Tony, over on the Decision Support Analytics blog, is running an interesting competition and the prizes have just improved - Neil and I offered a signed copy of our book to add to his list of potential prizes. You can check out the details here. I plan to write a couple of entries soon and I encourage any of you who blog to do likewise. Details of how to enter are here.
Cross-posted to both blogs

Posted by jtaylor in Business Intelligence • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

I was having an interesting discussion over lunch with a CTO/Professor who articles that all data is streaming data and I posted it to my twitter account as a question. Brenda took the trouble to reply, saying:

I say no. Most data is "at rest". Streaming typically corresponds to frequent state changes of the thing the data represents and of course, someone/thing cares to know about the data (thing) as it looks right now

You can find my twitter posts at twitter.com/jamet123 and Brenda's at twitter.com/bmichelson

Posted by jtaylor in Business Intelligence | Permalink | Comments (1) | TrackBacks (0)

I often post about the weakness inherent in many approaches to business intelligence - this one on getting a competitive advantage, this one on dashboards and this one on moving beyond BI come to mind - so it is nice to be able to recommend an article on analytics and BI. Dave Wells wrote a nice summary on this tpoic Business Analytics – Getting the Point. I highly recommend reading this article and, in particular, there is a great graphic :

This graphic makes a number of things clear:

• You must take action for intelligence to be useful
• You must track and analyze outcomes to understand what worked
• Your results must influence the next actions you take so that you can learn

Posted by jtaylor in Business Activity Monitoring • Business Intelligence • Decision Technologies • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

One of my favorite topics is the use of decision management, and the creation of decision services, to marry business intelligence to SOA. I saw an interesting article by Tobin Gilman, Senior Director of BI Product Marketing at Oracle titled Integrating BI Within Your SOA
. The article made some valid points about the power of data about processes to improve those processes through things such as Business Activity Monitoring or BAM. However the focus is all on people gaining insight from data. For instance:

When a business user gains insights that flag a business performance problem, some kind of action will typically need to be taken in order to address the problem. Often, the action may involve invoking a business process. If this is difficult to achieve, then there is often less value to the insight.(my emphasis)

Posted by jtaylor in Business Intelligence • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Intelligent Enterprise ran an article on Gartner: Emerging Technologies Will Help Drive Mainstream BI Adoption and it made me wonder what "mainstream" means in this context. After all, most of the technologies don't really change the basic premise of Business Intelligence - that the purpose of BI is to deliver insight to some kind of knowledge worker. But does this count as mainstream? I don't think so.

To make "BI" mainstream would mean that everyone in an organization - down to the people paid minimum wage at the front line - are making better decisions thanks to the understanding an organization has of its data. Indeed, it should go further and ensure that the machines that deal with customers (ATMs, Kiosks,websites) can also use this insight. Yet visualization, in-memory analytics and search integrated with analytics do not enable this - in fact they just make old-style BI prettier or faster. SOA and SaaS are more interesting as they enable decision services but they are still not enough. To make "BI" mainstream it must change from BI as reporting to something more decision-centric - it must focus analytic insight on the making of decisions in software not just in people's heads. This takes rules, predictive analytics, data mining and decision management not more "BI".

Posted by jtaylor in Business Intelligence • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

I thought it would be fun to highlight my 20 most popular posts from the last couple of years so here goes:

1. Keeping Predictive Analytics and BI on separate tracks
   
2. If IT wants to alter outcomes, it needs to automate decisions
   
3. Business rules, events and processes
   
4. Getting a competitive advantage from your data
   
5. Achieving Agility - some notes after Gartner
   
6. Introducing business decision management
   
7. Here's a way to put analytic solutions in the driving seat
   
8. Decision Technologies and Active Data Warehousing
   
9. SOA and Business Rules, perfect together
   
10. Business rules, routing rules, event rules
    
11. Decision Services
    
12. Decision management is critical to event driven architecture
    
13. Decision Management - another way to get the business to care about SOA
    
14. Marketing Analytics in a Post-Web 2.0 World
    
15. Little known ways to improve customer experience
    
16. More on rules and event processing
    
17. Business rules, desktops and knowledge buses
    
18. If dashboards are the end game, kill me now...
    
19. COBIT, SOX, compliance and business rules
    
20. Call for Presentations - the new EDM Summit
    

Posted by jtaylor in Business Activity Monitoring • Business Agility • Business Intelligence • Business Process Management • Business Process Outsourcing • Business Rules • Compliance • Decision Technologies • Event Processing • Innovation • Legacy Modernization • Predictive Analytics • Requirements • SOA | Permalink | Comments (0) | TrackBacks (0)

• How are you integrating business rules and analytics?
• How are you adding intelligence to your business processes?
• How are you putting analytics to work in your operational systems?
• How, in other words, are you using Enterprise Decision Management (EDM) to innovate your business? Your colleagues and peers want to know.

