Shell replaces ServiceNow with...ServiceNow, but there's strictly no customization allowed!

How do you take the step of actually presenting a business case to your CIO, where you say, 'I'd like to replace one version of ServiceNow with the same version?'.

That was the dilemma facing Rob Davenport, ITSM Chief Product Owner at multi-national conglomerate Shell, as the company looked to replace a heavily customized IT Guardian version to an out-of-the-box ITSM (IT Service Management) and ITBM (IT Business Management), without disruption to the business and against the backdrop of a global pandemic. It was, he says, akin to mid-air re-fuelling while moving from a freight plan onto a smaller, more manoeuvrable, faster plane at the same time.

Shell started working with ServiceNow around four years ago, he explains:

When we first started with ServiceNow, we had 32 tools spread across IT, we had lots of data issues, we had a dispersed IT environment that was really quite hard for us to change, modernize and improve upon. We first replaced those 32 tools with an ITSM deployment. ServiceNow ITSM was installed in around 2018 and it proved very successful. In fact, it proved so successful that we then deploy the IT Business Management module, Application Management, Resource Management, Demand...in 2019. That also proved very positive and well-received.

But at the same time these rollouts highlighted that there was a lot of ‘glue’ holding things together - 90 interfaces in total, either connecting to suppliers or other IT systems - as well as a lot customization that had been put in place based on the requirements of the business at the time. This created issues, Davenport recalls: 

We weren't able to really enable automation and new capabilities very easily. And of course, like most companies, [Artificial] Intelligence, Machine Learning, automation are all things that our CIOs are demanding more of us today. Fundamentally, we wanted to improve our reliable and secure operations. We wanted to have better control over our data and fewer data issues and have something future-forward in terms of a data model. And we wanted to, most importantly, improve the overall end user experience.

ServiceNow was brought in to assist with an assessment of where things were now and where they needed to move towards. This is when presenting that business case alluded to above came into play. In fact, says Davenport, there were some clear main drivers to make that case:

First up, we could reduce the effort, cost and time of doing our maintenance and upgrades of the platform. So can we move one annual upgrade, that would take three months to do, into two upgrades a year in less than six weeks? Can we get out tickets routed quickly? And simply, can I get tickets delivered from one support team to another in a much faster time with fewer errors?  Then can I address my data issues? So can I have a data model that thinks about products rather than applications or services? And finally, and most importantly, can I put the foundations in place so that I can then expand and enable new capability quickly? This, for us, is really the the big benefit of where we need to take the journey of Shell in terms of a modern IT-ready organization.

Deployment

The business case made, there was still the need to ensure that there was no disruption to operations during the replacement system’s introduction. In the event, Shell ran two live environments concurrently, with synchronizing of data through a bi-directional data replicator, explains Siraj Kabir, Shell Program Manager:

Once we were set up with the synchronization of data and mapping, we then had the foundation to add on the new processes, IT communities and suppliers in a series of waves, to move them from one environment to the other. We started with just about 200 IT staff; now it has turned into some 10,000 IT staff and suppliers being on boarded into the new environment. As we went through each wave, we also released more processes and new features. So far, this approach has been proven very successful, with only Shell customers left to move into the new environment. In the next couple of weeks, after this is done, we can then retire the old legacy system and hopefully fly with our new sleek, out-of-the-box system.

All of the progress to date took place against the shadow of COVID, a factor that had major implications for the approach taken to change management. Kabir says:

Generally, with a program of this size and complexity, we'd have gone with a ‘Big Bang’ change management approach, bringing every user into training, into forums, into discussions. That possibility was taken away with the COVID lockdown. So we thought about applying a very agile way of change management. We decided that we're going to deploy elements of change management process and then adopt as we go along.

One of the key differences that we had really picked up in this change management approach was how do we train our end users? To do that, we reflected back and we felt that as ServiceNow has been running in Shell for about three years, a lot of the end users are already familiar with ServiceNow aspects. So our change management training really started focusing on developing content about what is different. We also decided only to use existing channels and forums that were already in place. So far, this has really served us very well, starting with a number of early adopters and ramping up for all of our users as they came along.

ROI

There have been a number of tangible business outcomes already, including reducing upgrade times from three months to four weeks, says Kabir:

We are able to probably do that twice a year now. Our processes are 10% to 20% faster. We're able to turn on new features in days compared to previously in weeks. One of the remarkable things about this whole journey is there has been no operational impact or unplanned downtime throughout the program. And user experience has been overwhelmingly positive, despite encountering a few odd defects and data issues.

That said, there were some other challenges along the way, he admits:

The first one that we really confronted was how do we maintain a team spirit. Under that COVID lockdown situation, ours was a truly globally-dispersed team. To be able to get this team together and maintain a high team spirit, we immediately thought about putting maximum focus on care and making sure that we take extra steps to build teams. From singing, to reciting poetry, to making crazy roaring tiger sounds to clapping, we have done everything to be able to build that team spirit.

Secondly, as COVID started, the industry also went through a downturn. So we had to encounter cash preservation as a main consideration in our journey. To be able to do that, we had to really prioritize the scope. We really had to validly challenge that only essentials will be taken on board and all the 'nice-to-haves' would need to be deferred. But it also meant that we had to continuously challenge ourselves, but without compromising the quality.

Another issue related to the desire to be out-of-the-box and enforce a strict no-customization regime, something that parts of the business were resistant to:

We really had to learn to say no, not being harsh, but by proactively engaging with the businesses, telling them, preaching to them, the value of being out-of-the-box, and show it and thereafter convince them to be on board with this journey. One of the critical factors in this whole journey was that we got full leadership buy-in, all the way from the CIOs to our business leaders, also establishing very strong governance along the way to ensure that we can say no to customization requests.

As for learnings to date, the importance of senior level buy-in comes close to the top, he concludes, if that no-customization mindset is to take hold:

It has to include your CIOs, your business leaders, all feeling committed to zero customization. The next learning we have is making sure that we do risky things that are smaller scale before we go for a wider adoption. First, show it - prove it’s valid and then adopt. Other aspects that we've picked up along the way in terms of learning is that big blockers are rarely technical, but often organizational, [such as having the] right resources, contractual agreements with your suppliers or team behaviors.

---

For more diginomica stories from Knowledge 2021 visit our Knowledge 2021 event hub. Knowledge 2021 opened on May 11th and sessions are available to view on-demand until October 2021. This is the event registration link.

---