(Image by Martin Slavoljubovski from Pixabay )

Colgate-Palmolive has migrated hundreds of brand sites around the globe to a simplified and standardised platform using Adobe Experience Manager (AEM).

The digital marketing transformation project, known as Dragonfly, has involved the migration of 267 brand websites from a legacy content management system to AEM, across 14 brands.

Jon Schultheiss, Associate Director of Global Digital Marketing at Colgate-Palmolive, explains to attendees at Adobe Summit how his organisation faced "a great challenge". In some cases, the website migrations involved 50 language localisations. All migrations had to be completed to a hard 18 month-deadline, without any interruption to existing service. Schultheiss says:

That was our challenge, but we saw it as an opportunity. Thankfully, we not only had the mandate, but the unwavering support from the top leadership within our company. They ensured we had the necessary resources made available to us, so we could succeed, as well as the patience of the organization to retain the status quo for periods of time. This allowed us to make the transition as smoothly as possible.

The critical success factors to the project centred on gaining speed through simplification and standardization. Simplification focused on making things easier for content authors and creating a common set of templates. Standardization, meanwhile, involved all sites using a common user experience and interface, and a common methodology for content creation. He adds:

Those were our checks and balances - the driving dimensions that we would refer to when there was a discussion point or decision to go one way or the other. They became part of our vernacular; they became our guiding principles.

The right platform

Schultheiss says Dragonfly, and its attempt to support the migration of 267 brand websites, was a significant project. The starting point was the creation of a dedicated discovery team, who went through the sites, their variations and their dependencies. Schultheiss' team focused on how migration would take place, bringing together business managers, agency partners, and the internal development team. He explains:

We aligned on governance, ways of working, and the development and technical approach that we would take to meet our critical success factors. We decided to not build individual sites, or a set of sites, but instead create a factory of sites. This factory approach drove our capacity and capability during the migration, and has stood us in good stead post-migration, with now over 400 sites on this platform.

Schultheiss says Colgate-Palmolive selected AEM because it provided the best site templates and the right platform to help his team build its factory. Their factory-first approach focused on simplified development and standardised elements. The aim was to ensure the content team could complete digital marketing work without having to rely on input from the development team. This approach was known as "methodology over code".

Schultheiss says his team had to carefully manage the expectation of business stakeholders across the 18-month migration schedule. They made it clear that the creation of the factory was their first priority and that individual sites would not be ready for several months. They produced a structured communication plan, with sprints that demonstrated project progress. He says:

We shared our journey with our stakeholders, evangelising the story of how the work done for one of the brand families would affect and help the other families. We even held a second offsite, with our content team, to review how the factory could account for the variances across not only the content within one particular family, but provide best practices and standardisations across families.

It took eight months to complete the factory. Site migrations were then tested within the factory. What Schultheiss refers to as a prime site provided the pattern of content management for all other corporate sites. Unique content elements that could not be migrated were either changed or dropped, simplified or standardised. He adds:

This was a huge step in ensuring that other brands would follow suit. If you're trying to control and regression test when you vary from the prime site, you need to repeat that process across the other brands' look and feel; deviation was not tolerated.

New experiences, faster

Once the factory was up and running, Schultheiss went live with two prime brand sites. They spent another month on runtime testing and performance monitoring. This settling-in period allowed for optimization changes to be made before the team moved forward with the rest of the migration process. They pushed two more family sites live before, finally, pushing the Dragonfly prime site live. Schultheiss explains:

And it worked. We were ready to start the assembly line. For the next six months, we went on an aggressive go-live schedule to transition the remaining 262 sites in six months. We were able to accomplish this because of the dedication of our agency, development, and content teams.

Schultheiss says agile development processes were at the core of the successful site-rollout process. The team was able to benefit from sharing more than 6,000 user stories across its development platform. He adds:

We worked within a living, co-ordinated roadmap, so that each team would exercise just-in-time delivery. When there was the need for new content for prioritised marketing needs, that could be delivered with as little code as possible.

Schultheiss' team also used a regimented go-live schedule, with checklists and documentation. Anywhere between four to eight sites went live each day; on one particular day, the total reached 16. Schultheiss says the key to success was cooperation and coordination between different teams.

He says that building the Dragonfly project on AEM means his team has been able to stand up brand- and campaign-specific content at a higher velocity than before. Since the migration, Schultheiss says the company has been able to produce new experiences faster, in some cases, twice as fast as before, and at a fraction of the cost:

I think about how many different parallel streams we were running and it was a tremendous effort. We finished one month ahead of schedule. We went live with the 267 site in 17 months, but you don't do a digital transformation like this to hit a number - you do something like this for the climactic scenes to come. We are now leveraging our factory, empowering our content teams and exercising our dragonfly way of methodology over code.