BPM.com
  1. Peter Schooff
  2. Sherlock Holmes
  3. BPM Discussions
  4. Thursday, 29 June 2017
  5.  Subscribe via email
What do you think is more important to the digital business platform: process or data or something else?
Accepted Answer Pending Moderation
As processes are the means to solve your customer's problems, to me it will always be process. But to execute, manage and improve processes (for example with support of a dbp), data is (next to people, supplies, supporting tools, a way of managing) one of the biggest enablers.

Wrote in detail about it here: http://procesje.blogspot.nl/2016/05/does-new-gold-work-for-your-processes.html
Sharing my adventures in Process World via Procesje.nl
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 1
Accepted Answer Pending Moderation
People, Process, Data... you can't have one without the others.
Co-founder of Skore
Comment
Love and Marriage...
  1. more than a month ago
  2. BPM Discussions
  3. # 2
Accepted Answer Pending Moderation
What's more important for survival: air or water?
CEO, Co-founder, Profluo
Comment
  1. Emiel Kelly
  2. 1 year ago
  3. #4156
After being under water for 5 minutes, I'd like to say air ;-)
  1. more than a month ago
  2. BPM Discussions
  3. # 3
Dave Duggal
Blog Writer
Accepted Answer Pending Moderation
Two sides of same coin. Processes consume and create data. Events/state changes can instantiate processes. Metadata and real-time state can drive processes.
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 4
Accepted Answer Pending Moderation
Data + Processing = Information

(you need both).
Both need to be inventoried and treated as a reusable resource.
References
  1. http://www.phmainstreet.com/mba/ss090813.pdf
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 5
Accepted Answer Pending Moderation
Data is a consequence of process but process has data at core of the Platform.
More important is the architecture which binds them together to deliver on ALL business requirements AND readily understood just how this is achieved to put business in control of their processes not IT.......at fraction of cost....
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 6
Accepted Answer Pending Moderation
Businesses succeed or fail on the basis of decisions they make, relying on knowledge, experience, intuition, information, data etc.

All decisions are made from known knowns, known unknowns, unknown knowns, unknown unknowns.

The main contribution of processes is to improve outcomes via orchestration and governance.


So, data comes first.

Proof of this is some companies with no formal processes manage to stay in business.

The probability of success, all other things equal, increases when a corporation

a) has a set of best practices
b) encourages staff to make consistent use of these
c) staff follows the best practices but is able to deviate from these where necessary/appropriate.

Not much more to it than this . . . .
References
  1. http://www.kwkeirstead.wordpress.com
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 7
Accepted Answer Pending Moderation
Columns 1 and 2 (see @skemsley) on the Zachman Framework, need both. Data informs process, process acts on data, process updates data.
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 8
Accepted Answer Pending Moderation
While data has to be appropriately collected, preserved, and governed, let's face it: there are plenty of systems of record out there whose primary purpose is to deal with that. What BPM brings to the table is a way to make that data live, by pushing it through processes that consume, transform, or update it.

If we take data hygiene and availability as a given, then process is the name of the game for BPM. As long as we're not breaking anything, it's what we do with the data that's interesting.
http://www.bplogix.com/images/icon-x-medium.png
-Scott
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 9
Accepted Answer Pending Moderation
Process and Content Services are both needed to build a complete digital business app. And I would a add third element, Governance Services, which is becoming increasingly important, especially in regulated sectors like finance, health care, and government.
References
  1. https://www.alfresco.com/platform
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 10
Accepted Answer Pending Moderation
There's still no commonly accepted definition of what the term “digital” entails in relation to transformation -or, in this question, platform. In that sense one could also argue that there ought to exist a subset to a digital business platform which then could be a digital process platform.
If you would look at a business platform in a broader sense, than desirable features which could unlock the badge of honor “DIGITAL”, would almost certainly include everything that drives agility, quality and cost optimization. Process or BPM would have to be the frontrunners in that scenario. From there, one could derive the aforementioned subset, the digital process platform. The most most important aspects for digital here, I think, would be the management of business rules (including DMN and BRE) and of course everything related to business data (including process data driven AI).
NSI Soluciones - ABPMP PTY
Comment
  1. more than a month ago
  2. BPM Discussions
  3. # 11
Accepted Answer Pending Moderation
Architecture, by definition.

