CIOs, CFOs, and CEOs and Data Cleansing

This is my last full day in Las Vegas for Collaborate 12

I certainly didn’t win at the tables but I did win at the Collaborate 12 conference.

Old friends, new friends, new contacts, hopefully future customers.

MDM is on the rise and it is apparent at the conference.

Thank you to all the Oracle users that had questions this week.

One thing I did observe yesterday was a lot of people had full support from the CIO but…..

The CEO, COO, and the CFO didn’t feel the same passion.

MDM is going to save those people a lot of money and prevent a lot of headaches but they just don’t understand how important it is for them to figure out “what the heck MDM means”.

MDM doesn’t work without great defined business driven data governance coming from those people.

What doesn’t happen is the realization of how important MDM data is to an enterprise.

I saw many industries asking common questions this week.

Oracle Product Data Quality (formerly SilverCreek) was compared to Informatica a lot.  I don’t really know how the Informatica product compares to the Oracle product.  I do know that the Oracle product has had years of building universal product cleansing rules that would work for all product data.  I am assuming the same for Informatica.

Product data cleansing is different than customer data cleansing.   I can go into detail but you need to inquire about the differences yourself.  But at the highest level, you should take a look at the fact that customer relationships and data structures are far more static than product data.  Customer data content is typically the same basic core structure.  Product data can depend completely on the industry, the development processes, the definition of the product, the market channels, regional and global details, etc.  Product data can actually be defined as a result of a good MDM initiative where the customer data is typically cleansed and related in a good MDM initiative.

Experts will agree, disagree, elaborate, or admonish my opinions above.  Either customer or product MDM initiatives are not harder or easier than the other as a general rule.

As a consultant I would advise my client to run any data cleansing and data governance tool through a real test with actual product data.  Whether you pay for this test or whether you get it done during product selection is on you.  But, I would run my data through the tool and see which one gives you the desired results.  The one the does the best job wins.  Ensure that there is no real development effort in this endeavor so that the tool provides a means for business product expert to adjust without code.

Don’t buy a tool that is not easy to setup and use with your data.  It is not the software vendor’s responsibility to extract your data from your legacy system.  Once you have the data extracted into a modern file format like txt, csv, etc, then the tool should be able to pull the data in and analyze.  The tool should suggest changes and accept your feedback.

What is fantastic is if the tool will then deliver the product data cleansed into your PIM Data Hub instance without more code to be written.  Then it can be distributed in a manner that is according to the intense data governance policies that business has instilled into the entire organization.


















Posted in Extending Oracle PIM Data Hub Functionality, Oracle PIM, Oracle PIM Data Hub, Oracle PIM Data Hub Interface, Oracle PIM Implementation Activities, Oracle Product MDM, PIM Data Hub Infrastructure | Tagged , , , , , , , , | Leave a comment

Chicken or the Egg? MDM first then BI

This article is heavily based on the extremely informative presentation made by Dan Power, President of Hub Designs and Haidong Song, Product Strategy Director of MDM at Oracle.

The presentation was the best high level representation with specific instructions to organisations on how to conduct data governance.  Get this presentation at OAUG Collaborate 12 proceedings.

#10471 – Data Governance: Start From Where You Are

Before you call anyone to come and help you get your MDM intiatives started and organized, read this presentation!

The speakers were eloquent and accurate in presentation.  They answered the questions quickly and accurately (and realistically).

As a delivery practicioner, I was very impressed with the assembly of the content of this presentation from many different sources.  I have met Dan Power in the past.  I consider him to be a guru level advisor regarding all things MDM.

This presentation is the best “light” reading for executives and project initiative sponsors to date as far as I am concerned.  I believe my presentation on Sunday was part 2 or 3 to Dan’s and Haidong’s presentation on Monday.

I have already ask for my copy of the presentation because I am going to read and memorize the presentation.  I would always refer people to Dan for seeking expertise.  But in addition, I would recommend obtaining this presentation from OAUG, reading it, and seeking Dan and Haidong for questions afterwards.  Since Dan makes a living helping organizations with MDM planning and initiatives, I would advise not to try to dominate a lot of time before simply engaging his services.

