Data Fabrics in action - NatWest Group and Jaguar Land Rover walk the walk

One of the interesting side-plots of the recent announcement by Appian of its move into the world of Data Fabrics was that making that move would be essentially free to existing users of the company’s data and process management tools. Those tools are, collectively, the makings of a Data Fabric, so that should be enough provocation to other less well-endowed customers to extend their investment, as well as the as-yet-uncommitted to consider their next moves.    

It also means that the company has a range of early adopters available to talk about their experiences in building data fabrics and their reasoning behind making the move. A selection of them were on hand at the Appian Europe conference at which the company launched its data fabric offering, and here is an outline of what a couple of them had to say.

NatWest Bank

NatWest Bank had one of the most fulsome stories to tell. Steve Marjot, Head of Change at the Centre of Excellence of NatWest Group, talked conference delegates through a novel and constructive approach to the important processes of Risk Governance in banking.

This is an area that is obviously critical to every bank, but it is fair to say its importance has also become something of a hindrance to the flow of business. This is especially so when it comes to functions such as customers needing to be assessed as prospects for loans and other potentially risky services. Marjot’s target was to reduce, if not completely remove, the hindrance from the process:

It is really about how to get that important subject, Governance, under control and an integral part of the process, rather than an external road block to be negotiated around.

He found that there were 17 separate processes that impose some form of governance on change, each one a policy where the policy owner sets up a triage, which led to "hundreds and hundreds" of questions. This, in turn, led to a process, which led to some governance mechanisms for approval, all imposed on individual teams at the beginning of a change when least was known about that change. This, he said, introduced unbelievable inefficiencies in order to demonstrate that controlling risk was tested against the change framework.

This led him to the notion of changeless governance based on a single data model, a Data Fabric that allows those 17 processes to be brought together and re-architected into the flow of the work. This way risk can be controlled in the flow of the work, not imposed on top of it. This, not surprisingly, carried with it a powerful human element. Marjot explained: 

We had to start by engaging with every single one of those policy owners and say, 'We're about to re-architect your policy’. That terrifies you, because you love your policy, you have bought into the bureaucracy of your policy, because it works for you, because it manages your risk. But it doesn't work for us. Because what we want to do is create a colleague-led change journey. The three personas who matter in this space are the policy owner, the person implementing the change, and the person accountable for the outcome of the change.

This is where Appian came into the picture, as it had already played a part in NatWest’s commercial business lending processes. Here, the idea was to build a customer journey view of processes, based on customers’ actual experiences and needs. The decision was made to do the same thing for internal colleague journeys around change and risk governance using the same data modelling solution. This resulted in the creation of a central repository using common data, common rules, common workflow, and with the stated intent of relentlessly pursuing straight-through processing.

This year the bank has deployed the front end of that process, a single enterprise triage capability with the target of reducing those hundreds and hundreds of questions down to around 20. It has also deployed three of those 17 processes into Appian, and has three more deployed as minimum viable products, where work is continuing to mature them to the point they too can be deployed as full processes. Marjot said: 

With that front-end triage we've got logic built in, we've got proportionality, and materiality. And we're slowly, policy by policy, maturing that triage to make it as simple and direct as possible.

The important end result here is the time that can be saved by customers in obtaining a loan, which in turn saves the bank time in turning a customer enquiry into money out from the bank and earning interest. When Marjot started looking at this problem he estimated that the time taken by the existing Risk Governance processes amounted to an elapsed time of 73 days imposed on every single piece of change. This means that for every change the development team need to make to a process, there could be an additional 73 days elapsed time between being ready to release the change and it being in the production environment.

As he observed, these days that is simply an unacceptable delay, and with old-style customer loyalty to banks very much a thing of the past, that loan business will be snapped up by other, more agile sources of finance. Like all the mainstream retail banks, and many other industries and services, being able to react more quickly is now a crucial capability. Marjot concluded:

It's not enough to go from 73 days to 30 days. It's not enough to go from 30 days to one day. If we want to win, we have to go from 73 days to 73 minutes by the time we finish this.

Jaguar Land Rover

Kate Goodhart, Head of Intelligent Automation at Jaguar Land Rover, first turned to Appian to solve one of the major problems caused by Brexit, namely the ensuing supply chain challenges. Not surprisingly, the company has demand for its products from some 100 countries around the world. Fulfilling that demand has been hampered by supply-chain issues, making sure it can get the right parts at the right time so that a Just-In-Time production schedule can be maintained has, in Goodhart's words, been "a real struggle for us at the moment".

This is only compounded by the fact that, like every other vehicle manufacturer, it is also devoting much time and effort into going electric, which is a major challenge in its own right. She explained: 

We've made a bold statement that, from 2025, we're going to be full electric, which is really exciting. We're really proud of that. But that means we've got to change our manufacturing facilities, we've got to change our processes, and, again, the supply chain. So making sure that we're ready for that as well is huge for us.

The impact of Brexit for Jaguar Land Rover was, she said, absolutely massive:

The sheer volume and complexity of documents that we were going to have to deal with just to get parts across the border was just amazing: scary, to be honest. So initially, Jaguar Land Rover thought, oh, well, we'll just hire a huge team of people to sit and churn through those documents.

In the end it turned against this and targeted a digital solution. It looked at Appian, and a number of other automation tools, as the basis for a workflow package that the company’s tax team and various other teams used throughout that process to ensure that all relevant documents were in place when needed. She stated that it is a process robust and auditable enough to empower the company to use the legislation to claim some duty payments back.

Most important, perhaps, her development team have used Appian to build a process that has managed to protect the company from the one issue they feared most: component parts being stopped at the border and production being brought to a halt and without employing a vast army of clerical staff.

The importance of the supply train subsequently prompted the team to address a related issue, the management of problems with suppliers. This is a workflow that tracks the suppliers and their problems, which allows the company to maintain control of vendor quality and reliability.

As with the NatWest example, Goodhart has also found that working with Appian has helped create buy-in from the business staff, which she has found to be invaluable. She offered this advice:

Don't focus on the tech side of it, really bring it back to that business value. What is the benefit that you're going to give back through that automation or that workflow tool, and then put a number on it. Don't be scared to put a number on it and then use that to push it out to the business.

What I've found is that this has been quite helpful is trying to spark that competition between departments. So, for example, if when we've had workflows in finance and the tax teams, the purchasing teams are like, 'Oh, what's that?’ That starts to create competition and 'Well, they're using it, I want to use it.’ That's been really helpful for us.