We invite you to present at the Enterprise Decision Management (EDM) Summit OCTOBER 26-30, 2008 -- ORLANDO, FL  

Join us this year to share how the technologies and approaches of Enterprise Decision Management (EDM) have helped your organization deliver agility while managing risk, focusing on customers, or demonstrating compliance. Whether you call this Business Decision Management or Customer Decision Management or just Decision Management, we want to hear from you. The EDM Summit brings together managers, practitioners and vendors to talk about what works and to provide attendees with a host of practical ideas they can put to use in their own companies. These practical ideas don’t come from us, they come from you.

Most of all, we want real-life case studies. We want to hear what really happened, what worked and what did not, from the actual people who undertook them. Whether you want to show how you got started or how you have learned from experience, whether you want to talk about technology, people or methodology we want real-life cases. If you are a consultant or vendor, the best way to be accepted is to co-present with someone from your client’s organization. Real experience, not company positioning or marketing buzzwords, is what it takes to be selected.

Particular areas of interest include, but are not limited to ...

• Using decision services with BPM or SOA to put intelligence into composite applications
• Using business rules and analytics or data mining in combination
• Implementing adaptive control and champion/challenger testing
• The impact of the technologies and approaches of EDM on the software development life cycle

 Your presentation ideas are welcome in any of our mainstay topic areas, including ...

• Business Rule Management Systems and Engines
• Data Mining and Predictive Analytics Technologies
• Techniques and Methodologies for Data Mining and Predictive Analytics
• Decision Services, Business Rules and SOA
• Adaptive Control and Optimization
• Managing Decisions in BPM and SOA
• Moving to BI 2.0 / Operational BI
  
• Event-based Decision Management
• Compliance and Risk Management
• Organizational Change

Not on the list? Tell us about your own unique ideas! Top architect at a mainstream software vendor? Creator of a highly innovative product? We will consider your presentation for our Chief Architect's track. Highly selective. 

We Want to Hear from You ...

We welcome presentation ideas from all! Do you have a business success story? Best practices about how to use decision management technology? Significant progress in applying data mining to operational systems? EDM Summit is THE place to present on experience, proven solutions and new innovations in this exciting area. We bring together companies and experts from diverse industries in a unique and exciting venue to share their experiences. Do you know qualified colleagues who would make great Forum speakers? Click here to forward this message to a friend. Invite them to submit their Presentation Abstract for consideration!

CALL FOR PRESENTATIONS SUBMISSION DEADLINE: March 31, 2008  

To submit your presentation idea ...
Step 1. Please read the Speaker Agreement carefully.

Step 2. Complete the Speaker Abstract Submission Form.
Presenters will receive a full complimentary registration to the Business Rules Forum, including the two co-located conferences Enterprise Decision Management (EDM) Summit and Rules Technology Summit as well as to RulesExpo.

Got a question? Please email us at speakerinfo@businessrulesforum.com 

Posted by jtaylor in Business Activity Monitoring • Business Agility • Business Intelligence • Business Process Management • Business Rules • Compliance • Decision Technologies • Event Processing • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

Gary Cokins had a post this week When Analytic Solutions take the Back Seat in which he bemoaned the fact that customer facing processes so often fail to be improved analytically. For all the work done in supporting technical, back-office processes, customer service is often left out in the analytic cold. I have blogged a lot recently about how decision management can drive a better customer experience but Gary's point is a good one - all too often customer facing processes are not improved in this way. Part of the reason is, as Gary notes, a lack of will power on the part of companies and a failure to realize just how much damage bad customer service can do. But pat of it comes from failing to see how and where a customer facing process can be improved.
One of the clearest things when discussing customer treatment decisions in customer-facing processes is that companies massively underestimate the number of treatment decisions they make. Every time they display a webpage, list a voice prompt, make a customer service rep refer someone to their manager, generate a script for someone in a call center, display member details to a store clerk they make a treatment decision. These decisions are often made the same way for every customer, failing completely to deliver the kind of customer service people want. Find these decisions, focus on them and you will find plenty of ways to usefully apply analytics to improve them.

