Spontaneous Feedback Index (SFI) – Methodology
This page explains the methodology behind the Spontaneous Feedback Index (SFI), including the data sources, analytical framework, quality controls, and standardization principles used to transform spontaneous public feedback into a comparable customer satisfaction score.
The objective of the SFI is to provide a consistent way to analyze customer satisfaction based exclusively on spontaneous public feedback collected across multiple digital channels.
1. Purpose of the SFI
The Spontaneous Feedback Index (SFI) was developed to measure customer satisfaction using spontaneous public feedback rather than traditional surveys or questionnaires.
The index analyzes what customers voluntarily share online about their experiences across review platforms, marketplaces, and social networks.
By consolidating large volumes of spontaneous feedback into a standardized score from 0 to 100, the SFI helps organizations:
- understand overall customer satisfaction;
- analyze customer perception across multiple public channels;
- compare performance against benchmark reference datasets;
- evaluate changes in customer satisfaction over time.
2. Scope: What the SFI Measures
The SFI is currently calculated at the company level.
Yellow Tokens consolidates spontaneous feedback associated with the companies included in a project, regardless of whether the feedback originates from review platforms, social networks, locations, products, or other public channels linked to those companies.
The objective is to provide a single company-level measure of spontaneous customer satisfaction that reflects overall public perception across multiple sources.
3. Data Sources Used in the SFI
The SFI uses only public, spontaneous feedback.
Examples of supported sources include:
- Review Platforms: Google Reviews, TripAdvisor, Booking, Amazon, and similar sources.
- Social Networks: Facebook, Instagram, TikTok, X, YouTube, and other public social platforms.
- Public Customer Feedback: reviews, comments, recommendations, complaints, replies, and other experience-related content.
When a platform provides explicit ratings, Yellow Tokens uses the original customer ratings as part of the calculation.
When ratings are not available, AI models analyze the feedback and estimate a rating equivalent only when the content clearly represents a customer experience. Content unrelated to customer experience is excluded from the calculation.
4. Core Dimensions of the SFI
The SFI is built from multiple satisfaction signals derived from spontaneous public feedback.
-
Spontaneous Customer Satisfaction
Measures overall satisfaction levels derived from public ratings and customer feedback. -
Spontaneous Loyalty Signals
Evaluates positive and negative advocacy patterns identified in spontaneous customer feedback. -
Negative Experience Signals
Measures dissatisfaction, complaints, and other indicators of customer friction.
Together, these dimensions provide a balanced representation of customer satisfaction based exclusively on spontaneous public feedback.
5. Normalization and Standardization
Because spontaneous feedback originates from different platforms and formats, all satisfaction signals are standardized before contributing to the final index.
This process allows ratings, inferred ratings, satisfaction indicators, and dissatisfaction indicators to contribute consistently to the final score.
The resulting methodology produces a comparable score ranging from 0 to 100, regardless of the original source of the feedback.
6. Minimum Volume Requirements
To improve reliability and reduce the impact of small sample sizes, Yellow Tokens applies minimum feedback volume requirements before calculating the SFI.
These requirements help ensure that the resulting score reflects meaningful patterns in customer feedback rather than isolated experiences.
Organizations should interpret results with additional caution whenever feedback volumes are limited, even when minimum quality thresholds have been met.
7. Benchmark Comparison
One of the primary applications of the SFI is benchmark analysis.
Yellow Tokens can compare an organization's SFI against reference datasets maintained within the platform, helping contextualize customer satisfaction levels relative to similar market environments.
Benchmark comparisons are intended to provide directional context rather than official industry averages.
By comparing an organization's SFI against benchmark references, decision-makers can better understand whether customer satisfaction appears stronger, weaker, or broadly aligned with relevant market conditions.
8. Final Score and Interpretation
After standardization and aggregation of all satisfaction signals, the SFI is expressed as a score between 0 and 100.
Higher scores indicate stronger customer satisfaction and more favorable spontaneous public perception.
Yellow Tokens uses the following interpretation ranges:
- 85–100: Excellent (A+)
- 75–84: Strong (A)
- 60–74: Stable (B)
- 40–59: At Risk (C)
- 0–39: High Risk (D)
These interpretation bands are designed to provide a simple framework for understanding customer satisfaction while preserving comparability across organizations and benchmark datasets.
9. Limitations and Considerations
While the SFI provides a valuable view of customer satisfaction based on spontaneous feedback, it should be interpreted within its proper context.
- Spontaneous feedback is not a probabilistic sample of an entire customer base.
- Different industries generate different volumes of public feedback.
- Customer feedback may be influenced by seasonality, operational changes, or exceptional events.
- Public feedback tends to reflect stronger positive or negative experiences more frequently than average experiences.
For this reason, the SFI should be viewed as a measure of spontaneous customer perception rather than a replacement for operational KPIs, customer surveys, or other business metrics.
10. Why the Exact Formula Is Proprietary
Yellow Tokens openly documents the principles, data sources, and analytical framework used to construct the SFI.
However, the exact weighting structure, calibration logic, and scoring formula remain proprietary.
This approach helps preserve methodological consistency, prevent manipulation, and maintain comparability across datasets, industries, and time periods.
Similar approaches are commonly used in reputation indexes, rankings, financial indexes, and other composite indicators where transparency of methodology is balanced with protection of the underlying model.
Yellow Tokens periodically reviews the methodology to improve accuracy and ensure alignment with evolving customer behavior and digital platforms.
See Your SFI in Action
Build a project in Yellow Tokens and discover how customers publicly perceive your business through spontaneous feedback collected across multiple digital channels.
Analyze your SFI, compare results against benchmark references, and gain a clearer understanding of customer satisfaction at the company level.