Thanks,
AS
Comment
@Alexander.. Agree, because in the absence of proper architecture to accommodate workflow and workload management, the organization has a difficult time seamlessly inventorying their process templates, providing decision support to users/machines re which ones to engage and when, posting tasks as these become current along process fragment templates, capturing data, carrying out analyses on the data, being able to consolidate data from outside sources, providing real-time (hopefully predictive) decision support at the task level, based, in part, on the two data sources and based on information\data from corporate Kbases.

Might as well add to this the ability at Cases to accommodate goals/objectives and ways and means of non-subjective assessment of progress toward meeting these.
  1. more than a month ago
  2. BPM Discussions
  3. # 12
Accepted Answer Pending Moderation
Good answers above concerning whether data or process is more important for a #DBP. The answers can be summarized as "data is prior" or "data and process together" or "data and process and architecture all together" or "data and process and architecture and governance, even".

For a really actionable answer, we need more. First let's acknowledge DBP as "technology", which means artefacts which are force-multipliers for human brawn or brain applied to work. So now what is important for DBP-as-technology? What makes one DBP better than another at helping humans do more brain work (i.e. be productive)?

Let's begin then with the idea that DBP-supported brain work means business semantics ( "the semantics of brain work" as opposed to "the engineering of physical work" ).

How can we distinguish a DBP having better business semantics? We need to specify why one set of business semantic artefacts is better than another (meaning a set of business rules, business process models and foundationally, data models). We are now in the realm of ontology, even if we don't call it that. And we can make statements such as "process is logically derivative and/or emergent on top of data", in the same way that "sentences are emergent from words".

From this claim, data is logically prior to process. Data is the "vocabulary" of process. Better data enables better process; one could say you can tell better stories when you have a stable and rich vocabulary. But business data as vocabulary of work and business is part of a synthetic language, constructed substantially by executive fiat -- in other words "unnatural" and "maintained by continuous business investment". The problem is that the investment payoff in business semantics is both a long way from the investment decision and like all languages subject to strange governance issues -- thus investment payoffs from an investment decision in business semantics are very highly discounted and data management business cases are very hard to make. (The MDM problem may be an example of technology market failure.)

Result of continuous under-investment in business semantics: data chaos. Master data management ( "MDM" ) as an insolvable problem. Add to this the sociology of tacit knowledge and domain expertise, and it's a wonder that we have systems as good as they are. Even add to this the idea that data (i.e. vocabulary) is discovered not ex nihilo but in the course of work, i.e. in the course of executing formal or informal processes! (This sounds like a chicken and egg problem now.)

The idea of a DBP-starting-with-data almost suggests a recommendation for a monolithic Soviet-style top-down artefact of technology. But top down is a mistake, because the real world is messy and language is always evolving. For sure data comes first as both engineering and management domain. But recognizing priority doesn't get us very far along to assuring the best from powerful DBP technology -- if we are assuming that someone else will look after the data. My impression is that there is a world to be won. And it has to be won as the pace of automation and digitalization proceeds, otherwise the risks associated with the gap between available and needed business semantics will keep growing.

My own view is that there is a need for leadership at the top, leadership which nevertheless empowers and learns from the edge. This problem is perhaps the core problem of technology today. Who will step up? What is the business model that makes business semantics -- "data" -- a viable business?
Comment
  1. John Morris
  2. 1 year ago
  3. #4160
Tweeted from @JohnHMorris:
#DBP is the future! Good ones need gr8 #BusinessSemantics. But #Investment #risks terrible - http://bit.ly/2s9brbw - @PSchooff @BPMdotcom
@John.. Just wondering if MDM is just another name for "data warehouse"? i.e. removing duplicates, standardizing data (mass maintaining), and incorporating rules to eliminate incorrect data from entering the system in order to create an authoritative source of master data (from Wikipedia)".

Data warehouses, aside from serving as repositories, are not very useful in terrms of providing easy access to users of application systems.

First of all, most apps don't need all of the data in a data warehouse. In respect of the data they need, they don't want/need it in any one format (some apps need to get to individual records, others want to look at multiple records at a time). And, they certainly don't relish having to align data element names to some standard.

This is why we push generic data exchangers, where publishers of data can post data they own and are willing to share on a need-to-know basis. Both publishers and subscribers want to read and write (some subscribers enrich data and become publishers) data using their own data element naming conventions.