Posted by jtaylor in Business Activity Monitoring • Business Intelligence • Decision Technologies • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

Gary Cokins of SAS blogged about Predictive Analytics – Dream or Reality? today. He was responding to Juha Harkonen's post from a CFO conference. Essentially Juha was complaining that while a small number of companies are getting into predictive analytics, many are still struggling with basic performance management and KPIs. I have heard similar comments from folks focused on BI, especially operational BI (for instance in this morning's TDWI webinar).

I am going to challenge the assumption that data quality, data warehousing, data integration, reporting, OLAP and performance management must all come BEFORE predictive analytics and decision automation. In fact, if you look at those industries furthest ahead in the use of predictive analytics such as retail banking, this assumption is clearly false. Many banks have been using predictive analytics in automated decisions for years but are not much further along in the rest of the data/BI competencies than your average large company. Clearly one does not need an enterprise data warehouse strategy nor does one need to have solved all one's problems with BI and data before using predictive analytics. So why the persistent talk of it being "after" all those things?

Well, for one thing it is more complex in its detail so there is an inbuilt assumption that it must come after the others. It is also true that newcomers to predictive analytic usage often have done a lot in terms of data quality, governance, warehousing and reporting. Lastly it seems logical that the group that manages data should also manage predictive analytics.

In fact none of these things is axiomatic. Companies focused on improving operational, high-volume decisions may well find it easier to embed a predictive model than to handle performance management on that process in real time. Many predictive models are built in environments with incomplete data and many predictive modeling techniques are designed to come with these problems. Many organizations have both an analytic or decision sciences group AND a BI Competency Center and they don't overlap at all.

The best way to approach this is to decide which decisions you must improve and what you need to do that. Start with the decision in mind and don't assume you must complete everything at any given "level" before moving on in targeted areas - each decision has its own constraints and needs and should be addressed in that context.

Predictive Analytics does work and has worked for years. Many companies are using it. You don't have to finish BI before starting predictive analytics.

Posted by jtaylor in Business Intelligence • Predictive Analytics | Permalink | Comments (1) | TrackBacks (0)

Joe pointed me to an interesting article in his recent post discussing the power of data, and analytics, to replace "gut instinct" in business. This follows my post earlier this week on getting a competitive edge from your data.

I have to agree with this point of view - clearly companies now have so much data about what works and does not work and so many tools for understanding that data that there is no longer any excuse for just trusting "gut instinct"

Clearly the past is not a perfect image of the future and some things that might work well may not be allowed so it is not just a question of doing analytics on your data - you must use the kind of analytics that make predictions and model uncertainty (predictive analytics). You must also combine them with rules derived from regulations, policies, best practices etc and constantly challenge them with new and potentially better approaches - ongoing randomized testing or adaptive control.

One of the values of a more automated, data-centric approach is that it helps address some of the well known decision making traps because it is not clear that experts REALLY do better when it comes to making decisions. While it may be scary to let machines take decisions, it need not be.
Finally, you might also enjoy my review of Blink

Posted by jtaylor in Business Intelligence • Decision Technologies • Innovation • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

I was thinking about how you can get a competitive advantage from your data and how well traditional tools for analyzing your data fit.

Let's think about it . First you must act differently to get a competitive advantage. Therefore you cannot just understand something better, you must act more effectively. This means you must make decisions differently and that means applying rules and constraints to the insight you get from your data because the real world imposes regulations, policies and constraints.

If you are going to change the way you act it is not enough to improve your reporting on the past, you must also improve your ability to predict the future. This means using predictive analytics not passive reporting or analysis tools.

