• With 25% churn and complex IT systems, training costs are a huge problem for call centers.
  By automating decisions EDM can reduce the complexity and thus cost of training as a call center representative simply needs to know how to invoke the decision and pass on the answer to the customer, not how to make a potentially complex decision.
• Unnecessary calls to agents because the IVR system is unmanageable increase staffing costs and wait time
  A better IVR system, personalized to a caller and designed to allow automated actions means that fewer customers will want to “get human” meaning fewer staff and shorter wait times.
• Because the website is brochure-ware and does not let customers do anything, more customers call in than is necessary
  A website that supports true self-service, with automated decisions, means that more customers will self-serve and fewer will call in. Automating cross-sell and up-sell on the web even means you won’t lose a sales opportunity.
• Time spent by agents reviewing customer records or querying BI systems to try and determine if a refund can be paid costs real money as well as reducing the quality of experience for customers.
  Automating the decision, not the display of information, speeds response times and so reduces costs while making it easier for call center staff to get the “right” answer.
• Automation that is hard to verify for compliance with regulations can cost money for fines and manual audits after the fact.
  Using business rules to manage decisions makes for easier compliance as business owners and legal experts can directly review the rules being executed. Fewer errors, fewer fines, less costly review cycles.
• Customer service systems require constant change to cope with changing demographics and customer expectations.
  These costs too must be considered part of the cost. EDM helps build agility into the system making it easier and cheaper to change pricing, delivery, refund and other rules without spending on coding or, indeed, on training.

There are more than just these but you get the drift. EDM does not just personalize and improve customer service while building loyalty, it reduces costs too.

Comments

He starts by pointing out that “Successful BI enables the executives and professionals to make better decisions” to which I would add that successful enterprise decision management enables the systems in an organization to make better decisions. I would agree with him, though when he says that this “is a numbers play; presenting them, assisting their assessment, finding issues hidden in them, empowering action to resolve issues, and sometimes automating them…There will always be numbers in corporate BI”.

Cyril goes on to point out that there are various aspects to “Web 2.0″ including finding people with similar interests, sharing experiences, and creating knowledge from tacit resources inside people’s heads. Cyril discusses how these might complement BI but how might they complement enterprise decision management? There are two sides to this – how does EDM use social media outside the firewall and how does it use it inside the firewall.

Social media outside the firewall allows your customers, suppliers, partners, members of the public, whoever add their knowledge and opinion to yours. Once they have done this the question, from an EDM perspective, is what can you do with this information?

• You can mine it for insight

  The growth in text mining technology makes this ever more practical. You can also make it easy for people to build explicit links within their comments, allowing you to mine those linkages directly. For instance, you could use the product links people include along with text analysis of the “tone” of their note to see what cross-sell or up-sell you might use. This is harder than using structured data but increasingly practical. You might also find that the mere fact of participating in social media is a significant data point in customer segmentation models.

• You can enhance results with it

  When you give a person a decision, whether that person is a call center representative or the user of your website, you could include relevant entries from social media with it. For instance, if you recommend a cross-sell you could include positive comments about the product made by other customers.

• You can automate responses to it

  Again, using text analysis, you could decide that certain kinds of comments or interactions or contributions are indicators that outreach is called for and automate email or phone or social media-based responses inviting prolific posters to participate in some special program, awarding loyalty program points based on interactions etc.

  

  Social media inside the firewall allows your staff to collaborate in new ways that are relevant to EDM.

• Manage feedback on the results of a decision service

  Making it easy for those who deliver the results of a decision service to comment on how it was received might give you useful feedback on the results. Particularly when you are using adaptive control techniques to try multiple strategies, this could be very useful as the numbers might point to a particular strategy but your comments might show that it had a very negative impression on those who declined it, something it may be hard to put a dollar value on.

• Discuss assumptions in rules and models

  In the context of decision management it could be very useful to allow managers and analysts to discuss the assumptions inherent in specific rules and models using social media. The “wisdom of crowds” approach might allow you to tap into the tacit knowledge of more workers than the format rule or model review process could.

