Johannes Nygaard is Chief Digital Advisor at the University of Copenhagen and was a part of the CMS selection team that evaluated options and planned the process from beginning in 2017 to confirming the final selection in 2023.
By Janus Boye
It’s quite unusual for a CMS selection process to take 5 years, but as the digital platforms keeps growing in importance, year long evaluations are likely to get more common in the future.
This is in particular the case for large, complex and international organisations, where the task of migrating thousands of pages and custom-built development is a long, risky and expensive one.
I recently spoke to Johannes Nygaard who is Chief Digital Advisor at the University of Copenhagen and was a part of their CMS selection project. This project started back in 2017 and while the pandemic did delay things, the selection project was recently concluded with the selection of open source CMS TYPO3 as their new digital platform.
Let’s take a step-by-step look at what happened and see what you might be able to use in your own selection processes.
Finding a new CMS starts with ‘‘why”?
For over a decade, the University had worked with a small, local and proprietary CMS. While this clearly worked and was relatively low cost, the digital team also observed how other customers increasingly left for other platforms, how the CMS market kept innovating and how they slowly, but truly, became the only very large customer using that platform. In short: It was risky to not consider their options.
As Johannes said:
To avoid ending in a situation, where we might be forced to do a sudden change of systems, we wanted to both understand our needs and analyse the marketplace
From the onset, it was important for them to have the freedom to adapt the system to their needs, similarly to their existing setup. Also, it was at least equally important to become a part of a bigger community with many different customers on the same platform. This is in particular what they were missing.
My take:
It’s less common these days for large organisations to use a small, local and proprietary CMS, but the situation is actually not so much different from being five versions behind on a more well-known CMS. It’s a smart and risk-reducing move that the University gets this process kicked off on their own initiative. This clearly set the foundation for a process that did not have to be rushed and could include all relevant stakeholders.
The selection gets started: Needs analysis and informal benchmarks
In 2018, they received the green light to move forward and Johannes and his team took on two key deliverables:
Creating a list of about 25 similar sites, mostly from higher education, but also large local government sites, to find our which CMS they are using.
Writing a needs analysis based on must have requirements, and a longer list of things that could be delivered later on
Once these deliverables were ready, they tried to merge them and figure out which systems used among the 25 similar sites met their needs. This narrowed the list to four quite different content management systems, and as Johannes said, for internal reasons, the list of four, quickly became a list of three candidates.
My take:
I’m a big fan of this approach. I often find one of the two deliverables that the University did, but not the other. Either an organisation who has used months to create requirements, but not looked at the market and similar sites. Or vice versa, only looked at the marketplace, without putting much thought into the requirements. Combining these two is a smart way to keep a reality check and to narrow the list of hundreds, if not thousands, of potential vendors.
Narrowing to three potential systems and doing site visits
At this point, still in 2018, the list of potential systems was reduced to
Drupal
Sitecore
TYPO3
The team at the University of Copenhagen now booked a few site visits with customers of each, so that they could learn more about adoption and speak to users.
Reflecting on the 5-year process, Johannes highlighted this part as time really well spent. It helped shape the decision and by speaking to peers in confidence, without vendors and consultants, they learned a ton.
This phase was concluded late in the year with a set of recommendations. Before they were able to take action, the pandemic came and temporarily halted the project until 2021.
My take:
Speaking to peers is always a good use of time and a shortcut to unparalleled insights. Besides learning new things from behind the scenes, the benefit with this approach is also, that the knowledge doesn’t just reside with one person, but with the selection team. This is a very good investment that will pay dividends down the road.
Doing a pilot to better plan and understand the financial impact
During 2021, they then looked at the report with recommendations once again and also considered the financial impact.
Open source CMS TYPO3 was considered the best choice, both matching needs and with good references, but instead of just jumping ahead, they decided for an 18 month pilot project before migrating the whole lot. In this phase, they paid a local digital agency to support their efforts.
When speaking to Johannes, he also referred to the pilot as a kind of final proof-of-concept, with three main goals:
Build internal experiences, also technically, so that they could learn how much they could build themselves
To get to know the community better, including understanding the rate of development and the support ecosystem
To better understand the agencies offering TYPO3 services.
Nothing in the pilot was put into production - it was all just to learn, but the pilot also helped improve the estimations and thus making future budgeting more predictable. Budgeting both in terms of money and also better estimating the required internal resource allocation.
My take:
There are some important details in this phase. Sometimes pilots or PoC’s end up in production and then the learning slides to the background as the pressure of deadlines and production environments moves to the forefront. Also, in Denmark the list of experienced TYPO3 partners is small, so it makes more sense to take an implementation step-by-step and try to avoid creating an unhealthy lock-in to the digital agency.
Creating the final recommendation to make the decision
As the final step, the team wrote the recommendation to the management to make the change to TYPO3.
This was approved in March and as a part of it, they listed these three main reasons for selecting TYPO3:
Wide adoption among similar organisations, including many in higher education (notably Aarhus University in Denmark and many in Germany)
Consistent roadmap with improvements -
A well defined architecture, modular built
something more
My take:
Coming soon
Learn more about CMS selection
We have quite a few posts about this topic. This one from 2020 is written in a similar style: What can we learn from the W3C CMS selection process.
Also from Denmark, Aarhus University selected TYPO3 back in 2009.
In a helpful book from 2020, Deane Barker shared his 25 lessons learned about buying content technology and services. The book covers helpful advice specific to the market and it's filled with important lessons for selecting content technology. Also, unlike a few other business books, it tries not to make the wrong assumptions. We did a member call on the book launch and added a few more lessons: 25 + 3 Lessons Learned About Buying Content Technology and Services.
Finally, the conversation on CMS selection naturally continues in our peer groups and at our upcoming conferences.
Full disclosure: Both University of Copenhagen and TYPO3 are members of our community