Increasingly, for companies of any size, your systems are your business - you cannot do without them. Therefore these predictions and rules must go into your systems - your operational systems. Finally, of course, nothing is static so you must constantly challenge and assess the effectiveness of decisions

Given all this, then, traditional BI tools are not, perhaps, the best way to get competitive advantage. Instead, Enterprise Decision Management (EDM), or Business Decision Management as it is sometimes known, is the approach you need - it's an approach for automating and improving high-volume operational decisions. Focusing on operational decisions, it develops decision services using business rules to automate those decisions, adds analytic insight to these services using predictive analytics and allows for the ongoing improvement of decision-making through adaptive control and optimization.

Prompted, in part, by Nathan Jones' post on Using BI to gain a competitive edge

Posted by jtaylor in Business Intelligence • Business Rules • Decision Technologies • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

The Endgame for Business Intelligence is, according to CRM daily, dashboards. Say it ain't so! The idea that dashboards are the endgame for BI is frankly terrifying. Have you see the typical dashboard? To misquote Shakespeare, "Much color and expense, signifying nothing". Seriously, dashboards may be a step up from greenbar reports but the endgame? I don't think so.
First problem is that dashboards are still backward looking. Surely the endgame for BI must involve some kind of predicting of the future, some looking forward? Even if all the system is going to do is tell someone the predictions, at least the system should do that much work instead of making the user interpret the graphs and dials.
Secondly, who has time to look at all this stuff anyway? Surely what I want is for my business to run better. Sometimes that requires me to look at something, think about it, drill into it and make a decision to change a manual process. More often it requires that one of my systems or processes should start operating differently - take different decisions and act more appropriately based on the data.
Let's do as the author says and "Consider a real-world marketing director in Dallas". Why would I give this person dashboards for some of these things? Why wouldn't my online advertising engine just learn and improve from the click through and conversion data? Why wouldn't my CRM system make customer segmentation and treatment decisions that use the data? Why show dashboards for these things? The marketing director may not look at them, in which case nothing would be done and he/she may not interpret the data correctly even if they do. Indeed the article shows the weakness of its solution in its fourth paragraph when it says "The first behavior of business leaders we need to take into account is a tendency to be busy". Exactly. So why show them a dashboard when we could program the systems to ACT?
Enterprise decision management would use this same data not to display a pretty graph on some marketing directors desktop, but to make the CRM system, the website, the call center and everyone involved in the organization act more intelligently.
Dashboards as the end game? Let's hope not.

Posted by jtaylor in Business Intelligence • Decision Technologies • Predictive Analytics | Permalink | Comments (4) | TrackBacks (0)

There is some great "premium" content on ebizQ for those signed up as Gold Club members (its free, just sign up here) including this one - Forrester: It's Time to Reinvent Your BI Strategy
. In this report Boris Evelson gives some good guidance on BI strategy in general but I can't help feeling there is more to it. For instance:

Forrester defines BI as a set of processes and technologies that transform raw, meaningless data into useful and actionable information.

There's lots of good stuff about BI in the report and then Boris says this:

Enterprises can’t just focus on being efficient anymore. Squeezing more efficiency from operational applications such as ERP, CRM, or SCM, no longer helps enterprises to stay competitive. BI applications are needed for business processes and business operations to become more effective. For example, workflow and rules can be used to efficiently process a customer credit application, but BI analytics can effectively segment a customer population and target credit offers to specific customer segments for a better response and improved cross-sell and upsell ratios.

BI, rightly or wrongly, has become inextricable from the functionality of BI vendors and that means they are focused on analysis and reporting tools. The rules and workflow in the customer credit application cannot be improved just by using BI - you need to think of the credit decision separate from the processing and you need to embed not just rules but also predictive analytics into the decision. You might well have rules derived from data mining, the extreme end of the BI analytics spectrum, but even here you are going to have to have thought about the problem as one of improving the rules for a specific transaction and not as one of analyzing the customer base to understand it. This is the difference between analytics in BI (more sophisticated analysis) and the analytics in enterprise decision management (making more, and predictive, information available for a specific instance in a specific transaction so as to improve a specific decision).

Boris goes on to say that

"BPM might make processing a customer credit application less expensive, but analytics can use sophisticated customer segmentation to increase cross-sell and upsell ratios in real time during a customer interaction — when it counts"

Go beyond BI, even revamped BI, and use EDM to make your processes more effective.

