CRM and ERP News

I have decided to apply predictive analytics to see if James is going to post a blog entry today. (James tell me his first post this morning "wasn't much of a post", so we'll assume for our purposes that one doesn't count.)

Unfortunately, I did not have sufficient empirical evidence to apply standard statistical methods. Instead, I've had to rely on expert judgment to create a scorecard that will allow me to assess the likelihood that he will post.

I considered a variety of factors that probably come into play, such as whether or not he's nearing his book deadline, whether his mummy is visiting from England, and whether he can go outside to play cricket in relatively warm weather. Each of the potential values has been "binned", or sorted into ranges of values, and I have assigned a value to each bin: the larger the score, the more likely he is to blog!

When I take the conditions on any given day, and sum up all the associated scores, I have found that if the number exceeds 90, then he will blog. That means, I also have a simple rule: "If James's blog likelihood score is 89 or less, then I must post a blog entry, to fill the void."

Today, I found that his blogging score DID NOT exceed that value - primarily because his book is due IMMINENTLY. So, it looks like it's going to be up to me for much of this week. So, here goes...

I attended a TDWI webinar today by Wayne Eckerson on Predictive Analytics. This presented information from a recent survey TDWI conducted of its members (here's the report's executive summary and you can get the report here).

Now TDWI, and Wayne, are approaching this from the point of view of Business Intelligence/Data Warehousing and that gives them a certain perspective. Like me, Wayne and TDWI want practitioners to move beyond capturing the data to analyzing and using it. Indeed Wayne describes predictive analytics as a form of BI. I am not sure I buy that (after all there are serious differences between BI and EDM) as I consider predictive analytics to be more than "uncovering relationships" it is about creating executable models (I tried to explain this here) and I like this definition of analytics from Gartner). Indeed the presentation often seemed to focus more on what I would call "predictive reporting" than on real, executable predictive analytics. That said, I do agree that prediction builds on reporting, analysis and monitoring though analytics and EDM can also add value to monitoring, not just build on it, as I noted in this article on BAM and in this one on shifting performance management into action.

Some other thoughts:

• Wayne refered to Blink, a great book I reviewed here and that I, like Wayne, highly recommend
• It was interesting that the survey showed lots of exploration and development of predictive analytics (64%) but only 21% implemented partly or fully.
  Remember the survey was of BI/DW people and it has to be said that in many of our customers the BI/DW and predictive analytics group are completely separate! Nevertheless there is a clear and growing interest in predictive analytics amongst BI/DW practitioners.
• Of those adopting it only about a third are really measuring and iterating models
  This worries me as adaptive control, the constant monitoring and improving of analytic models, has historically been critical to success in this area. We say Enterprise Decision Management specifically to highly this aspect of decision automation.
• Wayne gave a nice list of target applications with marketing, forecasting, churn and fraud topping the list.
  It was interesting that budgeting and forecasting scored so highly, and campaign management for that matter. These do not lend themselves to embedded analytic models in operational systems so much as to predictive reporting. Indeed, Wayne gave some examples and a number were of predictive reporting while others were more EDM-like, embedding the predictive analytics into operational systems.
• Wayne noted that the credit card business is a massive adopter with big teams of analysts.
  This is one of the industries that seems to have separate analytic and BI/DW teams.
• Wayne noted that the ROI of predictive analytics is much higher that for other BI applications.
  This matches my experience me but then I can be cynical about the ROI for BI. While the ROI for EDM/predictive analytics is higher, the investment is also higher.
• I found it interesting that TDWI responders moved their analysts into their information management team as this is not what long-time users do.
  Clearly there need to be different paths to adoption for those who are adding predictive analytics to BI from those who adopted analytic modeling sometime ago (like banks, airlines).
• Data interaction is an important component in time spent and that increased the value of analysts being close to data.
  No argument from me
• Scoring and deployment seemed not to cause many problems from the respondents
  This sounds too easy for me and again I suspect that comes from the amount of predictive reporting involved here.
• Wayne is right that data mining/predictive analytic tools are better and that they reducing time to build models
  I disagree that they make it easy for someone without deep skills, however, as they still need real skill to use properly. They do not make it possible for non-modelers to build good models.
• Data warehouses can be really useful for modeling but there are issues in terms of how it is organized and set up
  For instance data with a time component is often stored by absolute time/date in a warehouse where an analytic modeler wants data in terms of "days since acquisition" or "months before cancellation". Extending a data warehouse to support analytic modeling is possible but a non-zero effort. As Wayne suggests you will likely need to create an analytic data mart.
• Wayne mentioned PMML and, while it is a good standard (and one Fair Isaac supports), it does not cover characteristics well (the pre-cursor calculations for a model) and this can be a real problem
  Wayne's examples were about predictive reporting - embedding models into reports - where this may be less of an issue

