When we talk about business or enterprise social networks, we expect to get value out of them from the level and quality of knowledge exchange and interaction.

In other words: is there a significant share of the Community posting on the network? Is the content posted interesting enough to engage post readers in replies, comments, discussions? At the end, this is one of the main goals when we establish this kind of software in our company, or when we engage in a public community.

I tried to analyse some groups on Linkedin, surely a big professional social network. The Group is an area where people can find each other based on some aggregation key, like a specific topic of interest, or a professional status, or alumni, etc.  People participating in a Group doesn’t know each other, so the level of interaction among people reflects the interest raised in people. To be more precise:

  • the number of Group’s members reflects the interest for the Group topic,
  • the number of Group’s posts and comments reflects the engagement level.

Now, some examples:

An Industry Association Group

Statistics about Information Technology - Community Assinform - Associazione Ita_2014-03-14_18-29-41

This is quite a large Group, with a relevant share of top level professionals. It’s growth rate has been constant over time, but not terrific.

Members are mainly located in Italy (single Country).

Statistics about Information Technology - Community Assinform - Associazione Ita_2014-03-14_18-32-31

Observing the activity graph above, we can see that the activity is low (one post each few days), and that the ratio of comments/post is low.

Looking at the Group activity, we see a large share of events announcements, and also that the number of frequent contributors is limited; one of them is the Industry Association itself.

The synthesis? This is a a Group that attract people, it’s a communication channel for the owner, but it’s worth looking at it not more than once a week.

A “topic centered” Group

This is an interesting example.

Statistics about Strategie competitive e modelli di business _ LinkedIn_2014-03-14_18-43-43

The number of Group’s members is huge, largely in the entry/middle professional levels. The growth rate was impressing, with an average of 2% week on week.

Statistics about Strategie competitive e modelli di business _ LinkedIn_2014-03-14_18-46-59

The above graph shows an impressive ratio of comments/post that is frequently above 100.  The frequency of posts is not high, an average of one per day.

To me, this appear as a successful initiative of the Group’s owner. Some people push forward the discussion, and the topics posted are clearly interesting. We can expect that many members are ready to post their comments because they are happy to show their professional attitude (and this is not bad), but their comments are meaningful, because a lot of other people engage in the discussion.

Looking at the posts sequence, we can see that, although there are some top contributors, there is a great number of different people that post to the group. It is really a Group, rather than a showcase.

During the last two years, in my company, we have started a series of projects to enhance our SAP ECC platform, in the direction of business integration across the software modules.

This happened because we have spotted some limits in the architecture of this ERP system. SAP is a well-integrated software suite, in his main areas (Financial, Logistics, Sales, Real Estate, HR, etc.) it offers a bulk set of functions, that cover the typical business processes in a very wide set of industries. So, where are these limits?

I will try to suggest some ideas and discussion topics across the posts of this blog.

My company operates in the Construction and Real Estate industry. Our business processes start from the acquisition and development of lands, until the sale of residential properties or the management of commercial properties that are leased out. SAP ECC was installed in late 2005, replacing a previous Italian ERP solution. Initially, SAP was not a successful project. Since 2008, we have started a complete re-engineering of SAP implementation, changing the design from a module/office centric design to a business-centric design.

Let’s start now from a small development that we are carrying on now: cash management by an economic business perspective.

We have a very basic configuration of the CM module: bank and cash accounts, customers and suppliers accounts, are classified according to categories that are relevant for our business. With these settings, we are able to perform a basic analysis of financial flows; this is important, because our business relies heavily on the financial supplies from banks, and our finance dept. needs to monitor the cash levels versus the financial planning.

Unfortunately, the simple analysis of bank accounts doesn’t allow us to understand how the sources and destinations of liquidity have influenced their balances. So we need to drill down this analysis across our business units, and across the development projects that are active.

A good perspective of our business is given by the Profit Center hierarchy: it is roughly divided in the following areas:

  • commercial real estate
  • residential real estate
  • property development
  • operating expenses
  • corporate finance

Each area is divided according to the management perspective of the underlying business: residential real estate, as an example, is detailed at building level; several buildings are grouped together according to the land on which they are built.

We are simply building a set of reports that allows our financial analysts to separate incoming and outgoing cash flows according to the building, building part, or single controlling-relevant business element that originates the costs or incomes from which the income or payment comes.

The key of those reports is a function that, given a cash management relevant posting, goes back to the invoice or other FI document, with a controlling destination, that is paid or collected.

I will discuss in some following posts why this solution fits our needs. The point now is the importance to link one area of the ECC system – Controlling – with another area – cash flows analysis, giving the financial analyst a perspective that is the same shared by the financial planners, the production manager, etc.

I am playing with an excellent platform for knowledge acquisition and architecting: Protégé from stanford.edu.

My main interest is the creation of OWL knowledgebases, and the application of Description Logic. The KISS project is maybe my greatest effort in this area.

Protégé is an excellent system for the knowledge engineer, but it is not easy to extract fragment of your ontology in a “user friendly” presentation format. I guessed that a tool like Velocity, from the Apache Foundation, would be a great help, as it allows to create e “template” of your required output, and then fill it at runtime with the java objects from the OWL API.


So I written a simple plugin for Protégé, integrating the OWL API and Velocity. The plugin is open source, you can access it from its website.

Yes.. more samples can help people without specific programming skills. In the past monts I was really busy, having now the responsibility for two functions; I will do my best to providemore documentation. However, remember that Velocity, the tool at the core of the plugin, was intended primarily as a programming tool; it is based on the Java(TM) object model, so, a non-programmer, will find it a little bit tricky.

Modeling standards can be a great resource for System Management. Field Engineers often are more oriented towards “practical” tecniques, underweighting the help that a good model can contribute to a successful management practice.

While I was managing Information System, I have worked with the Tivoli suite – from IBM – and developed some modeling tecniques to represent system events and to analyse the requirements for monitorning consoles. There is a paper, describing this, not updated, but perhaps interesting. I published it online after receiving a lot of requests from the Tivoli mailing list.

Another approach is described in my recent KISS project – described in another post – where I tried to build a knowledge model of the Information System and the related security and compliance issues.

The project uses a Description Logic-based approach for the conceptual modeling of the information system.

During the project I have used several new (form me) technologies, like the OWL language, DL reasoners, and I found this approach really fascinating.

I had confirms that

  • the adoption of XML-based standards allows a great level of interoperability among different stuffs,
  • open source technologies allow innovation with smooth investment curves.

The project deliverables are a fundamental tool, in my Company, for Compliance Risk management (according to the Basel Committee definition), and the conceptual models were used as a design input for a commercial application for Compliance management.

There is a web page dedicated to the project. You can find also another working paper, on conceptualization of norms, in my website.


October 6, 2006

I have just started this blog, as an english version of my hand-crafted Italian Web site at mcoletti.net.

The need of an english version of my website comes from the great majority of english-speaking visitors.

a presto!


Get every new post delivered to your Inbox.