Today, I saw a Linkedin advert from IBM showcasing their Analytics suite. Given the size of IBM and the products and services at its disposal, they have in fact given Analytics a significant attention by listing it on the drop down menu, i.e. if you landed on the IBM website, select Products menu option to see Analytics listed therein. From there, you need few clicks to get to what caught my eye, i.e. integration and governance.
As expected, IBM services the enterprise and the multinationals. And as a result, it has to deal with legacy, on-premise and cloud enterprise software and a large amount of data held in data silos. To overcome these issues, IBM and many others are resorting to Machine Learning (ML) and Artificial Intelligence (AI).
In a nutshell, they are trying to teach systems (software applications and data sources) to find patterns and relevance in data scattered across their enterprise software world. Many customers are buying into these early stage solutions without fully understanding how they can solve deep issues within the enterprise. Just like everyone had to have a mobile app strategy a few years ago without truly understanding why customers would bother installing hundreds of mobile apps on their smartphones starved of memory and space, simply to stay within the trend curve (my competitors are doing it so I must as well). Now the same is happening in the AI and ML space. That’s all fine if you have corporate budgets to burn.
The world of Virtual Integration
UnifiedVU is a lonely voice in the world of virtual integration. Whilst our solutions are drastically different from those of integration service providers such as IBM, Dell Boomi, SnapLogic, Informatica, MuleSoft, Cazoomi (I’m a Board Advisor at Cazoomi), Zapier, PieSync, Cloud Elements and others, we nevertheless bring a smarter technology to market.
Let’s take a shallow dive into IBM stack shown on above image:
1. Data integration
This is required if you must have data from one or more software applications and/or data sources in another software application. Data integration can be achieved in four ways:
- Direct integration between sources of data (software applications or data sources), i.e. one to one connectivity.
- Through data integration engines such as Cazoomi and Zapier
- Through middleware such as MuleSoft.
- Through data warehouses.
However, you need to question the motivation behind data integration requirement. Is this for:
- Further calculations, e.g. data coming from two sources are used to create new data?
- Simply to present data next to each other, e.g. financial data fromXero or Sage in Salesforce or SugarCRM, so that productivity can be improved?
If it’s the latter, then you do not necessarily need to spend your budget on expensive technologies such as IBM. Virtual integration service provided by UnifiedVU could be a low-cost alternative, that could provide significantly higher value than data integration. Given that UnifiedVU could be adopted for any role by choosing the connectors that matter most to you, we get one step closer to providing a solution that meets individual employee’s exact needs.
2. Data governance
Data governance issues could arise as a result of data integration (may not be the sole reason). The key focus here is to ensure that only those who should have access to a particular data are the only ones who actually have access to.
Let’s say, an Accountant using QuickBooks needs access to limited amount of sales data which resides in Salesforce. Should the Accountant be given a Salesforce account and then add restrictions there to manage his/her access? Is that possible at all? Perhaps! Unfortunately, this level of granularity is not available in many software applications. This also raises another burden as the Accountant would then need regular training to access the data, as one can expect, s/he is a soft user of the CRM.
But what if there is a better way? A better way that gives the Accountant limited access without compromising on sensitive data. Well, this is exactly what UnifiedVU offers. No more data breaches! No more additional training. No more wasted time. No more governance nightmares! We keep silos as they are and momentarily unify them without data exchange between systems. Data governance built in from the first day!
3. Data quality
Data quality issue arises when two or more data means the same thing. This could happen inside the same software application or the same data source. This could be down to misspellings, shortening of names or missing out names from a long name. Let’s consider Unified Health and Social Care Ltd. You may have used the whole name in your finance application such as Xero, SageOne or QuickBooks. But you may have shortened this to Unified Health in your customer support application such as Zendesk. When you integrate and transfer data from one software application to another, this could end up messing up the data of the receiving software application. A classic data quality issue.
Let’s explore company and trading names. Take Liquid Bronze Ltd trading as Malinko. You have a cat in hell’s chance of matching these. Well, you will be glad to know that we are currently experimenting with technology at UnifiedVU to address this issue. The key difference is that we achieve all this without any data exchange between software applications. If a data quality issue is noticed, we allow you an easy way to change the data in the connected software application by directly visiting the applicable record. It is our strong belief that nothing beats more than correcting wrong data at the original source.
4. Master Data Management (MDM)
The MDM was born when you no longer trusted data in which silo was the most correct. You created a new record combining data from all the applications. And once cleansed, this was then referred to as Master Data. It was then injected into silos where the original data came from in order to create the master data in the first place.
With UnifiedVU, MDM is irrelevant. We work with relevant small data rather than big data that everyone else seems to be obsessed with. Big data in one sense allows you in strategic decision making where you act on behaviours and patterns. With small data, you improve your workflow efficiencies.
This post was written in tongue and cheek to show that there is another way, which could avoid significant cost and bring organisation wide efficiencies and productivity gains without resorting to expensive solutions. This post was not written to undermine the great work done by IBM and others. The four areas highlighted are clearly areas that every CIO, CTO or IT Director worry about. With General Data Protection Regulations (GDPR) kicking in, this issue is now even a bigger item in CIO’s agenda. It’s none existent with UnifiedVU as we do not store any data other than user login credentials encrypted. As you can see, UnifiedVU was built from the ground up to remove these significant issues.
As I said from the beginning, there is another way. And it’s called UnifiedVU.
Do reach us if you like to discuss how UnifiedVU can deliver a solution that fits with your organisation.
Image credits: Images are taken from IBM and all copyrights remain with them.