Overall I was glad to see that Wayne is promoting predictive analytics as a next step for BI/DW adoption. BI/DW professionals need to look beyond predictive reporting, however, and understand the power of rules and analytics in combination to automate 95-99% of operational decisions - EDM, in other words. EDM is about analytically-driven processes and while these include predictive analytics, predictive analytics can also be used in decision support systems and reports (I discussed the differences between decision support systems and EDM systems with Dan Power)

If you want to know more about some of the techniques I wrote a post on the basics of predictive analytics in EDM while Kathy Lange of SAS wrote a nice introduction on DM Review and I think the Berry and Linoff book on Data Mining Techniques is great. If you do go beyond predictive reporting to embedding predictive analytics, don't underestimate the challenges of letting machines take decisions. There is more in the section on predictive analytics.

As a foot note I was reading my Computerwire newsletter at the weekend and saw an interesting piece about an Accenture study. Yesterday I got Rob Preston's InformationWeek "Between the Lines" email about BI still being in its infancy. There were some nice little factoids in these two pieces:

• Most of the business information that middle managers eventually get their hands on is useless
• Middle managers spend more than a quarter of their time sifting and searching through information
• 50% of that information is of no value to them
• 60% of middle managers said the problem stems from poor information distribution

Now Rob goes on to say that you may well have problems "even if all your data is clean, up to date, and easily accessible in a central repository" and that there is "plenty of analytical overkill" going on. I think, in contrast, that there is not enough focus on the use of information to make processes and transactions flow more quickly by automating decisions. Middle Managers want to get their job done. Do they really need data or do they need systems that automate more of the day to day?

Finally a shameless plug for a recent piece by Bloor Research that discusses Fair Isaac's range of EDM technologies and how they "define the top-end of the data mining market".

Technorati Tags: analytics, BAM, BI, EDM, Enterprise Decision Management, predictive analytics, ROI, data mining, TDWI, Wayne Eckerson

Towards the end of last year I asked various people to make some predictions for 2007. Here, in no particular order, is what they said. Interestingly pricing and customer-centricity both came up more than once.

Bruce Richardson, Chief Research Officer of AMR Research sent me this one (though I am not sure he is serious)

• EDM will revolutionize professional sports, particularly baseball and football. Take football, especially this week’s Wild Card match up between the Patriots and the Jets.
• As the Pats coach breaks down game film, he enters data on how the quarterback responded to the defense (or vice versa) depending upon which down, yardage needed to first down, field position, score, time left in the half or game, field conditions, etc.
• All of this gets coded and entered into a computer.
• All of the data is stored and analyzed, and rules are generated - If 3rd and 15 with plenty of time and only one blocker in the backfield, the defense should...
• Rather than consult those plastic-coated charts with plays on them, the coaches watching from the skybox enter the formations into a handheld or laptop.
• The computer responds with a series of probabilities based on the best historical matches as well as input from the team’s head coach.
• These results in the new rules.
• Eventually, self-learning robots replace players...

Mike Schoeffler, President of Profitdesk Software had this to say:

• A nomination for the next big decision automation in an industry: bank loan and deposit pricing. Behind the scenes, one industry after another has automated pricing of their products - airlines, clothing stores, supermarkets. Bank pricing is equally broken. It is often guided by C-level executives, who rely upon their intuition and art because the tools have not been available. The changes this year will be invisible to customers if implemented properly, but will noticeably expand bank profits, even in an inverted yield curve.

David Raab of Client X Client had a couple of hot trends:

• Complex event processing: the notion of linking and finding patterns in multiple customer activities and using these to guide customer treatments. No different from normal customer management, except that it requires more sophisticated technology and richer data, including customer activities, customer profiles, environmental inputs (location, competitor behavior, weather, economic news, etc.) and business situation (products, inventory, etc.) Yes, it's yet another perfect application for EDM.
• Marketing mix modeling: part of underlying marketing ROI and optimizing marketing investments as a result. It's becoming a feature of marketing resource management systems, which are a logical home because they have all the information on marketing programs to begin with. But you need to add more information on results and more sophisticated analytics to make it the mix modeling possible.
• Both trends build on the wider accessibility of customer data, both as captured in enterprise warehouses and as visible in place through service oriented architecture, and on customer data integration technologies that make it easier to assemble the data from different sources and relate it to individual customers. Both also point to a need for over-arching management of the customer experience across all interactions (their specialty at Client X Client).

