IBM On Business Analytics In Banking With @LaurenceTrigwel

I was on an IBM analyst call this morning on analytics in banking, featuring Laurence Trigwell, who is responsible for IBM’s business analytics strategy for the Financial Services industry.

In response to a number of challenges that banks (and other financial institutions) are facing, he identified three key initiatives for these companies:

  • Creating a customer-focused enterprise
  • Optimize risk and compliance
  • Increase flexibility and streamline operations

Obviously, analytics are pretty important to all of this, and IBM continues to focus on building their analytics portfolio with acquisitions such as Algorithmics and capabilities such as Watson. As with all of their portfolios, the IBM analytics portfolio contains a number of different products, making it a bit difficult for customers (and probably IBM sales reps) to wade through the potential solutions; however, for the purposes of this presentation, Trigwell has organized them around the three key initiatives listed above.

  • For creating a customer-focused enterprise, you need performance management, business intelligence and predictive analytics in order to gain insights from customer data to “provide the most profitable offer to the right customer at the right time”. Although this sort of “next best action” view of analytics for customer focus is part of the picture, I think that it misses the point completely: the stated goal (in quotes above) is really focused on the enterprise, not the customer, since it has enterprise profitability at the core. Not that profitability isn’t important, but if you’re going to be truly customer-focused, you need to have some goals stated as actual direct customer benefits. It might be the same thing, since providing an offer that is best suited to a customer’s needs means that they are more likely to accept it, which in turn results in great profitability in a probabilistic sense, but using language that is actually customer-focused when you’re talking about becoming a customer-focused enterprise is important to shift corporate culture.
  • For optimizing risk and compliance, you will need risk management, governance, performance management, compliance and business intelligence capabilities. I don’t disagree with this, and analytics (as well as other things, such as BPM) can definitely turn risk management and compliance from just overhead to a competitive advantage, if you do it well.
  • For increasing flexibility and streamlining operations, you will need performance management, predictive analytics and business intelligence capabilities for identifying process bottlenecks and operational inefficiencies. This is really about displaying and reporting on what’s happening, not fixing the problems, so not a sense-and-respond view of analytics or how it fits into process management in general.

In each of these three scenarios, the capabilities that he is suggesting that you need are actually IBM products: from three to five products for each scenario if you want to go all out. They have bundled some of these into industry solutions, but it’s not clear what the level of integration between the products is to create a seamless solution, or if they’re only integrated on the PO. Clearly, IBM needs to do some rationalization of their analytics portfolio to reduce the number of products (as they need to do in some of their other portfolios), although their usual strategy is to allow acquired companies to run pretty much untouched for quite a long time before starting to merge products. That strategy is good for the sales teams and the executives of the acquired company, but not necessary good for customers who have to deal with an increasingly large and bewildering array of products that overlap in functionality. The case studies that he discussed typically used only one product, either Algorithmics or Cognos, although one used both Cognos and SPSS, so I’m not sure that there’s much of an appetite for multiple analytics products applied to the same initiative.

In all cases, he talks about analytics as identifying/reporting on issues/situations, but not much on how analytics need to fit together with other systems to make it all happen. He touches on a bit of this with some of his case studies, but it would be great to see the analytics for banking shown in the context of other IBM products that can really make the three initiatives real, such as CRM, BPM and ECM.

Since banks and other financial institutions are my main customer base for consulting, and they’re all IBM customers, it will be interesting to see how they roll out some of the newer IBM products and solutions, especially in combination, in the years to come.

2 thoughts on “IBM On Business Analytics In Banking With @LaurenceTrigwel

  1. I seriously question the actual practical benefits of these products and their approaches. There is a mathematical theory and rigor behind it but that does not mean the data and the model that they come from produce anything of value.

    You are absolutely right that even in case you gain some valuable insight from analytics you still need to turn this into an action that will change the future outcome into a desirable direction. Is this improved outcome a benefit for the customer or for the business bottom line?

    People ignore that mathematical rigor does not turn fiction into fact but it leads to what Nassim Taleb calls NAIVE INTERVENTION.

    http://isismjpucher.wordpress.com/2013/01/05/naive-intervention-part-1-from-antifragile-to-models-behaving-badly/

  2. Max, I’m just starting on Antifragile and will let you know how it affects my thoughts on all of this!

    Analytics tools in the hands of people who don’t understand analytics are often not of much use. It’s not just about understanding the mathematics behind it (although someone needs to have an idea of that), but about the meaning of what is being analyzed and the results. It’s very common for people to get caught up in in the “pretty graphs” aspect of analytics, especially just after they see a fancy vendor demo, but those graphs need to show something of relevance for the business, and also provide some sort of actionable information.

Leave a Reply