Posted by jtaylor in Business Intelligence • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Forrester recently published an excellent paper - The Dynamic Business Applications Imperative (John Rymer and Connie Moore) - that you all should buy and read (if you don't have a Forrester subscription it's a steal at $279). This paper defines a Dynamic Business Application as

"A software system that embodies a business process and is built for change, adaptable to business context, and information rich."

"Most business applications are too inflexible to keep pace with the businesses they support."
"...they force IT pros to spend too much budget to keep up with evolving markets, policies, regulations, and business models."

"IT's primary goal during the next five years should be to invent a new generation of enterprise software that adapts to the business and its work and evolve with it."

• Imagine Apps That Evolve With Your Business And Its Work
  
  Apps that evolve with your business must allow business people to change how things are done in those applications - using business rules to manage decisions, for instance - while also applying the data your business collects to improve - using predictive analytic models to inform decision-making, for instance. Many of the requirements for evolution in an application are around evolving decisions, putting decision management at the heart of this strategy, particularly when it comes to building for change (something about which I have blogged before).


• There's Nowhere To Hide: All Business Processes Are Subject To This Trend
  
  Absolutely not. For all that various aspects of these applications have traditionally been used in certain domains, the pressures that drove those domains to be early adopters are now affecting everyone and every process. You will have to take control of the decisions in your information systems and business processes, no matter what.


• Progress To Dynamic Business Applications - Start Your Journey By Resetting IT's Relationship To Businesspeople
  
  This reset can be achieved in the core logic of your systems, not just in the periphery. Using business rules to empower business users to manage the decisions in their applications changes this relationship while also increasing your tolerance for (inevitable) change. I blogged before about something Forrester called Concurrent Business Engineering

• Business Process Management


• Business Intelligence and Performance Management


• Decision Management

"the tools are at hand, and pioneers in service-oriented architecture (SOA), business process management (BPM), and business rules ... have begun showing us the way. The time to start on this journey is now."

Technorati Tags: agile, John Rymer, business rules, decision technology, change time, Connie Moore, decision service, decision management, BI, business intelligence, BPM, business process management, Forrester, dynamic business applications

Posted by jtaylor in Business Intelligence • Business Process Management • Business Rules • Decision Technologies • SOA | Permalink | Comments (0) | TrackBacks (0)

Sandy has been at the Forrester event and posting some great stuff on business process management, business rules and business intelligence. Today I read five posts that you shouldn't miss:

• Forrester Day 1: Automating Business Processes panel
• Forrester Day 1: John Rymer
• Forrester Day 1: Rich Phillips, Maritz Travel
• Forrester Day 2: The three B’s
• Forrester Day 2: Skipping class

She posted the slides for her presentation in the last one and slides 21-24 are particularly excellent. There were a number of points made in these that I wanted to call out:

• Business process management and BPMS products bridge the gap between end-user computing and full on application development. Business rules, used to automate decisions, can also bridge this gap by putting the business in control of decision-making logic in processes, in legacy systems and in enterprise applications
• As John Rymer pointed out, business rules help you build for change so you can deliver systems that respond rapidly and safely to changing business conditions.
• One of the best ways to get control of processes being managed using BPM and developed on an SOA platform is to build decision services that encapsulate and manage the decisions within them.
• One of the great values of considering business rules, business process and business intelligence as a set is that rules-based decisions are a great platform for embedding analytics into process as I discuss in this article
• It is this combination that drives agility, as I have noted when discussing agility as defined by Gartner

Technorati Tags: business rules, business process, business intelligence, Forrester, business agility, Sandy Kemsley, column2

Posted by jtaylor in Business Agility • Business Intelligence • Business Process Management • Business Rules • Decision Technologies • SOA | Permalink | Comments (1) | TrackBacks (0)

Andy Bailey wrote an article recently called "Where BI has fallen short". Andy's focus is mostly on how BI has fallen short of helping managers and knowledge workers and he outlines four pillars of an ideal solution - Information Access and Search, Collaboration, Knowledge and History, and Rapid Configuration.  All of these are good points but I think he is missing the importance of operationalizing this (I know, horrible word). What I mean is, how do I bring this better decision-making to bear on the operational systems that act on my behalf, that drive my business? How does my knowledge and history get rapidly combined with information  when I am talking not about people, but about a system?