For more product specific presentations, I would recommend my presentations as well as a few others:

Other speakers: If I didn’t include your presentation in my list above, my apologies.  I am directing people into the content that I have personally reviewed and enjoyed.  It is no reflection on the quality of writing by the other presenters at Collaborate 12.

The title to this article is because of one important quote that I will probably misquote.  I really liked the statement that I will approximately repeat and reiterate from Dan Power, “You should probably put in your MDM initiatives first before you start your BI (business intelligence) initiatives. ” That is the rough end of the quote.  I will continue with my own opinionated elaboration.

This is not to say that BI is not also relevant to other areas and data not covered by MDM hubs.  But, to be fair, instilling great data governance practices into your organization will lead to far more focused and decision driven business intelligence.  The details of master data management will enable business intelligence reporting versus hinder the BI.  The success of an MDM initiative from top down will lead to successful BI not the other way around.

Dan Power writes for several publications including his own.  You can find him at or the online magazine

I should be selling my services on the blog.  But technically by leading you towards services like Dan Power offers, I am preparing you for a successful Oracle based MDM technical solution for your purely business driven intiative.

I hope this helps.  I am now going to go have fun at Collaborate which happens to be in the sleepy little town of Las Vegas.





#10471  – 数据管理:从你身在何处。

你叫任何人来帮助您得到您的MDM intiatives启动并组织之前,请阅读此演示文稿!








这并不是说,BI是不是也与此有关的其他区域和数据不包括MDM集线器。不过,公平地说,灌输伟大的数据治理到您的组织的做法将导致更集中驱动的商业智能和决策。主数据管理的细节,使商业智能报告与阻碍了BI。 MDM主动从上而下的成功会导致成功的BI而不是其他方式。






Posted in Oracle PIM Data Hub, Oracle PIM Implementation Activities, Oracle PIMDH, Oracle Product MDM | Tagged , , , | 1 Comment

Functional versus Technical: Why Functional is more successful!

There are several schools of thought on methodologies to implementing a proper product MDM solution with Oracle PIM Data Hub.

I divide them into two different schools. Although both should have the same outcome, they don’t currently.

Technical waterfall style implementation projects fail or appear to be successful but are truly failures.

Business driven iterative implementation projects take longer typically but are far higher chances of success and long term benefit for organizations.

PIM initiatives are the business’s way to taking technology and making it work for them.  Not the other way around.  If your PIM Data Hub project starts with technical resources versus pushed from the top down as a required initiative then you are certain for failure.

How can a term defined by wikipedia as stated in a previous article show predominately managed by IT?

“processes and technologies focused on centrally managing information about products, with a focus on the data required to market and sell the products through one or more distribution channels. A central set of product data can be used to feed consistent, accurate and up-to-date information to multiple output media such as web sites, print catalogs, ERP systems, and electronic data feeds to trading partners.”

Who in your organization sets policy for your product data?

You need to answer that question honestly.

Oracle PIM Data Hub is not a development project primarily.  It is a tool to satisfy a very serious business initiative.

I’ve been on some projects where business pushed it off to the IT people to implement.  Bad business people!! shame on you!

I know it is overwhelming and intimidating to take on such a big project but in your own personal lives you have done more complex tasks.

You have to start controlling and governing your product data.  You are somehow doing this today even if it is with multiple systems.  Somewhere in the past you have given business policy to IT and told them to build a program to do that work.

You need to start the process of establishing true uniform product data.  The best place to start is at the beginning of the PIM project, not at the end.  If you need help, ask for it.  Acting like you know what you are doing or overplanning a fairly less complicated implementation can get you into trouble or your boss into trouble.

Functional processes on an implementation are business driven, not technology driven.  You should be able to identify the work that you need done technically.  Any gaps in the Oracle PIM Data Hub functionality that you as a business have identified through the thorough process of knowing the functionality meets your business requirements in detail will be much more accurate than gaps that are “guessed” in advance of this effort.

Don’t leave it up to your IT staff nor contractors to decide your product data policies and processes.  You need to understand them and how they will exist in the new Product Hub environment in detail yourselves.  That doesn’t mean that you are learning how to write code or other technology skills.

