Baylor University's Oracle ERP rollout - a masterclass of a pandemic pivot
- Summary:
- Baylor University provides an exceptional example of how business and IT come together to assure implementation success.
We are starting to see how COVID-19 changed the manner in which EPR implementations are undertaken and how a fresh approach is bringing demonstrable value. Baylor university's Oracle project is a case in point. But first, a modest rant.
ERP implementations have been stuck in a time warp where the practices of the last century carried over into the 21st almost unchanged. Projects limped across the implementation finish line, budgets were routinely smashed. As for benefits, they were often illusory or simply not measured. Then along came cloud implementations with all the promise of eliminating CAPEX, shortening implementation timelines and continuous delivery. And while we saw some progress, it has been at the margin not the mainstream. Or at least that was the case until COVID-19 came along and upended traditional implementation methodologies.
Almost overnight, gone were the endless meetings about meetings, status update meetings, functional hash out meetings, all to be replaced by Zoom meetings where getting stuff done became the order of the day. Gone were the waterfall projects where users are an afterthought, replaced by IT and the business working together, albeit remotely. What did Baylor do and how did it work? For that I spoke with Jon Allen, Asst. Vice President and Chief Information Security Officer Baylor.
The story starts in 2016-17 when the university was endeavoring to come up with a solution to its legacy landscape that included some 20 add-ons to what passed for a platform which was supposed to support the entire HR function.
What we really had was a set of barnacles that weren't performing. It was very painful.
Oracle comes back refreshed
In 2017, Baylor looked at both Workday and Oracle. At the time, they thought Workday had the edge but a necessary delay meant the project went on hold until 2018.
It was very interesting because what Oracle brought to the table in the Spring of 2018 looked like an entirely different product and we felt that at that stage, Oracle was on par with Workday on HR but more advanced in financials. We were running demos not Powerpoint so we got to see the real thing as it was in development at the time. That also meant we could see the roadmap ahead for six months and how features we needed were getting into the product. We saw how the customizations we need are available. We saw the live system and were able to ask questions around the things we need to do and so got a really good feel for what each system offers.
The implementation project kicked off in August 2018 in the traditional manner with Huron as the implementation partner. But it quickly became clear that Baylor didn't have what was needed to pull off a project designed for digital transformation.
Fortunately, we had the leadership in place who had bought in and said, we are supporting this, regardless of what it takes. And I think in retrospect, that's one of the key themes through this project. We were committed to absolutely doing what the solution provided that was going to map to our business, making sure we did things like completely rebuild our chart of accounts. I'm an IT guy. I had no idea what the chart of accounts was or what the value was in it. And then they talked about change management. I thought that was how we make sure to get code in production. I had a lot of learning to do!
By the time the pandemic struck, Baylor was in the testing phase. Like everyone else involved in projects of this nature, it had to decide whether to continue or pause. Baylor decided to continue but realized it had to do everything remotely, in this case using Microsoft Teams.
I'd say that working remotely worked out well for us because your communications have to be so much better and you're not for example chasing people down.
Change management - the key to success
For me, the interesting part of the conversation centered around the change management topic. Allen noted that in his view, the outsized number of failed higher education projects is largely down to a failure to take change management seriously.
In every project you're looking to cut corners and change management is one of those because it hasn't been the focus. If you're looking at a place to trim as I've talked to other schools, that seems to be the corner that we notch out. The view has been that users will figure it out but that's not worked. This time we took change seriously and to be fair, it was Huron who pushed us into taking that approach because higher ed organizations are kind of loose federations and if you lose users then it's really hard to get them back. We involved the business users early on, started training while developing the system and making sure that our communications were prioritized. What's interesting though is that the pandemic actually helped us along with that because as I said earlier, we were forced into different ways of working that complement but also challenged us to be better at both understanding the user concerns and then training them to the way the system works. I think looking forward, it's going to change the way we approach projects, and making sure that that is a foundational piece of any of these projects rather than an afterthought.
Turning to the way that Teams enabled the go live, Allen said that he doesn't believe the University will go back to implementations the way they were done before. Rather, Teams enabled the go live and that in turn has changed the way people see value in collaboration approaches that are used meaningfully.
It's not so much about where I am, but how I communicate.
Talking about how this project has changed the way the business and IT work together, Allen said there is a mutual respect that has allowed both IT and the business to better understand the challenges and opportunities that each face. Allen says that working relationship is, to some extent, assisted by the quarterly update cycle during which needs are agreed and prioritized. But it is also evident in the way that functional experts are configuring the solution while at the same time ensuring the IT technical people have the time and space to manage identity and security topics. Allen playfully refers to this new breed of functional person as a 'funky tech,' a term of endearment.
As we closed out our conversation I asked Allen about takeaways from which others can usefully learn. Alongside the change management topic, he advises keeping things as simple as possible.
There's always the temptation to recreate the past because of an exception that occurred years ago. So in our old HR system for instance, it took 26 steps to hire in a new person. That just didn't make sense anymore.
My take
Did Baylor get it right all the way through? No. Risks were taken on payroll that, in hindsight, were a step too far. But issues were quickly fixed. Even so, as I listened to Allen, I was impressed with the infectious enthusiasm and passion with which he told the story.
As we meander through the fallout from the COVID-19 pandemic, it is clear that the so-called best practices of the past are past their sell by date. Baylor provides an excellent example of what can happen when an organization sets its mind to success, is prepared to take implementation advice but is especially receptive to change management as a foundational element of the implementation project. Critically, in my view, is the recognition that switching delivery methods is not just essential but adds value to the overall implementation and delivery processes both now and into the future. It helps that modern software is capable of configuration by functional experts and in this, Oracle's cloud products have made significant advances in the last couple of years, especially on the user interface side, which Allen describes as way better than what they first saw in 2017. I concur. The Redwood UI, introduced in late 2018, was a huge step forward and long overdue. It's good to see customers reacting positively because as regular readers know only too well, a crappy UI usually means a crappy experience. Where once you could kind of get away with that, the current generation of users will no longer tolerate a poor UI.
Going forward, it will be interesting to see how the collaborative methods Baylor is using evolve to deliver further projects at scale.