Iconsulting was born twenty-two years ago on the hills of Bologna. To have the idea of the company are two partners, Simone Fiocchi and Federico Ravaldiwhich come from a context close to the research. Iconsulting immediately has a strong verticalization on the topic of data analytics. At the time it’s a bit of a white fly, when in the IT world the management application is the master and topics such as data mesh and data fabric are still unknown.
Today in Iconsulting almost 400 people work and the company has broadened its range of action: in fact it deals with issues of advisory and help companies in the date digital transformation, an all-round digitization process to allow for make data driven decisions come on information content derived from the data and its analysis.
We met Riccardo Piva, Senior Manager of Iconsultingto explore some issues related to the way in which it is possible today get value from data. Specifically, how businesses can benefit from a framework like data fabric.
– How does Iconsulting operate today?
We are technicians, people who join the company developing code. We’ve always been used to hiring people who grew organically within the corporate context. All management was also born from the development of the code and over time it has turned somewhat into a commercial perspective, but the soul always remains technical.
This allows us to provide independent vendor vertical consultancy: We like to say that technology is not a problem. It is true that in recent years there have been many technologies, however we are not bound to any. Instead, we are bound to a context, which is that of data analytics in all meanings which today may have, but across all technologies and methodologies that the market offers and uses.
– Let’s focus on one topic: the data fabric.
For some years, new areas, technologies or methodologies have been born every month with the word given in front of them. Data fabric it came a bit overwhelmingly in the sense that it’s not a technology. It’s a framework which is grafted into several technological contexts. It is something broad, as it is his alter ego, the data mesh. They arrived together, we explored them understanding their differences and points of connection. In fact, the starting point is very similar in both cases.
A company like Iconsulting must be able to capture issues like these and be able to explain them to their customers because very often they are very theoretical concepts, frameworks or paradigms. In these cases, the first step is to search for information available in the literature. After that we start talking about it with customers and collect feedback. We also try to understand how they can be put into practice within an organizationhow this can derive value from it and what are the most interesting points e And more effective in terms of grounding.
– Which customers are you talking about?
We move on enterprise segment because these companies are the ones that they use data more from an analytical point of view. However, we do not address a specific sector. We move from automotive to healthcare to public administration e in each of these contexts we have top management with great experience of the market and the sector itself. In Iconsulting there are also transversal technical figures who bring all the more part of delivery, of technicalities within the various contexts.
However, it should be emphasized that the approach to paradigms such as the data fabric is much broader than that of technologies and you can’t talk about it with all organizations: a maturity of the companies is necessary because they are complex subjects.
If we talk about it to our clients it is because we identify it within their context this type of framework can create benefitscan solve problems. In that case, we treat the topic trying to focus on elements that can give an immediate benefit so that it is very easy to understand.
There are other contexts where it is the customer himself who says he has heard of a certain framework and so asks us what our point of view is.
– You talked about digital transformation. At the basis of this concept there is not so much a purchase of new technologies but a change of mentality to put data at the center of the strategy and therefore move in a data-driven way. In your experience, what culture of data is there in Italian companies?
It depends on the sector, digital transformation how much value the data can actually bring in terms of information and advanced techniques. There are still contexts, including enterprise ones, which are linked to the old way of managing data from an analytical point of view and who therefore find it difficult to make that transition. However, we are no longer at the levels of a few years ago and it is rare that we find ourselves having to explain why analytical data should not be on a transactional system. Today we have a different obstacle: being able to navigate within a chaos of technologies, paradigms and methodologies also very different from each other. And also of professional figures who have common skills but totally different objectives in the field of data analysis, such as those who do, to use an obsolete term, business intelligence and those who actually do data science.
However, regardless of the sector, there is on average in the Italian enterprise segment a fairly high level of knowledge among the interlocutors who still understand the issues. Maybe sometimes they are just interested and have you just need to be oriented in the best way and to find the right compromise because it becomes difficult not so much to understand the concepts but to place them within the single company reality.
– How does the data fabric fit in this context?
As I said, today there are different professional figures, such as i data engineer, i data scientist, i data analyst o i business analyst, who have different goals and objectives within the organization, but who they must be able to work as one in a highly divergent and complex technological environment. The process of evolution of organizations has followed different paths. And this was the starting point from which the desire to rationalize and create a new change in the context of data arose. She was born there need to add concepts that aim to bring order or simply to make data-driven initiatives faster, in terms of time to market quality data and not duplicated as much as possiblewith information content that is easily searchable and addressable by everyone within an organization, that the information within the data is consistent and the data is qualitatively aligned with the reality or the final objective of the business. And that all this it has to happen very quickly and it has to respond to data driven business needs.
The data fabric wants to address these issues according to a sort of natural evolution of the logical data warehousewhich used structured, semi-structured and unstructured data within environments managed ad hoc with specific tools and amalgamated through the data virtualizationwhich could create a common environment for all these very different worlds in terms of consumption.
The data fabric theme is something that descends into the depths of these separate worlds, creating an effect of technological glue superior to data virtualization. It is an item that seeks to technologically and methodologically amalgamate a set of different elements, making them more united and more harmonized. This, from the outside, makes them look like a single ecosystem.
A basic element in this process is the metadata. The information content of the data allows you to answer questions, therefore, it is the information that is needed, but the metadata plays an essential role in obtaining this information.
– In the company, on the one hand there are those who transform raw data into data with the opportunity to be worked on and on the other hand there are those who must take advantage of the information that can be extracted from this data. Very often, however, the one who does one thing is not the one who also does the other. How do you manage to make the two figures work harmoniously?
There is no optimal way, it depends a lot on the context in which we move. We try to understand the needs of the business to translate them into concrete and technical aspects through data. Some tasks are repeatable and automatable thanks to intelligence and metadata. This allows you to create more value with average quality data because data quality check mechanisms can also be automated. And the fact that all systems speak fairly uniformly allows you to create a more transversal value than all corporate worlds, even without a virtualizer on top.
This effect of freeing up space and making the work that is done within the data platform even more automated and intelligent goes for the benefit of business figures that in this way they can spend their time not making ends meet or having to load huge tables, which unfortunately still happens very often, but understanding the processes and optimizing them as much as possible. And that it also means optimizing costs.
– There was talk of self-service. Is it really possible in the company to have a business that produces reports on its own without having to resort to IT and therefore align with its times?
Software come Business Object or Microstrategy are enterprise products who reason with a very complex semantic data model, transversal and complete of the company. Therefore they are much more centralized in the use and for this they are slow in terms of evolution and maintenancebut they cover a very large sphere. These came used for self servicebut in a less dynamic way than today.
Currently, ci are much more dynamic data visualization products, come Power BI for the Microsoft world, Tableau o Amazon Quicksight which was born on AWS. There are many products of this type that actually no longer work intensively on the enterprise model but operate on small contexts. This has helped the self service a lot in terms of analytical data consumption.
Then there are technical people using Python code or R code to write notebooks and work on the data to obtain poor quality graphs and tables independently. These tools have helped this kind of mechanism a lot and have fostered a change in corporate mentality.
However, some problems arise, such as for example find the right data or even take a data that is not yet available. The idea of data fabric is by simplify this process through intelligence. This can be done via the next best action o next best recommendation. For example, you might have an AI system that, when a self-service user creates a new dashboard, provides suggestions based on what that user is using proposing the information that might be most useful the type of analysis within the specific data context or that it indicates which is all the information related to the job you are doing. Or, even, maybe it shows what a colleague from the same business area who is working on similar tables has used.
Self-service is simplified and offers even more value. Thus the time to market is reduced and the quality and completeness of the activities are enriched.