You need to learn as a business how to configure off the shelf systems like you know how to build spreadsheets and documents.  Configuration of “off the shelf” systems are not that complicated anymore.  They are uniform skills and are not different across other organizations.  Oracle PIM Data Hub is the same peice of software for everyone with the same set of configuration steps available to use.  There are exceptions with some of the verticals but the methods are all identical and practically important to all.

A functional specification for a solution to resolving a gap in Oracle PIM Data Hub functionality written by the client with strong support for the PIM Data Hub functional experts will produce faster development and better unit testing than one written without deep basic configuration knowledge from the client themselves.

It takes at least three to four months on average for an organization to gain PIM Data Hub design knowledge in relation to their own product data for a high quality functional specification to be written by business.

Exception: If you know you are going to build interfaces, then it is possible to start solutioning a technical architecture early in the project.   But the data mapping comes from the business gradually so you must plan to evolve these types of specifications.

Thanks for the reading of the article!


有几所学校的思维方法,以实施适当的产品MDM解决方案与Oracle PIM数据平台。









Oracle PIM数据平台的不是一个开发项目为主。它是一个工具,以满足一个非常严肃的倡议。





功能上的实现过程是业务驱动的,而不是技术驱动。你应该能够识别你需要做技术上的工作。 Oracle PIM数据平台的功能,你作为一个企业通过深入的过程中知道的功能已经确定的任何空白,满足您的业务需求的详细比的差距,“猜”在这方面的努力将更加准确。


您需要了解作为一个企业如何配置现成的系统,如你知道如何建立电子表格和文档。 “现成的”系统配置不那么复杂了。他们是一致的技能和其他组织之间是不一样的。 Oracle PIM数据平台的是一样的peice,软件的每个人都可以使用一组相同的配置步骤。有一些的垂直,但有例外,均相同,并实际上重要的所有的方法。

Oracle PIM数据平台的大力支持PIM数据平台功能专家编写的功能由客户端的解决方案,以解决差距的功能规范会产生更快的发展和更好的单元测试,而不是一个没有深刻的基本配置知识编写自己的客户端。




Posted in Extending Oracle PIM Data Hub Functionality, Oracle PIM, Oracle PIM Data Hub, Oracle PIM Data Hub Interface, Oracle PIM Implementation Activities, Oracle PIMDH, Oracle Product MDM, PIM Data Hub Infrastructure | Tagged , , , , , | Leave a comment

Collaborate 2012 Las Vegas

Collaborate 12 Las Vegas Nevada

April 22-26, 2012
Mandalay Bay
Convention Center
Las Vegas, Nevada, USA

I am here and enjoying a nice room service breakfast.

I am in Las Vegas Nevada USA for the Collaborate 12 Oracle Application User Group Conference.

April 22-26, 2012
Mandalay Bay Convention Center
Las Vegas, Nevada, USA
I am speaking twice during the conference on two different subjects.  If you are a reader of the blog (which I hope you are regularly reading), you will recognize some of the topics within my discussions.
Today on Sunday 22 April 2012, I will be speaking on the planning of a PIM Data Hub implementation.
On Wednesday, 25 April 2012, I will be speaking about item integrity rules.
I hope you can come by and see the presentation.  Please give me a good presentation rating because those are very important.  It seems my competition has been very unfair on their ratings in events past.  I need the support please..  I confess.  I am probably a better writer than presenter in communicating my desires to you for success.
But if you cannot attend, please get a copy of the proceedings.  I am sure everyone is going to ensure they have joined the OAUG at
Have a great week!
Viva Las Vegas
Posted in Oracle PIM, Oracle PIM Data Hub, Oracle PIMDH, Oracle Product MDM | Tagged , | 5 Comments

Is it worth doing? PIM Data Hub Implementations

How do you get value from MDM initiatives like Product Data Hub?

Why buy the software?

It usually takes twice the amount of money minimum to implement PIM as to buy the software.  Is it worth it?