• Exception handling

  Documenting and discussing how an exception was handled and how that worked can be useful as you try and gradually automate more and more decisions. Understanding how a class of exception could be handled and what the consequences were is critical to getting the more complex situations automated effectively.

• Mashup reporting and decision management

  Allowing users to combine the results of decision automation with reports and search results that clarify or extend the decision as well as peer opinion and discussion might help with adoption and with usage. Being able to see the numbers that should drive a change in the rules, the discussion of how those rules were developed and the rule management environment itself in one place can only be an effective tool.

While it might seem that EDM, with its focus on automation has little to do with people-centric web 2.0 technologies, clearly there is a role for them to be used together.

A final comment – I agree with Cyril that human nature is not really going to change even when the current generation of workers is replaced by those growing up on Facebook etc. Like him I also think “they will be much more ready to participate in corporate collaborative activity” so making more of these kinds of things practical in terms of having a critical mass. I also think that this audience will be much more willing to “mash” things together than their parents are.

“May you live in interesting times”

Comments

• Operational – delivering decisions in operational, transactional systems
• Low latency – something I have discussed a couple of times before in the context of risk management and decision latency in active data warehouses
• Granular – I would say transaction-centric as it is about using a single transaction to drive decisions not aggregating a bunch for reporting
• High availability – obviously these systems cannot go down, even while the way they make decisions is changing constantly

The authors talk about a "BI service at a decision point in a process". Me I call these "decision services" and they are a crucial design element if you are to bring "intelligence" to bear on a process or system. They go on to give three examples and it seemed to me that these are close to what I call Enterprise Decision Management or EDM.

1. Their first example is a TSA system and my question is why bother with the reports at all given the need to "cater to bookings made between the report generation time and flight departure time" that resulted in "a service that accepted the name of a customer and returned potential matches from the TSA lists." This service could have been used as someone booked to flag them in a more event-based way. Indeed, this story illustrates why traditional BI ideas (reports) are so useless in operational systems. The report was produced because producing reports is what people are used to.
2. The second example is an EDM example through and through – rules and analytics working together to recommend actions to the field.
3. The third is another great EDM example – next best action being the primary customer-centric use of EDM (I presented about this at Teradata’s conference, for example). Analytics and rules driving recommendations (actions) while more traditional BI technology is used to display useful contextual information that will help the conversation go smoothly. The feedback loop is another common EDM feature and when the "NBA function was packaged and deployed as a Web service" it is clearly a decision service.

All in all a nice article with some good examples of EDM.

I don’t think that "Operational BI" is the right name for these kinds of decision-centric applications any more than I did when Keith Gile wrote about it or when I consider "BI 2.0" By and large the BI vendors do not understand this kind of system and all the dashboards and reporting tools they have developed are of little use in building them – there are critical differences between BI and EDM. Even the data you collect in your data warehouse may well be aggregated and processed in a way that makes it less helpful when you are trying to build these kinds of decisioning systems. If you think of these as "BI" systems, even "operational BI" systems you will bring the wrong perspective to bear. Think of them, instead, as decisioning systems.

Comments

Tag:

In it Venkat Devraj makes an interesting case for using decision automation, what I would call decision management, to automate database administration tasks. His notes about why scripts are a rotten approach to this kind of automation are perfect and apply to all possible uses of decision management.

Staying on an IT operations front it made me wonder what else might be amenable to decision management. How about these:

• Network reconfiguration when equipment is added/removed/fails
• Engineer assignment to problem equipment using location, urgency, skills etc
• Monitoring the network to see which service level agreements are going to be breached and proactively informing the agreement holders
• Routing helpdesk calls based on source, urgency etc
• Managing backup schedules using rules and predictive analytics to optimize the approach
• Self-service diagnostics for patches and versions to help people set up valid stacks (like Sun Preventive Services for instance)

I am sure there are others but these were just ones that popped into my head based on customers, especially in telecommunications. I am not sure what the prompt for using EDM to manage these might be (in terms of scale) but an IT department that believed in the approach and was having a hard time selling it to the business might consider them as internal proof points.