Many mission critical apps need to build their own histories - pointing/linking to some remote data store means that the links can break and unless you go to blockchain, you have no guarantee that someone will not change data you are relying on to be "history".

A mess . .




  1. John Morris
  2. 1 year ago
  3. #4164
+1 @Karl blockchain, trust and data stores. As for data warehouse, DW is just one component of the data management universe. Too often point solutions solve a tactical problem but contribute to system-wide entropy. I'd be surprised if you disagree! : )
No disagreement
@John, Glad that somebody mentioned the word "model", because the relationships between data and processes are as the following:
- data models are used to define process models
- process models are used to create process instances
- process instances are used to create (new) data instances
- (old and new) data instances are used to execute process instances
- (old) data instances are used to improve process models
...
@Akexander ..Is there any room in your list of transitions for insertion of "process template" ?

i.e. I build a process, I compile it, creating a template, users (people/machines) see the templates, they engage processing and as and when the first step along the process becomes "current", we have an instance of the process (template).

Useful? Or is this just hair splitting?

Rationale: A user can stream a case record (customer, law enforcement case) onto a template, thereby creating a process instance (process template instance, actually). The routing of the 1st step determines who gets to see the 1st step post to their InTray but nothing goes to to the case history unless/until the user types in data AND clicks on COMMIT.

So, whereas you can destroy process instances simply by backing out of the 1st step (terminate the instance), you cannoot destroy a process template because "others" typically roll out process templates.

Another source or perplexion to many of my customers is the notion of "sub-case"

Say I get an order for 5 custom snowmobiles. I can make 5 cases but then i have to invent a master case to represent the customer order (the customer placed one order, not 5 orders).

Seems better to have the notion of "sub-case" but here whenever you bring in data from the outside world or a user processed some task, they have to navigate to "case\subcase\pathway\step"

It's easy to introduce errors "case\item 1\carry out assembly\ . . ." when the user/machine encodes data to "case\item 2\carry out assembly . . " whereas a graphic workflow clearly showing the pathways for items 1-2-3-4-5 may reduce indexing errors.

This category of Case really is job-shop operations, quite different from an order for 1,000 assemblies, all the same where any incoming, say, parts, are for whichever of the 1,000 need them on a first come first serve basis (i.e the incoming parts go to inventory and the order feeds off of the inventory)
@Karl, sure. There are many dependencies between modes (templates) and instances of data and processes. Thus architecture is more important for DBP.
Karl feels like the process template is a UI form designed for that instance with capability to only engage/submit to main process when all data ready i.e can be "saved"...maybe even shared by other users to complete? The records of activity only recorded on submission and if timed out can escalate as required...or am I missing something...?

Alexander absolutely the architecture of the DBP is very important get that right the rest follows...?
@David.
The process template is a linked set of tasks to be used for all instances. Very common for a user to have 25-50 active cases all using the same process template and at different steps along these. i.e. each record holder gets a private instance of a process template when they stream the record onto the template, OR another way of looking at it, when an instance of a template is loaded into the Case.

In our Cases there is no "main process" - I suppose you could say the Case itself is the "main process" - a new Case typically starts up empty or with a specific default set of process template instances that "worked well" for other Cases, otherwise, the user just sees a menu of process templates and they pick one, or two or more.

Data is saved incrementally, at each step along a process template instance, as and when the user indicates they are "done"- it's not uncommon to have 25 open Cases with some Cases suspended at steps where the user is waiting for a piece of information.

No different for a step waiting for a data import from say, a supplier, advising that a part has been shipped which might prompt a user to start up a "receive shipment" process in anticipation of the shipment.

I will make a brief video, much easier to explain things in a video.
  1. more than a month ago
  2. BPM Discussions
  3. # 13
Accepted Answer Pending Moderation
John
Yes DBP is the future but requires getting back to basics how business actually works..its people and support needs...quite simple. That's the easy bit the high risks come with getting the market to take it seriously as it challenges a mature market dominated by large vendors and their large supply chain. Self interest also resigns internally as business still reluctant to take on IT. But this will change as business understand how it delivers and removes risks of implementation. When this happens the new era of adaptive enterprise software starts...and at fraction of cost of old complex silo hard coded ways....just a question of time. .....
Comment
  1. John Morris
  2. 1 year ago
  3. #4161
+1 @David: "back to basics how business actually works"!!! In other words no "black box" and "make it so No. 2" -- instead, "take responsibility and own your work". It's what Japanese automakers did in the 60's.
@David/John... Not sure I agree that "business management" is "quite simple" - if that were true, why do 80% or so of new business ventures fail?

