Skip to Content

Before you are able to optimize a process, you must first understand it. Thus, in every project improvement project, the BPX should try and collect as much information as possible about the existing process. In the BPM methodology, this is called the AS_IS Analysis Phase. The old CAF methodology also has various tools to accomplish this requirement. This phase of data collection usually involves the detailed examination of a process and the attempt to discover what roles are involved, what steps/tasks are involved, what business objects are present, etc. 

Collecting data

The question is how do you collect this information? If you are lucky enough and the process is documented in some sort of tool (SAP Enterprise Modeling Applications by IDS Scheer, etc.), then you can use that as your foundation. This ability to find the process documentation – irregardless of whether it is up-to-date or not – is based on the assumption that the process is even documented.  Depending on the process maturity of the organization involved, the process may also exist without appropriate documentation.   Of course, you must realize that the process documentation may no longer be accurate. It is often the case that the actual process as it exists in reality has evolved beyond the officially documented one.

Irregardless of whether process documentation exists or not, there are various ways to collect additional data. One useful method to collect information about the process is to interview those involved in the process. Such discussions can take place with individuals who are actually involved in the process – who are responsible for particular tasks in process instances. The BPX might also talk to the “process experts” – individuals who may not be actively involved in the daily manifestation of the process but have knowledge of the process at a meta level.

How a historian collects information

Since this information collection is such an important part of process improvement projects, it might be useful to look at how other fields perform similar work in order to improve this activity in projects in which BPXs are involved.

Historians are also forced to collect information on their research areas. If you are a historian and are doing research on in the Industrial Revolution, how do you discover how individuals from that time viewed their lives? One way to go about this task would be to start by reading the books of other historians who had written on the topic. However, you might be interested in “hearing the voices” of those who actually lived during this period – without the interpretation of historians / analysts from a different era.

Since these individuals of interest are all dead, you wouldn’t be able to interview them directly. You might start with newspapers and diaries to get a feeling for the period. However, the problem with these two sources is that they are restricted to a particular class of society that was literate and had the time, energy and means to collect their thoughts in a form that survived over the years.

What about the workers who toiled in the factories under horrible conditions. I assume that there are few first-person impressions that describe their daily experience. If you were interested in the impressions of their lives, you would probably have to rely on impressions that were collected indirectly. 

To get a more complete view of the individuals in question, the historian must examine various information sources.  The idea is that a reliance on only one source of information would be biased and incomplete. You need a multi-faceted approach to gain the deepest understanding of the object in question.  Of course, there is no assumption that all data sources are of equal importance or relevance. Depending on the subject being examined, one data source may provide higher quality information than another. It is, however, critical to realize that these various sources are complementary to one another.

The following points can be learned from historians:

1.   Collect information from as many different sources as possible to get a more complete picture of the analyzed object in question

2.   Try to find ways to directly capture the “voice” of those individual in question rather than depending on third person analysis.

If we apply these points to the job of data collection for BPXs, we get the following picture that gives an example of the use of various data sources.

Why is this multi-faceted approach important?

  • Because BPXs often rely on one source of information about existing processes rather than using multiple sources.
  • The analysis phase in most process-related methodologies is usually not described in any detail – leaving much of the data collection choices to the BPX.

Of course, some of these data sources are implicitly mentioned in process-related methodologies. What usually happens, however, is a dependence on the more traditional sources (SAP Enterprise Modeling Applications by IDS Scheer, etc.) rather than going into the field and actually interacting with those involved in the process.

This “field research” is especially important when finding out how processes are really experienced by those directly involved in them. A process may be documented in its “ideal” state but those who are directly involved in the process may have created various exceptions to this ideal state. Unless you interview individuals at various levels in the organization, you may never discover what really happens on the shop floor. As we know, attempts to optimize the ideal process rather than the existing process are often destined to failure or least resistance.

Note: We had a long discussion regarding the importance of capturing these exceptions at the recent Sapphire in Orlando in a NetWeaver Strategy Session Business Innovation Life-cycle.

Data collection in a Web 2.0 world

Compare the difficulty of the historian collecting information about the Industrial Revolution with the job of present and future historians who will be asked to analyze the present Web 2.0 society where almost everyone – irregardless of their “class” – can document all aspects of their lives for future generations to read.

Thus, modern technology has generated an explosion of information that will allow future generations to have better insights into the existence – irregardless of how banal – of individuals.

