Every business has silos of data that don’t glue together, business processes strung together with documents and human links. In this article, we examine how a no-code application fabric can transform how your business extracts value from data.
Plastering over the cracks
Anyone who’s led a digital transformation project doesn’t need telling that data quality is the most prevalent killer of IT projects. It comes down to the fact that enterprise IT isn’t purchased as a single platform, but in bits (in the form of software applications bought to service specific departmental functional needs. Try gluing all of these discrete data structures together and enterprise architects generally find that data exists across their business—and typically beyond—without logical ‘joins.’
What businesses need is a reliable way to plaster over these cracks, to create a uniform and joined-up single version of the truth.
What is a table join?
It would be fabulous if every application in your business identified a customer with the same unique code or identifier. Equally, it would be truly wonderful if every system identified suppliers, products, and organizational departments in a uniform way. But they don’t. And that’s a problem.
At NDMC, we’ve encountered businesses using accounting software, Service Management, and Customer Relationship Management (CRM) software that lacks a common identifier of the key data tables that bind business data together. In one particular case, a motor retailer had set up their accounting software to record transactions of every retail customer using the same nominal ledger code. This meant they technically had ONE customer in their ledger although they were doing in excess of $200 million in retail business!
Correcting shortcomings in data structures
It’s not easy to overcome absent data linkages. You can imagine the scenario: you are using an accounting system that records customers in one way, and you want to tie the customer accounting record with details of their service behaviors or communications habits. This means, somehow you need to recognize ‘the customer’ in these systems in order to match it to the accounting record that details the status of the customer, orders taken and sales processed, etc.
Looking for a Customer Data Platform to bring your customer data together? Discover how NDMC Consulting can help.
Fortunately, modern Cloud Application Fabric solutions like Encanvas are there to help you.
As the name suggests, an Enterprise Application Fabric creates a common, unifying technical architecture for the applications you create, deploy and use across the enterprise. Harnessing the latest private-cloud technology, they are most likely today to use No-Code software development principles to speed up the time-to-value of new applications. Additionally, they will provide up to 60% of everything you need to build a new application ‘out-of-the-box’ including the common core datasets that every new app will require (such as details of Users, Roles, Suppliers, Customers, Processes, Policies, Org Structures, Locations, etc.).
Establishing this common core set of tables for your business means that every successive applications development can be delivered faster, and without the pain of (each time) having to ‘plaster over the cracks.’ This all sounds encouraging, but it doesn’t answer the fundamental question of HOW to create those pesky ‘joins’ between data that don’t exist today.
To do that Application Fabrics need to enter the world of Fuzzy Logic.
Going ‘fuzzy’ to overcome table data linkage issues
Application Fabrics like Encanvas contain fuzzy logic technology to bind silos of data together in order to create the common core tables you will need for your application fabric.
How this fuzzy logic works in essence is to use weighted logic (i.e. if/then queries) to test relationships between two or more tables, and successively build up a list of probabilities to suggest either a relationship exists or it doesn’t. Perhaps the simplest example I could use to get your head around this comes from customer address data.
Let’s say you want to find the common customer tables between two systems that have no common unique identifier. Perhaps, systems do include a data table that describes the address and contact details of the customer.
It might be possible to execute a fuzzy logic query that goes something like this:
‘See if the name of the company is written in the same way’
‘Then, check to see if the first line of the address reads in the same way’
‘Then, the country’
‘Then, the postcode’
‘Then, the name of the individual recorded as the primary contact’
‘Then, the phone number’
‘Then, the email’
‘If all of the above are consistent, then the probability is that we can assume the match is complete.’
An Application Fabric blends application Platform-as-a-Service (aPaaS), with integration Platform-as-a-Service (iPaaS)
Every business ideally wants ONE platform to manage its IT. The emergence of Application Fabrics is the inevitable result of this move towards a unifying enterprise architecture in the cloud. As cloud computing evolves, this demand for greater unification of the technology building blocks becomes inevitable.
Want to learn more about Application Fabrics? Check out our article here.
Ian Tomlin is a marketer, entrepreneur, business leader and management consultant. His passion is to help make great ideas happen. Relentlessly optimistic about the potential of technology for good, Ian’s 30+ year career has focused around the intersect of strategy, technology and marketing. He writes on subjects including enterprise computing and organizational design. He also works as a consultant and advisor to the executive teams of PrinSIX Technologies, Answer Pay and INTNT.AI, helping to rethink their marketing in order to tell their brand story.
Ian has founded a series of successful businesses including NDMC Ltd (2003), Encanvas (2006), and Newton Day Ltd (2019). He has written books, articles and guides on brand, digital transformation, enterprise applications, data science, workforce management, and organizational design. He can be reached via LinkedIn or Twitter.