Barb von Halle and Larry Goldberg of KPI had a couple:

• There will be at least one, maybe two, new business-friendly software products available by 2Q2007 through which business-focused (not technology-focused) rule authors can write rules independent of a BRMS. Possibly with generation into 1 or 2 BRMS products
• There will emerge a business rule formalism that we believe will enable business rules and Enterprise Decision Management to integrate well in an SOA world by 3Q2007.

Henry Morris of IDC said

• In 2007, we will see delivery of more industry-specific composite applications that combine transactional, analytic, and collaborative decision-making tasks. Pricing decisions (which have a distinct vertical flavor) will be a popular subject area for these applications -- monitor changes to demand, analyze new trends, adjust the price in the transactional system, monitor, etc. Some will be packaged; many will be custom-built. The packaged applications will require extensive on-site customization. Providing an interface for model-based configuration and customization (at a level of abstraction suitable for business analysts) will be a differentiator for the packaged composite applications.

Sandy Kemsley, my fellow blogger over on ebizQ, had a piece on 2007 trends here and highlighted this one:

• There are two more Web 2.0 characteristics that I think we're going to start seeing in BPM in 2007: tagging and process syndication. Tagging would allow anyone to add free form keywords to a process instance (for example, one that required special handling) to make it easier to find that instance in the future by searching on the keywords. Process event syndication would allow internal and external process participants to "subscribe" to a process, and feed that process' events into a standard feed reader in order to monitor the process, thereby improving visibility into the process through the use of existing feed technologies such as RSS (Really Simple Syndication).

Ian Turvill, the other regular blogger on this blog, had a couple too:

• Increasingly, retailers will see Enterprise Decision Management as a key way in which they can drive customer centricity across the organization. A "Customer Centric" enterprise is one that delivers marketing, sales, and service to strategically distinct segments in very different ways so as to maximize the overall customer portfolio value. Customer Centricity is an approach which has been used by leading retailers, such as Best Buy and Tesco, as well as other recognized consumer service organizations, such as Harrah’s and Royal Bank of Canada, to achieve substantially superior financial performance. Customer Centricity involves five key steps:
• Understanding the current and potential lifetime value of customer relationships
• Explicitly choosing and focusing on a select set of customer segments
• Understand those customers in depth, including how to build larger relationships with them
• Engineering a "winning" value proposition across these customers’ entire shopping experience, including marketing, sales, and service
• Aligning the enterprise around delivering that proposition
• Insurance companies will adopt Decision Services to get better Decision-Making across the board. Insurers are recognizing that the benefits of predictive analytics and decision automation that they have seen in personal lines underwriting can also be realized in other functions and in other lines of business. They will respond by creating a separate decision-making capability an inherent part of their overall information technology architecture through the creation of "decision services". A similar prediction has been made before. In a July 2005 report, Karen Pauli, a Senior Analyst at TowerGroup, stated that: "The insurers that will survive this competitive environment are those that leverage technology to advance their business and operations. . . . One of the primary ways insurers are transforming their infrastructure is by supporting the insurance value chain in a modular way, such that applications are not all in one and can decouple and share some core components. . . .". What makes 2007 different is the rapid and widespread adoption of Service Oriented Architectures, which make it now possible to deploy the applications which exploit the SOA approach.

Craig Dillon, who heads up our ScoreNet business here at Fair Isaac, had another (he and I presented on Decision Service Providers recently).

• Technologies that control business rules will enable companies to increasingly move key decision points outside their four walls, enabling new classes of business process outsourcing. This will enable a new class of business – BPO networks with embedded business rules running at the hub.

As for yours truly, I have a couple:

• Every BI vendor will talk about having predictive analytics as a core component of their technology. Few of them will actually manage this.
• Every BPM vendor with long term prospects will partner with a major BRMS vendor and multiple integrated solutions will start to emerge.
• The idea of decisioning and decision services as a class of development will become mainstream or at least more mainstream.
• BRMS vendors will devote much of 2007 to improving the ability of business users to test, check and manage business rules
• Predictive Analytic workbench vendors will focus more and more on deployment of the models, not just development of them
• Some useful standards will be published and increasingly adopted in business rules and analytics, increasing the speed of adoption
• A really great book will get published on EDM and all the readers of this blog will buy multiple copies for their friends and family enabling the authors to take well-earned vacations in paradise.

One more thing. There was a nice 2007 Trends article over on Intelligent Enterprise about which I blogged here.