As I said last week, I think the votes are in and decision technology should be part of your BI stack so that you can share intelligence with your systems. This was the main trust of Smart (Enough) Systems (recently excerpted on ebizQ - Book Excerpt: Smart (Enough) Systems). BI has fallen short in how it helps people make decisions. It has fallen much shorter in how it makes systems smarter.

Technorati Tags: BI, BI 2.0, Business Intelligence, decision management, Smart (Enough) Systems, smartenoughsystems

Posted by jtaylor in Business Intelligence • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Yes, that's right, part of your Business Intelligence stack. Regular readers of this blog will know my opinion on this but I am not alone. Check out these "votes" - articles by independent writers of various kinds:

• Reactive and Proactive Business Intelligence by Gary Holden (the article that prompted the post)
• Business Rules and Business Intelligence by Robert Blasum
• Business Intelligence and Business Rules Make Great Bedfellows by David Straus
• Toppling the BI Pyramid by Neil Raden (co-author of Smart (Enough) Systems)
• A Road Map to Intelligent Business by Mike Ferguson
• Plus these three articles by Cyril Brookes
• Building Automated Decision Making into BI System Design - A Methodology Overview
• Decision Automation in BI - Selecting Decisions to Automate
• Decision Automation in BI: Design Guidelines for Business Analytics and Rules

Of course you should also consult my article on how to put your performance management into action and the whole BI section of the blog, especially decision technology as a platform for BI.

So, doing BI? Better be doing rules too....

Technorati Tags: BI, BI 2.0, BRE, BRMS, business rules, competing on analytics, data mining, decision management, predictive analytics, Smart (Enough) Systems, smartenoughsystems, Business Intelligence

Posted by jtaylor in Business Intelligence • Business Rules • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Well I'm back and this week the main ebizQ site published an excerpt from the book I just wrote with Neil Raden, Smart (Enough) Systems. The excerpt is from Chapter 10, EDM and the IT Department - Book Excerpt: Smart (Enough) Systems. You can buy the book here (and you should, if you have not already)! There is more information on the companion site www.smartenoughsystems.com.

Technorati Tags: adaptive control, analytics, business rules, data mining, decision management, decision service, EDM, enterprise decision management, Neil Raden, predictive analytics, Smart (Enough) Systems, smartenoughsystems, SOA, James Taylor

Posted by jtaylor in Business Intelligence • Business Rules • Decision Technologies • Predictive Analytics • SOA | Permalink | Comments (0) | TrackBacks (0)

I had an interesting call with Aaron from SeaTab last week. SeaTab is a SaaS BI vendor focused on retail, CBG and supply chain. Aaron was briefing folks as SeaTab had a new release out - we did not get into the details as I was more interested in general questions but you can see the press release they put out here. Aaron went through the basics of the product, describing it as very user configurable, down to calculations in individual reports. The integration of structured data from almost any source is a big focus and the product boasts that it requires no physical data warehouse or marts - it has no pre-defined dimensionality. Indeed there is no ETL or DW or BI - all of this is handled by SeaTab. Users upload flatfiles incrementally to a virtual DW where they are de-normalized and tokenized for performance. SeaTab works with customers to define a logical data model so have some data definitions and the logic/no-dimensionality allows for quick updates for changes to data sources. While SeaTab is mostly focused on reporting for large numbers of users - front line e.g. sales tracking, reorders - they do some monitoring and automated responses e.g. reorders based on some calculation involving stock levels and activity. So far so good but how does this relate to decision management? Well in a couple of ways.

Firstly you can do decision management using SaaS also - for instance Fair Isaac has the concept of a decision service provider that does exactly  - and if you do then SaaS BI is interesting as a way to provide support for those decisions that could not be automated 100%. Secondly it is true that many organizations find the development of the basic analytic environment they need to understand data so that they can move on to predictive analytics is time-consuming and costly to set up and a SaaS solution might allow for quicker development of infrastructure suitable, longer term, for decision management. However, the ease with which SaaS BI can be integrated with decision services is perhaps the most compelling reason for considering them together.

Because decision services have well define interfaces, encapsulate all the complex business logic and are easy to change they are easy to integrate. Even well designed decision services do not automate 100% of decisions, however, and it is not unusual for a referred decision (one that could not be automated) to need a tightly focused report or visualization to accompany it. The use of SaaS BI might allow rapid integration to support decisions so that the decision service could return either a decision or a the reasons a decision could not be made along with a URL to a SeaTab page that would help with the manual decision-making.

