We believe every data set contains a story, with clues about customers, markets, shifts in what is happening, and more. With so many directions a business can go in, it is important to read these clues, and understand any signposts hiding in your data. At Soundience, we are experienced and skilled in finding trends, relationships, and other patterns in your data, specializing in attractive and insightful visual presentation of results to make them real.
Approaches we use
We use a wide variety of approaches, usually in appropriate combinations, to yield insights:
- Standard statistical techniques such as regression
- Rich visualization, to help find patterns that are not apparent from the numbers
- Machine learning techniques, such as cluster detection and dimensionality reduction
- Social network analysis, to find networks of relationships between people or organizations, and identify key nodes or weak links in these networks
- Our own MarketPulse tool, for allocating impact to different underlying drivers
- Neural networks, for mapping more complex higher-order relationships between inputs and key outputs
Making sense of multiple data sources
We work with many sources of data, including company databases, public data sets, web server logs, customer service records, client billing records, or data collected from surveys.
We are experts at cleaning and combining data, to merge multiple data sets so that new relationships can be found. We often present the cleaned-up data in a navigable web interface, to help us and our clients explore the data.