Wednesday Nov 21, 2012

When should I use a Process Model versus a Use Case?

This Blog entry is a follow on to https://blogs.oracle.com/oum/entry/oum_is_business_process_and and addresses a question I sometimes get asked…..i.e. “when I am gathering requirements on a Project, should I use a Process Modeling approach, or should I use a Use Case approach?”

Not surprisingly, the short answer is “it depends”!

Let’s take a scenario where you are working on a Sales Force Automation project. We’ll call the process that is being implemented “Lead-to-Order”.

I would typically think of this type of project as being “Process Centric”. In other words, the focus will be on orchestrating a series of human and system related tasks that ultimately deliver value to the business in a cost effective way. Put in even simpler terms……implement an automated pre-sales system.

For this type of (Process Centric) project, requirements would typically be gathered through a series of Workshops where the focal point will be on creating, or confirming, the Future-State (To-Be) business process. If pre-defined “best-practice” business process models exist, then of course they could and should be used during the Workshops, but even in their absence, the focus of the Workshops will be to define the optimum series of Tasks, their connections, sequence, and dependencies that will ultimately reflect a business process that meets the needs of the business.

Now let’s take another scenario. Assume you are working on a Content Management project that involves automating the creation and management of content for User Manuals, Web Sites, Social Media publications etc. Would you call this type of project “Process Centric”?.......well you could, but it might also fall into the category of complex configuration, plus some custom extensions to a standard software application (COTS).

For this type of project it would certainly be worth considering using a Use Case approach in order to 1) understand the requirements, and 2) to capture the functional requirements of the custom extensions.

At this point you might be asking “why couldn't I use a Process Modeling approach for my Content Management project?” Well, of course you could, but you just need to think about which approach is the most effective. Start by analyzing the types of Tasks that will eventually be automated by the system, for example:

 

Best Suited To?

 

Task Name

 

Process Model

 

Use Case

 

Notes

 

Manage outbound calls

 

Ö

 

 

A series of linked human and system tasks for calling and following up with prospects

Manage content revision

 

 

Ö

Updating the content on a website

Update User Preferences

 

 

Ö

Updating a users display preferences

Assign Lead

 

Ö

 

Reviewing a lead, then assigning it to a sales person

Convert Lead to Quote

 

Ö

 

Updating the status of a lead, and then converting it to a sales order

 

As you can see, it’s not an exact science, and either approach is viable for the Tasks listed above.

However, where you have a series of interconnected Tasks or Activities, than when combined, deliver value to the business, then that would be a good indicator to lead with a Process Modeling approach. Another good indicator, is when the majority of the Tasks or Activities are available from within the standard COTS application.

On the other hand, when the Tasks or Activities in question are more isolated, tend not to cross traditional departmental boundaries, or involve very complex configuration or custom development work, then a Use Case approach should be considered

Now let’s take one final scenario…..

As you captured the To-Be Process flows for the Sales Force automation project, you discover a “Gap” in terms of what the client requires, and what the standard COTS application can provide. Let’s assume that the only way forward is to develop a Custom Extension. This would now be a perfect opportunity to document the functional requirements (behind the Gap) using a Use Case approach. After all, we will be developing some new software, and one of the most effective ways to begin the Software Development Lifecycle is to follow a Use Case approach.

As always, your comments are most welcome.

 

 

 

Saturday Nov 03, 2012

The Birth of a Method - Where did OUM come from?

It seemed fitting to start this blog entry with the OUM vision statement.

The vision for the Oracle® Unified Method (OUM) is to support the entire Enterprise IT lifecycle, including support for the successful implementation of every Oracle product. 

Well, it’s that time of year again; we just finished testing and packaging OUM 5.6.  It will be released for general availability to qualifying customers and partners this month.  Because of this, I’ve been reflecting back on how the birth of Oracle’s Unified method - OUM came about.

As the Release Director of OUM, I’ve been honored to package every method release.  No, maybe you’d say it’s not so special.  Of course, anyone can use packaging software to create an .exe file.  But to me, it is pretty special, because so many people work together to make each release come about.  The rich content that results is what makes OUM’s history worth talking about.  

To me, professionally speaking, working on OUM, well it’s been “a labor of love”.  My youngest child was just 8 years old when OUM was born, and she’s now in High School!  Watching her grow and change has been fascinating, if you ask her, she’s grown up hearing about OUM.  My son would often walk into my home office and ask “How is OUM today, Mom?”  I am one of many people that take care of OUM, and have watched the method “mature” over these last 6 years.  Maybe that makes me a "Method Mom" (someone in one of my classes last year actually said this outloud) but there are so many others who collaborate and care about OUM Development.