About the only thing "basic" about business is you need the next piece of business to come in before you run out of money to keep the doors open.

How you evolve strategy, how you fund initiatives, how you track progress toward goals/objectives is, IMO, difficult, not to mention poorly done by most organizations.

Of course, the only real requirement is that you have a competitive advantage, but there are many ways to building/increasing competitive advantage and if a corporation does what all of their other sector members do, they will not end up as leaders of the pack or be successful staying there.

  1. more than a month ago
  2. BPM Discussions
  3. # 14
Accepted Answer Pending Moderation
John and Karl raise good points..
First John...it was Edwards Deming who was credited with turning around Japan's fortunes with his bottom up "systems thinking" which in reality reflects the BPM discipline. Digitisation of operations does indeed achieve putting responsibility into hands of workers and now customers....

Karl
Businesses is simple we have proven only some 13 task types with links address all user needs.....Why new ventures fail is a quite different issue. Sadly UK is not a good place to successfully cross the exploitation gap yet has a very innovative culture....U.S. has culture which sees that opportunity and all credit does support such exploitation for both home grown ideas and those from UK...! As you indicate the key to success is keeping a momentum in sales in the new technology and this has much to do about national culture not so much about management in the venture. I agree getting to that leadership does require that competitive edge ....and new supporters in buyers....and Governments are the biggest...UK has much to learn from US.....
Comment
+1 re bottom up "systems thinking"
@David... Can you provide examples of 2 of your "task types"?
Karl
Here they are....does it all!

First what is a business process? It is series of linked tasks where each task is a “step” in the process. It has a clearly defined start point and one or more expected outcomes.

Next what are these “tasks”? Our research established them as follows;

• The Form task is the task that the user will be mostly concerned about. This is where data is entered into and extracted out of the database. It can be a “simple” display form or a complex interactive form. This superseded by the web report/form task below but used for quick first cut/ prototype of the application.

• The Web report/form task is used to hold the path for Java Server Pages/forms to run across the web. Utilises Ajax to ensure once only entry of information with intelligent grids. This includes a Tag Library and ability to directly to data stores as required

• The Normal task “halts” the process for an off-line activity. It is a very useful development aid in a process, but should be used sparingly in a “live” environment.

• The Program task allows you to “call” applications such as Word, Excel and specific program used in a process.

• The Pending task places the process into the “Pending” tray of the user concerned. This is a very useful task that is used alongside deadlines and delays in a process.

• The Report task enables a report to be generated via any report application based on a previously defined template.

• The Calculation task can contain calculations involving almost anything including dates, numbers, algorithms, strings etc and aids rules creation.

• The Sub-process task allows the process to move to another, or ‘sub’, process.

• The Event task bundles the same task together in multiple runs and waits for another process to action it. .

• The VB Script task allows the use of Visual Basic code. This task can have many different functions according the requirement of the developer but only used in client server environments.


• The Import/Export task handles the movement of “bulk” loaded data into and out of the database. It can be both completed by the user and/or the system according the specifications.

• The Server Side Message Queue task handles communication between the defined process and many other external systems, such as legacy systems. It’s more popular use is in the sending of e-mails from inside the process.

• The Finish task recognises the process has ended. As far as the user is concerned, the “run” of the process will disappear from the trays. At this point, that particular run will be placed into the “Process History” tray of the manager.


These task need to be “joined” up to create the flow of work and this requires flexible links. The order in which tasks are processed is determined by the links between the tasks. Each link in a process has at least one source task and at least one target task. When the source task is complete, the link triggers the target task. These links are very powerful in terms of flexibility and conditions as required. This aspect has proven to be a significant contributor to business rules.
@David.. Very interesting. I will read these and get back with comments
  1. more than a month ago
  2. BPM Discussions
  3. # 15
  • Page :
  • 1


There are no replies made for this post yet.
However, you are not allowed to reply to this post.