making to structured, data-informed planning in an increasingly complex eCommerce environment.
Bruno, during your session “Mastering Demand Forecasting and Navigating Amazon’s Vendor Central Order Volatility”, what were the main takeaways attendees gained from it?
The purpose of the masterclass was to give Amazon Vendors and eCommerce managers practical insights into demand forecasting, including the core principles, key forecasting concepts, and quick wins that can help improve forecast accuracy in eCommerce, with a particular focus on Amazon Vendor Central.
Demand forecasting sounds straightforward in theory, but where do companies usually get it wrong when working with Amazon?
In many cases, eCommerce managers, especially Amazon Vendors, struggle with how to use data effectively for forecasting. They often have access to large volumes of internal and external data, but lack the right internal processes, structure, and analytical approach needed to turn that data into accurate and actionable forecasts.
Vendor Central is known for unpredictable ordering patterns. What causes this volatility, and how should brands adapt to it?
This volatility is often driven by the way Amazon places orders, which can fluctuate significantly based on its own algorithms, stock position, and short-term demand signals.
To adapt, brands should combine Amazon data with their own internal historical data, including ERP data and sales performance across other channels. They should also invest in internal data analytics capabilities to improve visibility and decision-making.
From your experience, what are the biggest risks of poor forecasting, both financially and operationally?
The biggest risks are usually financial and operational.
Financially, poor forecasting can lead to excess stock, higher storage costs, markdowns, and tied-up working capital.
Operationally, it can result in stockouts, lost sales, inefficient supply chain planning, and products sitting in the warehouse longer than expected.
If a company wants to improve its forecasting and reduce dependency on Amazon’s volatility, what is the first practical step they should take?
Companies should gather as much relevant historical data as possible, organise it properly, and use it consistently to create a more reliable forecasting model that is not solely dependent on Amazon’s ordering behaviour.
As eCommerce continues to evolve across technology, customer behavior, operations, and digital growth, conversations like these will once again be a key part of Balkan eCommerce Summit 2027, where industry leaders, brands, and innovators from across the region will come together to exchange ideas, practical insights, and real-world experience.
More information about the upcoming edition is available at balkanecommerce.com