Comments

Tag:

Add to Del.icio.us | Digg Furl

Bookmark WebProNews:

More than many other businesses, TV ad sales are heavily reliant on telephone negotiations, paper purchase orders and personal relationships. The industry is so old-fashioned that orders for commercials are often processed manually by a TV network

Amazing. I knew TV stations were notorious for being technology luddites but even so… This range of problems seems ideally suited to the technologies and approaches of Enterprise Decision Management or EDM. Business rules are already in use at several media outlets for handling ad pricing. The rules allow even complex sets of conditions to be applied to come up with the right price, freeing employees to spend time with clients rather than grinding through the mechanics of calculating the ad price.

 Business rules also deliver agility, allowing the rules to be changed and managed more aggressively and consistently. Clearly predictive analytics could be applied also to create some dynamic, demand-based pricing. Meanwhile actually slotting the ads purchased into ad breaks is another great rules-based problem, and one where inferencing really matters – check out this story about slotting. Optimization, another EDM technology, could also be applied to manage slotting options. Meanwhile analytics are getting used in keyword bidding and online ad issues are creating some interesting opportunities.

EDM would be perfect for NBC, it even has the same number of letters…

Comments

Digg | Reddit | Furl

Bookmark WebProNews:

About the Author

VP of Product Marketing with a passion for the technologies of decision automation. 15 years designing, developing, releasing and marketing advanced enterprise software platforms and development tools. Across the board experience in software development, engineering and product management and product marketing.

http://www.edmblog.com

I was struck by one comment from their survey in particular:

Seventy percent of respondents want knowledgeable and personalized service, and nearly 70 percent say that first-contact resolution, whether by phone, email, or the Web is also a primary driver of customer satisfaction

I have blogged before about how improved decision-making can make your staff seem more knowledgeable and how it can personalize service. Indeed, speed is one of the dimensions of decision yield for this reason (although the emphasis in Decision Yield is on precision, consistency, agility, speed and cost as a set). First-contact resolution also requires decision automation – only decision automation gives you the control you need while still empowering front-line staff with decision-making power. You get strategic alignment along with front-line automation and speed while ensuring consistency and precision across all your touchpoints.

Good CRM and good self-service need EDM.

Digg | Reddit | Furl

Bookmark WebProNews:

VP of Product Marketing with a passion for the technologies of decision automation. 15 years designing, developing, releasing and marketing advanced enterprise software platforms and development tools. Across the board experience in software development, engineering and product management and product marketing.

http://www.edmblog.com

The article said “Gartner research vice president Nigel Rayner will tell delegates at its Business Intelligence Summit at the end of the month, that traditional thinking around BI needs to change if businesses want to reap any benefits.
Rayner is quoted as saying:

‘BI has been around a long time and people do get returns out of it, but it has been technology and infrastructure driven and is not addressing the needs of senior management,’

Now it seems to me that not only does it not address the needs of senior management, it does not support the needs of operational staff either. Not management, not analysts, but people working on the front lines helping customers. It also does not help the software that support customers.

Neither BI nor Performance Management will make these high-volume transaction-focused systems better. The systems cannot read reports or look at dashboards. The front-line staff have neither the time nor expertise to draw conclusions from data presented to them. Both front-line staff and front-line systems need decisions – actions to take – about customer treatment. Enterprise Decision Management or EDM is about applying your data to these front-line customer treatment decisions to improve them. BI needs to do more on Performance Management for sure, but you also need EDM to put your performance management into action and give you strategic alignment from your KPIs and dashboards to your front lines.

EDM is not BI but it builds on it.

Comments

Digg | Reddit | Furl

Bookmark WebProNews:

VP of Product Marketing with a passion for the technologies of decision automation. 15 years designing, developing, releasing and marketing advanced enterprise software platforms and development tools. Across the board experience in software development, engineering and product management and product marketing.

http://www.edmblog.com