How can and will these new Web 2.0 (or Enterprise 2.0 ) technologies influence the job of the BPX who is performing data collection.  Let’s take a quick look at the impact of these new tools through a focus on one particular technology: microblogging.

Micro-blogging is a form of multimedia blogging that allows users to send brief text updates or micromedia such as photos or audio clips and publish them, either to be viewed by anyone or by a restricted group which can be chosen by the user. These messages can be submitted by a variety of means, including text messaging, instant messaging, email, digital audio or the web.

The most popular representative of this technology is Twitter – there are other examples of the microblogging in/for the enterprise including Yammer, ESME, etc.

Marco ten Vaanholt recently wrote a Why twitter is not (yet?) all that valuable for BPXers ( and others) about Twitter where he questioned its value as a tool that exists outside the organization. I’d like to suggest that such microblogging and other Enterprise 2.0 tools will soon be a part of every organization

I’m not agreeing or disagreeing with Marco – I’d like to take a different perspective on the whole issue. I’m going to explore the use of microblogging and other Web 2.0 tools for data collection purposes in process improvement projects.  The BPX might mine such tools to find out how individuals in an organization view particular processes

What are some of the issues that might be involved with such usage?

  • Because workers are associated with a multitude of processes in their daily lives, it would be necessary to somehow associate individual messages to particular processes. This might be achieved with tags that represent particular processes.  
  • People might respond differently if they know that BPXs or others might use their tweets in process improvement projects. Especially in countries with strict personal privacy-related policies (for example, Germany), this usage might be controversial
  • Using Twitter as example of how many tweets individuals produce on a daily basis, the sheer number of tweets to analyze would be a job for heavy-duty search / analysis technology. Business Objects is an obvious choice here.
  • There is no requirement that process-related tweets really contain useful information. “I hate invoicing” – doesn’t really provide information that is useful to the BPX – other than a general indication that the process might need improvement.
  • Individuals can but there is no mandatory rule that they must create messages about a process. Thus, data collection based on microblogging and other Web 2.0 tools might be skewed towards a certain type of process user. Other types of data collection would be necessary to supplement microblogging-based data.


In most process improvement projects, BPXs are acting more as historians who are looking at the Industrial Revolution rather than historians looking at modern society. When BPXs collect information, they usually look at material that is indirect to the individuals rather than talk to the individuals themselves. When they talk to these individuals they usually work through intermediaries rather than conversing directly with the individuals and the shop floor and capturing the process in their own words.

Of course, the use of complementary data collection methods / sources has a certain disadvantage in that each of the data sources usually ends up as an artifact in different format.  If these artifacts remain separate, then the situation in process improvement projects becomes more complex, because those involved have no single view on the process. It is necessary to analyze these different artifacts and create one harmonized view – usually in the form of a process model with the appropriate documentation.

Note: We also discussed strategies for the usage of artifacts in process improvement projects at the recent Sapphire in Orlando in a NetWeaver Strategy Session Business Innovation Life-cycle. In this session, we also discussed whether the data collection task is partially the responsibility of the Business Analyst role rather than the Process Modeler. Irregardless of which role performs the task, I was pleased to see that those designing the next generation of SAP process-related tools are well aware of the importance of multi-faceted data collection and are examining various ways to support it.

To report this post you need to login first.


You must be Logged on to comment or reply to a post.

  1. Former Member
    Hi Richard,

    Thanks for your interesting blog.  While Web 2.0 tools allow easy creation of data by “masses”, the challenge is how to turn it into useful (process)insights. For adhoc (situaltional) processes possible such data collection combined with analytics can provide useful insights, but actually improving process execution is hard without having the needed details. In case its a interesting thought as to how live feedback can be attached to running processes and later/dynamically used to optimize them.


    1. Richard Hirsch Post author
      I like your idea of attaching real-time data from corporate Web 2.0 data to running processes. Since you can attach BI-based data to processes, you could feed Web 2.0 data to BI to use the same mechanism.


  2. Former Member
    Hi Dick,

    Many thanks for this interesting insight. I particularly like the comparison with the historians. I personally have always seen any analysis work (which is what they used to call BPXers, right?) as a bit of a “private investigator” job.

    I totally agree on getting as many angles as possible. This sort of approach also can strengthen user acceptance later. You always need to try to talk and involve as many people as possible (obviously to a degree).


    1. Richard Hirsch Post author

      Exactly. The idea is to involve as many people as possible. The main difficulty is moving all this data into one single cohesive format. I don’t whether BPMN by itself is enough – somehow you have to supplement the model with other documentation.



Leave a Reply