Technorati Tags: AMR, analytics, BI, BPMS, BRMS, EDM, Enterprise Decision Management, IDC, insurance, KPI, Marketing, predictive analytics, ScoreNet, SOA, standards, customer centricity

I am getting caught up on book reviews over the break. Today's is Data Mining Techniques by Berry and Linoff. This is one of the classic works on data mining and well worth the read.I really liked the book both because it is well written and because, although it drilled into a fair amount of detail about some of the techniques, it started each new section off at a high level. This allows someone without a statistical background, such as me, to read as far as I can in each section and then skip ahead to the next technique. This is a nice change from books that simply get more and more detailed as page follows page, preventing you from gaining an overview of the subject. The book introduces data mining and a methodology for applying it, talks about some of the applications in "Marketing, Sales, and Customer Relationship Management" (as the subtitle puts it), walks through some statistical techniques and then spends the bulk of the book on various data mining techniques. It wraps up with a nice summary of how data mining plays with other technologies and with some practical advice on getting started.

One of the best summaries of where data mining, and indeed EDM, fits is given early in the book where an enterprise is encouraged to:

• Notice what its customers are doing
• Remember what it and its customers have done over time
• Learn from what it has remembered
• Act on what if has learned to make customers more profitable

The authors point out that Data Mining is focused on the "Learn" stage or, as they put it data mining suggests but businesses decide. EDM, of course, is concerned not only with learning but also with acting, most particularly acting by automating decisions in front-line systems. Merely finding patterns is not enough - you must respond to the patterns and act on them, ultimately turning data into information, information into action and action into value.

The methodology section, and the subsequent notes that relate to applying these techniques in real life, talked about the feedback loops between steps in data mining - there is not a linear "waterfall" sequence of steps but constant iteration and learning. They also emphasized the importance of finding the right business problem at the beginning - start as someone once said, with the end in mind. This was reiterated when they quote Voltaire who said "Le mieux est l'ennemi du bien" ("The best is the enemy of good"). In other words, don't get hung up on trying to find the perfect algorithm, perfect answer. Instead build something that is good, that works, and learn and improve over time.

The authors made a big point out of the value of data mining for "mass intimacy", where you want to treat customers differently and there is a business reason to do so but where customers are too numerous to be assigned to staff. One of the issues they pointed out was that staff must be trained in customer interaction skills while also using all the data you have. This can be a real challenge and is one of the reasons I prefer an EDM approach, where the decisions those staff need to make are automated, to other approaches. By giving them the decisions they need you free them to work on the relationship (as I have discussed before). The value of data mining, and EDM, in building a customer-centric organization cannot be overestimated.

Some random snippets of useful stuff from the book:

• A model "can result in insight" and "produce scores". The first kind is used in EDM largely to product rules while the second is often embedded directly in the decision services being built
• Analysis can be directed (find the value of something) and undirected (find structure)
• Data visualization is very useful during the initial exploration of information.
• There is some discussion of the difficulty in deploying models when the step involves"a programmer takes a printed description of the model and recodes it in another programming language so it can be run on the scoring platform". EDM's focus on automating the deployment of models into a rules-based decision service is designed to address this issues.
• Besides coding the actual model, data transformations are also a big issue and remain one even in EDM.
• Decision trees are "powerful and popular" for classification and prediction because they can be represented by, and represent, rules. Indeed decision trees are a cross-over artifact between rules and models that are critical in EDM also. One of the things that makes trees particularly useful is because they need less data preparation as they can handle all kinds of variables well.
• The authors emphasize repeatedly the importance of time series data e.g. detecting early signs of attrition by tracking all actions of checking account customers in the time up to when they leave a bank. The time-based signatures thus created are great predictors. They note also that this is one of the weaknesses of data warehouses when using them for analytics - they tend to arrange data by absolute time/date when the analytics are more useful relative to an action or event.
• The value of neural nets is noted but the problems neural nets have with respect to traceability and explicability are also noted. This makes neural nets great for things like fraud detection, where results matter and reasons matter less, and poor for things like credit assessment where regulators expect to see compliance with rules.
• The section on market basket analysis and association rules is very good and describes these forms of undirected analysis well. They point out that these can, if you are not careful, describe the history of marketing promotions rather than genuine decisions to purchase products together. They also give some good examples of using product hierarchies to generalize where some products are much lower volume than others.
• They describe a pyramid with operational data on the bottom, summary data next, the database schema on top of that followed by metadata and finally busienss rules - what's been learned from the data.
• They worry that"rules" are not actionable but I think this is because they focus on rules that describe the data not on rules that describe the actions to be taken

You can buy the book here and it should definitely be on your bookshelf.