Of course I also think the SeaTab folks should think about how they offer decision management SaaS in addition to their SaaS BI but the BI section of the blog goes on and on about why I think that so I won't repeat it.

Technorati Tags: BI, business intelligence, SaaS, SeaTab

Posted by jtaylor in Business Intelligence | Permalink | Comments (0) | TrackBacks (0)

Timo Elliott over at the BI Questions blog had some great cartoons on this topic - this one about "gut-feel" and this one about what executives want from BI. These are both genuinely funny but they also point out a serious issue - that gut-feel may be overrated but that it will not necessarily be replaced with better decision-making just because more data is available.

It seems to me that there are two aspects to this problem - the problem of executives (who want the magic button for strategic decisions) and the problem of workers (who want a somewhat magical button for day-to-day stuff). Starting with the second one, you can (and should) focus on the challenge of automating decisions not just supporting them. After all, front-line workers have less time and less experience with data analysis and so are easier to overwhelm with data. They are also, perhaps, not the people you want making "gut-feel" decisions about your customers.

Clearly this does not work for the strategic decisions, however, as they are not repeatable enough to lend themselves to automation. Often these decisions are about decision strategy - how aggressive should I be about pricing decisions, about retention, about risk. If you think about each of the operational decisions (micro decisions) and automate them then one of the side effects is the ability to run simulations of how a change in strategy might impact these decisions. This is, of course, not something you could do if those decisions were being taken manually. This allows you to consider the macro as well as the micro decisions in a systematic way.

While this kind of decision simulation has not got to the point of being able to say "Let's say you want to save millions of dollars - you just push this button here", it is at the point where you can say "which of these three approaches should generate the best return, given the real-world constraints on my business" and then have an easy way to pick the rules that seem to work best and push them into production without pain.

I blogged about another Joe McKendrick post that seems relevant here - Sharing intelligence with your systems - and you might like this article on shifting your CPM into action. Meanwhile Cyril had a nice post on this too Decision Automation in BI: Design Guidelines for Business Analytics and Rules 

Technorati Tags: automated decision making, business agility, business rules, competing on analytics, decision automation, decision management, decision service, micro decisions, optimization, predictive analytics, simulation

Posted by jtaylor in Business Activity Monitoring • Business Intelligence • Business Rules • Decision Technologies • Predictive Analytics | Permalink | Comments (0) | TrackBacks (0)

Joe McKendrick had an interesting post yesterday over on the BI in Action blog - To 'Compete on Analytics,' Intelligence Has to be Shared. I reviewed the book to which he refers, Tom Davenport's Competing on Analytics on my other blog a little while back (here) and think that decision automation offers another way to "share" intelligence - only with your systems, not just with your people. At one point in this post Joe says:

"Add to this the fact that most end-users do not have access to the latest BI tools, and still have to go through IT or other departments. "

Now the question I would ask is do you, in fact, want most end-users to have access to the latest BI tools? Do your bank tellers, retail clerks, truck drivers, greeters or even call center representatives need or want business intelligence tools as they are currently defined? Could they use them if they had them? Is their turnover low enough to justify training them? Joe went on to point out that

"Overall, the survey found, fewer than 10 percent of employ­ees have access to BI and corporate per­formance management tools"

That does sound too-low to me but still I think that the solution is not to give everyone BI or performance management tools but to make the systems with which they work smarter and better informed. BI tools alone, even if widely deployed, will not "work at the speed of front­line decision-makers". If I have seconds to make a decision, I need a system that is focused on that decision. After all, those who know first don't win, those who act first do! In addition, performance management is more than performance monitoring. Most corporate performance management tools are really just good for monitoring. You need to be able to change the way you (or your systems) decide in response to your monitoring before it can be considered management. As my co-author Neil Raden said in a great piece over on Intelligent Enterprise about BI 2.0 :

Rethink analytics - Informing people to make better decisions is out; changing the nature of work is in

In the interests of fair disclosure, Joe wrote a very nice testimonial (here) for the book Neil and I recently completed on this topic, Smart (Enough) Systems.