It can be worth it.  It depends upon whether you’ve used the right criteria to make your decision.  There are lots of nice graphic arts out there showing what PIM as a process will do for your organization.  However, I believe the PIM Data Hub solution is a business driven process with a technical solution.  Therefore, it is more expensive within than on the surface.

Wikipedia does a good high level job in defining “Product Information Management” and the principle consumers.  I am going to copy and paste the definition and support it within this article to attempt to help you purpose a PIM Data Hub initiative if you find you need one.

Wikipedia link above states: “Product information management or PIM refers to processes and technologies focused on centrally managing information about products, with a focus on the data required to market and sell the products through one or more distribution channels. A central set of product data can be used to feed consistent, accurate and up-to-date information to multiple output media such as web sites, print catalogs, ERP systems, and electronic data feeds to trading partners. PIM systems generally need to support multiple geographic locations, multi-lingual data, and maintenance and modification of product information within a centralized catalog to provide consistently accurate information to multiple channels in a cost-effective manner.”  The definition continues into another paragraph that states different channels of product data to the last sentences which state: “PIM in this example represents a solution for centralized, media-independent data maintenance for providing purchasing, production and communications data for repeated use on/in multiple IT systems, languages, output media and publications. It also provides a solution for efficient data collection, management, refinement and output.”

On a high level I can break down the above paragraph.

1) Business process driven implementation using a technology solution.  More business involvement with less IT solutioning (off the shelf hub software).

2) Centralized system focusing on the information for marketing and selling products, but not forgetting the rest of the product data.  It is a centralized set of product data.

3) Sending this product information to other distribution channels.

4) Supports multi languages and multiple geographical areas

5) Product data collection, maintenance, and management.

6) The system is feasible to use, maintain, implement, understand, integrate, and other mission critical cost basis.  It should be technically feasible to implement.  It should be proven to be possible by other clients who are using it (not always from the same industry).

All of these points describe business processes and standards versus technology.  If your organization is not ready to take these points to the literal, then no hub product will be successful in your future.  You should be doing these things to the best of your ability now.

Product Hub doesn’t have to replace the other systems if they do a great job at product definition for your organization.  However, if the legacy system is too hard to get data in and out of, then consider the basic design of that system to be at fault and give replacement a serious consideration.

Is it worth it?  If you cannot dedicate more internal product experts to learning how to use the system software to meet your product MDM needs,  then you probably should consider cancelling or delaying the project until you can afford to use those resources.

It is worth it if you are losing customers because you can’t keep consistent on your product data to the point where you are producing invalid orders/quotes.

It is worth it if you have very complex products that require lots of rules in relationship to product feature and combinations of features.

It is worth it if you have very complex rules relating to who can buy your product by geographical, language, size of customer, customer type,  or other product obligations like royalties.

It is worth it if you have many front end product creation systems that need synchronized often to eliminate redundancy.

It is worth it if you have to secure your product data in many different aspects of the same product data file.

It is worth it if you have to deliver a lot of web content data to multiple web based ordering or marketing systems depending on how many and how often.

It is worth it if you are tired of building and managing your own software applications to produce the same functionality that Oracle provides.  This would also need to be measured by the cost of maintaining the software every year and still getting new functionality at the same rate Oracle delivers it.

It is worth it if you are looking to integrate a lot of product change management into the lifecycle of a sellable product.

It is NOT worth it if you need a complex pricing system for your product.  It is worth it if you are integrating to a complex pricing system.

It is NOT worth it if you need it to be a complex design system like Agile/Oracle PLM.  But it is worth it if you want to integrate to Agile.

It is NOT worth it if you have very limited tight budgets and no internal business resources to heavily contribute to the design of the project (hands ON!).  This is not something to throw at the consultants and expect to receive something back useful.

I don’t believe I was “complete” in all the facts stated above by any sense of the word.  But what I have mentioned above including the references to the wikipedia definition, I believe to be true.

I hope this helps you plan a better product MDM project.

Posted in Oracle PIM, Oracle PIM Data Hub, Oracle PIM Implementation Activities, Oracle PIMDH, Oracle Product MDM | Tagged , , , | 1 Comment