Big Data is a much newer term. I tend to think it's a new name for something people have been doing for a long time: analyzing data to create valuable insights.
The characteristics of data analysis have changed with the development of information, we see exponential growth in volume, we see variety, the speed and changes in the truth of the data;
The so-called "Four V's of Big Data" (volume, variety, velocity, veracity). The same phenomenon applies to Credit Management.
You can always find someone these days who can predict something for next to nothing. Predicting bankruptcies of potential customers and suppliers. But make no mistake, qualitative predictive insights (predictive scores/insights) are not a commodity.
How can you judge the quality of these predictive scores? There are no hard criteria except to wait and eventually do an analysis to see if the predictions are actually correct. This is not an option, because you don't want to take unnecessary risks with your company's profits and losses, do you?
My answer is: just like anywhere in life, there are no free meals. Quality has a a price, the cost of data collection is a long-term investment, an investment in data quality (timeliness, completeness, accuracy, ...) and in creating insights. Because predictive insights that are based on incomplete data, incorrect data, untimely data or data that is of low quality for any reason are too expensive: they not only lead to failure to predict real bankruptcies, but also to too many "false positives": you will be deterred from making the right choices because an incompetent party is predicting bankruptcy of (and preventing business with) companies that are financially sound.
This is my advice also for credit management: data quality is the key to success, predictive insights to manage your credit risk are of high value. Whose data quality do you trust the most?