Technorati Tags: analytics, BI, business intelligence, competing on analytics, decision automation, decision management, Neil Raden, predictive analytics, Smart (Enough) Systems, smartenoughsystems, performance management, Tom Davenport

Posted by jtaylor in Business Intelligence • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Cyril Brookes had a great article this week - Decision Automation in BI: Design Guidelines for Business Analytics and Rules - where he talks about applying rules in decision automation as a way to enhance BI. Worth reading for all you BI types.

P.S. This is my 200th post on the blog! Wow.

Technorati Tags: BI, business rules, decision, decision automation, business intelligence

Posted by jtaylor in Business Intelligence • Business Rules • Decision Technologies | Permalink | Comments (0) | TrackBacks (0)

Michael Dortch blogged a post over on the BI in Action blog today - BI, BPM, and SOA: Alphabet Soup that's GOOD for You! Inspired by his comments I thought I would see how many TLAs I could get in a title while still writing a coherent post. I managed 9 (yes, I know, BI is not a TLA technically but whatever):

• BPM or Business Process Management is about defining, managing and controlling the business processes that underpin your business
• BPO or Business Process Outsourcing is about contracting with someone else to run one of these business processes for you
• BI or Business Intelligence is about understanding your business by analyzing the data you have
• CPM or Corporate Performance Management (sometimes called Business Performance Management) is about monitoring the results your business is achieving through analyzing the data you collect
• SOA or Service-Oriented Architecture is an approach to building an application architecture from loosely-coupled component services
• EDA or Event-Driven Architecture uses events and the responses systems take to these events as the primary organizing principle of systems
• CEP or Complex Event Processing involves correlating many events, often related to different business processes, and then automating an appropriate response to these events
• BAM or Business Activity Monitoring alerts businesses to problems, issues, goals met or other indicators of how well a process is executing, typically in real-time

So now all I have to do, having ransacked Wikipedia for definitions, is tie all this together

• If you automate a business process with BPM, how do you get straight-through processing if people must make all the decisions?
• If you outsource a process with BPO, how do you keep control of the critical decisions in that process?
• If your BI systems tell you what worked in the past, how do you apply that to decisions you will take in the future?
• If your CPM environment tells you something is going wrong, what decisions can you take to respond?
• If you are using SOA to be more agile, what happens when a service makes decisions that must change often?
• In your EDA, are you just going to tell people to act or are your systems going to take a decision to act in response to an event?
• Once you have correlated your events in your CEP system, how do you decide what should be done?
• When your BAM dashboard tells a manager you have hit a goal, they can change their decisions but how do they change the decisions taken by their systems?

Decisions, decisions, decisions. And that brings us to the 9th TLA - EDM or Enterprise Decision Management. Enterprise decision management, or decision management, is an approach for managing and improving decisions. It involves separating out the operational decisions in your environment, automating them using business rules and predictive analytics, and then managing and adapting them over time to ensure they reflect changing conditions. As you would expect, these are topics I have written about a lot. You might start with this post on decision services as they are key to embedding automated decisions in your application architecture. Regardless of where you stand on SOA and EDA you should check out this post on SOA and EDA and why decisioning complements both and this one on reasons to automate decisions when adopting SOA. I also wrote this article on rules and SOA and this post on being event-driven and decision-centric. There's a fair bit on the blog about the intersection of BPM and BI and I wrote this article on how business rules can be a platform for bringing BI to bear on BPM. I have blogged about why rules are needed in CEP and on how decisioning complements BAM as well as this article on shifting your CPM into action. There some stuff on how rules and decisioning can make BPO work better and some of this is summarized in this post about driving overall agility. There's also a lot more on these topics and others on my other blog, www.edmblog.com, to which you can subscribe here.

Phew. I am worn out by all these acronyms.

Technorati Tags: adaptive control, automated decision making, BAM, BI, BPM, BPO, business process, business process outsourcing, business rules, CEP, decision automation, decision service, EDA, EDM, enterprise decision management, event-driven architecture, predictive analytics, SOA, service-oriented architecture

Posted by jtaylor in Business Activity Monitoring • Business Agility • Business Intelligence • Business Process Management • Business Process Outsourcing • Business Rules • Decision Technologies • Event Processing • Predictive Analytics • SOA | Permalink | Comments (3) | TrackBacks (0)