Doug Cutting, one of the co-founders of Hadoop, recently said that “Google is living a few years in the future and sending the rest of us messages.” That comment resonated with me when I met up with the team behind Rage Frameworks, a small firm located near me in the Boston area. If you look up Rage, you’ll see it listing itself as a Business Process Automation vendor, and as a “Cool Vendor” from Gartner in the BPM (Business Process Management) area. Yet, while business process modeling is important, this is no glorified workflow and integration vendor. Rather, the firm has a big data / analytics platform called Real Time Intelligence (RTI) that is being used to build multiple native solution areas in such areas as credit risk, market analysis, and even securities pricing forecasts.
For example, the solution has a highly predictive bankruptcy modeler, and I was fascinated by the demo of the software that showed how its securities risk index statistically correlated with stock prices on a lagged basis a few weeks out (i.e., it appears to predict stock price). No, we didn’t demand proof of back testing, but it was funny because our sister site MetalMiner has some metals forecasts that do this. But we’re not in the business of trading, and neither is Rage. Of course, there definitely are some very interested parties kicking the tires.
The secret sauce here comes from the platform. Basically, it is an expanding knowledge base that serves up its insights as a service. The knowledge base is fed by over 18,000 data sources, including not just credit bureaus and other similar databases (e.g., structured and semi-structured content), but also the mass ingestion of unstructured content from a slew of web crawlers, RSS feeds, and “patented extraction components.” Then, the data has to auto-magically convert all this potential data noise into the sentiments and “signals” that indicate various forms of company risk (i.e., supplier risk), which can be aggregated into a composite RTI supplier risk score (available out of the box, but can be “tuned” and trained to a client’s particular needs).
At this point, you might be saying to yourself:
- This seems massively complex, complicated, and expensive
- I’m skeptical that this magic black box can truly extract and infer predictive supplier risk information from such data feeds and unstructured text.
- How does it deal with private suppliers vs. public suppliers?
- This also seems like OpenRatings (now part of D&B) reborn, so how is it different?
- How does this solution deal with supplier risk beyond financial risk score modeling?
- How do I make the solution tailored to my specific risk management needs rather than a generic model?
- And how about the broader picture of supply chain risk? The domains become vary specialized.
- Similarly, how can I use it for supplier/market intelligence beyond risk?
- How does this thing tie to my internal systems?
- Is anybody actually using it and getting value from it?
Let me answer the first and address the rest in a follow-up piece.
In terms of the cost and complexity issue, the supplier risk solution is merely served up as a simple subscription cost per supplier per year. No minimums. If you want to monitor a few dozen (or even a few hundred) critical suppliers, this is a great option.
But, while the supplier risk product is straightforward commercially, I am only scratching the surface. Basically, it’ll let you be your own D&B. And the real power of this lies in the “ontologies” that let you customize the solution to the semantics particular to your business domain. It’s like training a spend analysis tool to auto-classify text to a custom commodity taxonomy, but on a much broader scale. Now, imagine that the attributes describing that domain are tied to the structure data models of complex value chains. The world becomes very interesting – especially in the hands of service providers. And we haven’t even addressed the AI-based category management development service that uses such “Natural Language Processing on steroids.” It’s actually more than NLP, and I’ll dive into this in the next piece, but let me just say that this intelligence platform will be truly disruptive in the market, especially in the hands of large practitioners and service providers who can tune it to their strategic supply intelligence needs.
You will definitely want to stay tuned for the next editions in this series where we'll dive into the solution, some case studies, and the many implications that this firm will have on the market, especially for market intelligence service providers and for intelligence / analytics related solution providers.