I’ve thought about writing this blog entry for a long time just to reflect on how far the Method has come. Each release, as I prepare the OUM Contributors list, I see how many people’s experience and ideas it has taken to create this wealth of knowledge, process and task guidance as well as templates and examples.  If you’re wondering how many people, just go into OUM select the resources button on the top of most pages of the method, and on that resources page click the ABOUT link.

So now back to my nostalgic moment as I finished release 5.6 packaging.  I reflected back, on all the things that happened that cause OUM to become not just a dream but to actually come to fruition.  Here are some key conditions that make it possible for each release of the method:

  • A vision to have one method instead of many methods, thereby focusing on deeper, richer content
  • People within Oracle’s consulting Organization  willing to contribute to OUM providing Subject Matter Experts who are willing to write down and share what they know.
  • Oracle’s continued acquisition of software companies, the need to assimilate high quality existing materials from these companies
  • The need to bring together people from very different backgrounds and provide a common language to support Oracle Product implementations that often involve multiple product families

What came first, and then what was the strategy?

Initially OUM 4.0 was based on Oracle’s J2EE Custom Development Method (JCDM), it was a good “backbone”  (work breakdown structure) it was Unified Process based, and had good content around UML as well as custom software development.  But it needed to be extended in order to achieve the OUM Vision.

What happened after that was to take in the “best of the best”, the legacy and acquired methods were scheduled for assimilation into OUM, one release after another.  We incrementally built OUM.  We didn’t want to lose any of the expertise that was reflected in AIM (Oracle’s legacy Application Implementation Method), Compass (People Soft’s Application implementation method) and so many more.

When was OUM born?

OUM 4.1 published April 30, 2006.  This release allowed Oracles Advanced Technology groups to begin the very first implementations of Fusion Middleware.  In the early days of the Method we would prepare several releases a year.  Our iterative release development cycle began and continues to be refined with each Method release.  Now we typically see one major release each year.

The OUM release development cycle is not unlike many Oracle Implementation projects in that we need to gather requirements, prioritize, prepare the content, test package and then go production.  Typically we develop an OUM release MoSCoW (must have, should have, could have, and won’t have) right after the prior release goes out.   These are the high level requirements.  We break the timeframe into increments, frequent checkpoints that help us assess the content and progress is measured through frequent checkpoints.  We work as a team to prioritize what should be done in each increment. Yes, the team provides the estimates for what can be done within a particular increment.  We sometimes have Method Development workshops (physically or virtually) to accelerate content development on a particular subject area, that is where the best content results. As the written content nears the final stages, it goes through edit and evaluation through peer reviews, and then moves into the release staging environment.  Then content freeze and testing of the method pack take place.  This iterative cycle is run using the OUM artifacts that make sense “fit for purpose”, project plans, MoSCoW lists, Test plans are just a few of the OUM work products we use on a Method Release project.

In 2007 OUM 4.3, 4.4 and 4.5 were published.  With the release of 4.5 our Custom BI Method (Data Warehouse Method FastTrack) was assimilated into OUM.  These early releases helped us align Oracle’s Unified method with other industry standards

Then in 2008 we made significant changes to the OUM “Backbone” to support Applications Implementation projects with that went to the OUM 5.0 release.  Now things started to get really interesting.  Next we had some major developments in the Envision focus area in the area of Enterprise Architecture.  We acquired some really great content from the former BEA, Liquid Enterprise Method (LEM) along with some SMEs who were willing to work at bringing this content into OUM.  The Service Oriented Architecture content in OUM is extensive and can help support the successful implementation of Fusion Middleware, as well as Fusion Applications.

Of course we’ve developed a wealth of OUM training materials that work also helps to improve the method content.  It is one thing to write “how to”, and quite another to be able to teach people how to use the materials to improve the success of their projects.  I’ve learned so much by teaching people how to use OUM.

What's next?

So here toward the end of 2012, what’s in store in OUM 5.6, well, I’m sure you won’t be surprised the answer is Cloud Computing.   More details to come in the next couple of weeks! 

The best part of being involved in the development of OUM is to see how many people have “adopted” OUM over these six years, Clients, Partners, and Oracle Consultants.  The content just gets better with each release.  

I’d love to hear your comments on how OUM has evolved, and ideas for new content you’d like to see in the upcoming releases.

About

OUM Logo
The Oracle® Unified Method (OUM) is Oracle’s standards-based method that enables the entire Enterprise Information Technology (IT) lifecycle.

Read:

· Brief

· White Paper

· Customer Program Data Sheet

Connect:

· OUM on Twitter

· OUM on LinkedIn

Search

Archives
« November 2012 »
SunMonTueWedThuFriSat
    
1
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
22
23
24
25
26
27
28
29
30
 
       
Today
Feeds