Technology never fails to amaze me. Here I am on my mobile, in Bizagi’s USA office, checking out my bank balance as quickly as if I was in one of HSBC’s UK branches. Most of us never stop to think how it happens – and unless you work for HSBC’s IT department, you’ll probably never find out. But I’d bet pretty strongly that these guys have implemented Data Virtualization capabilities.
Data what? If you’ve only a vague understanding of the term, then you’re not alone. According to a survey by the BI Leadership Forum, 37% of respondents said they simply didn’t know enough about it to consider deploying it in their businesses. So I thought I’d take this opportunity to explain what data virtualization is, and why it’s so relevant to BPM.
Every business has multiple data systems. And thanks to customers like me who want not only to check my bank balance, but review my payment history, pay my credit card, transfer money and download information about an interesting new mortgage offer, businesses are under pressure to connect these systems at an alarming rate.
That’s if they can be consolidated at all
. Yes, we can put a good process in place. But each time my BPMS wants some of that data, it has to go to the relevant database, and connect with it - creating thousands of links and ‘spaghetti strands’
. The data and the process model get stuck together, each adding layers of complexity and making the resulting process bloated and sluggish.
What data virtualization does is hide all that ugly data behind business objects – in our case, the forms, processes and sub-processes that you see in Bizagi. The process map that you create then becomes the service that you access to bring up the information.
Separating the data from the process, has many benefits. For a start, if you want to make a change to the process, you don’t have to wait for IT to build new data. IT can simply point the process to the virtual tables in Bizagi’s virtualization layer and update it.
Take another example: when you need to update a legacy system with a new one, you just have to integrate it once. You replace one link – not thousands of spaghetti strands. This adds up to fast, agile and low-cost development – exactly what’s needed to keep up with those data-hungry customers!
Making the concept real
So are there any downsides? Well, the clue’s in the title: virtualization is a complex software and the concept can feel pretty abstract to your business users. To make sense of it, you need to spend time modeling a decent ‘virtual view’. This is where Bizagi comes into its own: our Modeler
is truly easy to use. You won’t find any In/Out data variables on our process maps!
And in fact you don’t need to worry about data at all at the modeling level – it is all mapped at our Studio module once you map your processes. You do need to think about your data strategy holistically because if you do so on a tactical level you will end up with more spaghetti strands in the future.
Whether your process is led by HR, Marketing, Sales, Purchasing or any other area of your operations, your people can sit down together and create a process map that shows exactly how the data should look and where it should go. All your IT department needs to do is point the map to the various data sources and the virtualization takes care of itself.
So back to my bank statements… what I truly love about data virtualization is that most of the time, your customers won’t even know you’ve implemented it. They don’t care how many legacy systems you’ve had to connect with, in how many countries. All they know is that they got a great service when they wanted it. And that’s exactly as it should be.
To find out how Data Virtualization and agile BPM architecture can benefit your business, please check out our B for Business
whitepaper emphasizing why data matters in today’s customer-centric world.