SmartScout strives to provide the most accurate and insightful data for Amazon brands and sellers, addressing the challenges inherent in estimating data on Amazon’s marketplace. Through a combination of advanced methodologies, strategic partnerships, and proprietary systems, we refine data accuracy and offer reliable estimates.
This outlines our core data processes, explaining how our product-level data is aggregated and the steps we take to handle unique challenges, such as child variations and the complexities of the electronics category.
1. Overall Data Calculation Process
Our data collection begins at the product level, where we analyze the product page and look closely at the Best Sellers Rank (BSR). This baseline data is crucial, as the BSR serves as a directional indicator of product performance, helping us generate revenue and market share estimates.
However, since Amazon's own BSR system can fluctuate, we implement additional layers to ensure precision:
BSR Weighted Average
To counteract the daily volatility of BSR, we average it over the last 30 days. This weighted average smooths out potential anomalies, giving a more stable basis for calculating estimates.
Brand, Seller, Subcategory, and Category Aggregation
Using our foundational product data, we then build aggregated insights across brands, sellers, subcategories, and overall categories. This scaling allows us to provide clients with a comprehensive view of performance across various Amazon dimensions.
Triangulation with Third-Party and First-Party Data
We continuously validate our estimates by cross-referencing our data with external sources, including Keepa's API, and integrating insights from large brand conglomerates with direct sales data. This multi-source triangulation strengthens our data accuracy.
Amazon’s “x+ Bought in Past Month” Data
We also factor in Amazon's “x+ bought” insights to refine our sales estimates further. By layering these data points, we achieve more precise volume estimates, ensuring that SmartScout’s insights stay as close to the actuals as possible.
2. Handling Child Variations
Amazon product listings often include multiple variations (or “child” ASINs), each potentially influencing the parent listing’s sales and visibility differently. Managing these variations accurately is key, and our approach includes:
Weighted Sales Estimation Based on Child Ratings
For items with child ratings, we distribute units sold proportionately based on each variation’s ratings, assuming that higher-rated children are likely to contribute more significantly to sales.
Equal Distribution in Absence of Child Ratings
When child ratings are missing—often for low-volume or new items—we distribute sales evenly across variations as a fallback. Though less precise, this approach allows us to maintain data continuity without compromising the overall dataset.
3. Electronics Category Estimation
The electronics category presents unique challenges, primarily due to Amazon occasionally not assigning a category BSR to ASINs here. SmartScout has developed specialized solutions to manage these gaps:
Subcategory Sales Models
Leveraging partnerships with prominent brands in electronics, we’ve created sales models at the subcategory level, allowing us to estimate sales despite missing BSR data. These models are informed by real sales data from our partners, which cover most subcategories in this category.
Performance Benchmarking
By continuously monitoring and comparing our subcategory models with known sales data, we maintain a high level of accuracy, especially in cases where other providers struggle.
Summary and Final Thoughts
In the Amazon marketplace, data estimation is inherently complex and challenging. At SmartScout, we leverage a combination of extensive resources, unique partnerships, advanced sales modeling, and robust internal algorithms to ensure our data remains accurate and actionable. While estimates may sometimes deviate, our multi-layered approach—from weighted averages to cross-referencing with external data—positions us as a leader in data precision. We take accuracy personal. We'd love the specific insights that we might be missing so our engineering team can look at it.
Clients can feel confident that SmartScout is constantly refining its processes to stay ahead of industry standards. With our commitment to data accuracy and our comprehensive approach, we strive to provide insights that are as close to actuals as possible, delivering a clear and reliable view of Amazon’s complex landscape.