Technorati Tags: algorithms, analytic application, business intelligence, business rules, CRM, customer insight, customer segment, decision automation, EDM, Enterprise Decision Management, marketing, predictive analytics, segmentation, statistical analysis, data mining

I saw this post by David Raab over on his blog - Business Intelligence on Smart Phones: Not Just Humbug. Like the author of the original article to which David refers (Power Of A Data Warehouse In The Palm Of Your Hand by Elena Malykhina) I am cynical. Her comment that "It remains to be seen how many mobile professionals actually need to slice and dice data from handheld devices" really struck a chord with me. I don't see even weary road warriors wanting to do "traditional" BI on a smartphone. But as David correctly points out the follow-up question is interesting:

The more intriguing question is what new business intelligence functions a smart phone platform would make possible

Now substitute the works "decision management" for "business intelligence" and I think you are on to something. One of the the differences between BI and EDM is the focus on taking action using insight gained from data rather than showing someone the data and helping them gain some insight. I would say that David's examples are all, in fact, EDM examples. They use the information the phone has (position), insight from the data the company has (fraud likelihood, wait times) to take an action (dispatch the person with the phone to a particular place, tell them to do or not do something). I don't see traditional BI vendors having much to offer here - the whole reporting/OLAP infrastructure they have developed is predicated on knowledge workers doing analysis. If you want to take advantage of mobile devices you need to think about automating decisions for the person holding the device. For instance:

• Use mobile phones held by maintenance engineers to track their location and then use analytics to predict which pieces of equipment are most likely to fail soon and rules to assign the nearest, qualified engineer before sending the directions on where to go to the engineers phone.
• Don't show them reliability graphs or travel times, tell them where to go to make best use of their time
• Use the mobile phone of a real estate appraiser to find out which risk zones a property is in and what the predicted difference is between a house inside and outside that risk zone
• Don't show them a picture of the risk zones
• Use a doctor's mobile phone to route them to the most useful hospital during an emergency based on predictions of patient load, the hospitals they know and their specialties
• Don't show them graphs of wait times and pie charts of specialties needed
• Use a customer's mobile phone to make them an offer at a store that is nearby having predicted that they are likely to buy it, checked that is in stock there and estimated that they are more likely to respond in person than to an email promotion to the website

And so on. Automate decisions and use mobile devices to provide context for those decisions and to deliver decisions to people out and about. Don't send them reports. Please.

I have blogged before about the value of location information in automation of decisions and on location intelligence with EDM and wrote an article in BI Journal (subscription required) with Ed Gandorf of MapInfo "Driving Decision Automation with Location Intelligence".

P.S. An extreme example of this might be something like pay-as-you-drive insurance as described by my colleague Ian.

Technorati Tags: analytic application, analytics, business intelligence, business rules, customer experience, customer insight, decision automation, EDM, Enterprise Decision Management, personalization, predictive analytics, mobile device

Ian, my fellow blogger here at edmblog.com, pointed me to this article You Must Market To Algorithms, Not Just People. The article nicely summarized the growing role of algorithms in marketing and, as predictive analytics and even rules can be considered a form of algorithm, it made me want to talk about enterprise decision management or EDM in this context. Let's start with the key concept:

As more human behaviors emit trails of digital residue, the more opportunities reside for algorithms to harness those human-induced data and become information intermediaries

Absolutely right. And these data are available in such volume that reporting on them is not going to help anyone - you have to build insight from the data so that you can use it, not just report on it with BI. This means building predictive analytic models based on the data that can be embedded into your operations - algorithms in other words. In reality you must also combine rules - about the user's preferences so as to maximize the customer's influence on decisions and about policy and regulation to ensure compliance. This is particularly true for any business subject to the Long Tail. Automating decisions in this way can let you improve the customer experience and scale 1:1, personalized communication to thousands or millions of customers.

As Max says, "all behaviors .. create halos of metadata, which algorithms process, mediate and disperse to others" and you need to account for all this information in your decisioning. Max gave some examples of algorithms and I thought it would be fun to show how EDM-like some of them were:

• Restaurant recommendation
  Well you could use predictive algorithms to segment people by the kind of restaurants they like/use and to build recommendations by comparing to other people. The customer can set rules for price, location etc as well and the combination comes up with restaurants.
• Real-estate
  Looks like this one was just rules - rules from the customer about properties in which they are interested
• Travel
  More preference rules plus models predicting occupancy and handling dynamic pricing as a result
• Music playlists
  Rules and predictions for like and dislike
• Fraud
  Rules and analytics, a classic EDM one.

Technorati Tags: analytic application, analytics, business rules, customer experience, customer insight, customer segment, decision automation, EDM, Enterprise Decision Management, fraud, long tail, marketing, predictive analytics, algorithms