Now that we know Digital Experience Platforms (DXPs) were not immune to challenges of usability, scalability and uptime last March, the question is how to prepare for the next big spike.
And the spikes are already here, one year later. Just ask any state around the U.S. that manage COVID-19 vaccine appointments.
Being ready for a digital onslaught requires a combination of the right flavor of cloud computing technology, the right people and a solid digital emergency plan, according to Irina Guseva, principal analyst at Gartner who first reported the usability, scalability and uptime problems in her Magic Quadrant for Digital Experience Platforms report released in January. Customers of a majority of Digital Experience Platform providers reported those problems to Guseva in the March 2020 time frame. The “failure” many experienced in March forced these practitioners to get their house in order, she added.
Having the Right Flavor of Clouds
So how do you start? Organizations must find the right flavor of the clouds on the continuum of Infrastructure-as-a-Service (IaaS) to Platform-as-a-Service (PaaS) to Software-as-a-Service (SaaS), according to Guseva.
What Is Infrastructure-as-a-Service (IaaS)?: With Infrastructure-as-a-Service (IaaS), a brand essentially buys or rents server space from a vendor. They can then in theory manage their applications — from operating system to middleware to runtime — without assistance from the IaaS vendor. IaaS is often used to describe “cloud services” or “managed infrastructure services."
What Is Platform-as-a-Service (PaaS)?: With Platform-as-a-Service (PaaS), vendors give customers the same server space and flexibility, but with some additional tools to help build or customize applications more rapidly. PaaS vendors handles things like runtime, middleware, operating system, virtualization and storage, although the client or customer manages their own applications and data.
What Is Software-as-a-Service (SaaS)?: Software-as-a-Service (SaaS) handles the technical stuff while at the same time providing an application (or a suite of applications) that the client or customer can use to launch projects immediately — or at least, faster than they would do with an IaaS or PaaS solution. SaaS vendors often use IaaS or PaaS Solutions to support their suite of applications, handling the technical elements for their customers.
According to Guseva, the ideal spot is to be between PaaS and SaaS, with SaaS becoming increasingly more pervasive. “Yes, you can say, ‘I need my private cloud because I have PII and whatnot,” Guseva. “If you do have those requirements, then you really need to have the proper talent on your team to support your ship when the sea gets stormy, and that's usually not the case.”
Guy Hellier, vice president of product management for customer experience management at OpenText, a DXP provider, told CMSWire that his organization saw many on-premise customers last March accelerate plans to move to the cloud with managed services to drive down internal costs with improved security and reliability.
Related Article: IaaS vs PaaS vs SaaS Cloud Computing Architectures Compared
Why People Matter
Guseva said many practitioners report a desire for “real clouds” and not PaaS because they want to avoid building their own open-source container environments through Kubernetes. According to Guseva she's been hearing practitioners tell her the same things. “I don't have the people to do that. I don't have the money to hire people to do that. I just want a cloud, where all the infrastructure is taken care of. And I can just focus on the creation of the experiences and optimization of those experiences and being cool and innovative.”
Organizations shouldn’t expect a back end developer who is not trained in AWS to suddenly have to configure a way save your ship from sinking. “Whether it's internally, or whether it's through your systems integrator partner, you'll have to have the right people,” Guseva said.
The most successful companies were already well on their way to executing a comprehensive digital transformation for their organization, according to Michael Gerard, chief marketing officer at DXP provider e-Spirit. The ones that had in place a culture of agility, enabling rapid response to the new normal, also succeeded. “As cited by many studies, COVID-19 sped up digital transformation by five or more years for most organizations,” Gerard said.
DXP Users Digitized Complex Services
John-Paul Syriatowicz, CEO of DXP provider Squiz, said clients during the initial digital onslaught of a year ago needed to get more services and information online, create new online channels to engage, become more flexible and be able to update sites rapidly.
They approached the pressing matter with straightforward informational and transactional sites with no issues, Syriatowicz said, but the extent to which they could take these initial responses further depended on how much work they’d done until that point to digitize services, integrate systems and build teams that could move rapidly around areas such as content development.
“Once the initial needs were addressed,” Syriatowicz said, “our clients began looking at how they can be more resilient and responsive for whatever comes next — digitizing more complex services; integrating systems to better personalize content, offer new tools and automate processes; and bringing multiple sources of information or interaction into one platform for streamlined customer experiences.”
Mature DevOps, Testing Led the Way
Organizations that do well in this arena even with increased platform growth had a few characteristics in common, according to Kelly Rusk, Sitecore architect for Rackspace and Sitecore MVP:
- Had more mature DevOps and testing methodologies
- Acknowledged that while a DXP platform provides base functionality, there are several items they need to have accountability for — aka, "not buying the hype that often comes with a DXP: that the platform is all you need," Rusk said.
- Decided on which DXP platform to leverage by first evaluating business requirements vs. buying a DXP and letting the technology drive what the business adopts. "A good example," Rus said, "are some of the marketing features of DXPs. They offer powerful insights, AI, and other aspects. But if an organization only requires base information, they can quickly increase non-governable complexity by becoming in love with the data, while lacking the actual information their organization requires to make decisions."
- Had a documented governance plan that detailed not only their DevOps and development practices, but how to execute and who is involved during a crisis such as a disaster recovery event.
- Reevaluated which features they actually needed to be successful by taking a hard look at the user journey. 'Many organizations are too focused on using every feature of their DXP vs. identifying what their users demand and building to that," Rusk said.
"Many organizations overdeveloped their solutions trying to keep up with the full features of their DXP," Rusk added. "This was not the time to overreact and overextend but rather ensure business requirements drove solutions. Several organizations went too fast attempting to meet perceived user needs, and missed key development and DevOps milestones resulting in a less stable platform."
Related Article: Is Your Organization Ready for a Digital Experience Platform?
CMO, CIO Alignment Matters
General characteristics of those who were able to handle the digital onslaught included good alignment and collaboration between the CMO and CIO, according to Vahe Kassardjian, director at Deloitte Digital.
Organizations in a strong position also have:
- A roadmap that provides a reasonable (not overly accurate) three to five year view and allows the execution of immediate projects. It also prevents projects that would handicap the long-term plan, according to Kassardjian.
- Loosely coupled teams, processes and technologies and clear deliverables and roles and properly architected technologies.
- A mature agile mindset with a focus on a few core principles.
- A "think-ahead culture," as opposed to organizations who favor — sometimes gratify and nurture — heroic acts for fixing problems that shouldn't have been there in the first place.
"Monolithic is not good and must be abandoned," Kassardjian added. "On the other hand, too many organizations suffer from excessive complexity, over-customized systems, capability gaps and overlaps, traumatic migrations when replacing/upgrading parts of their home-assembled stack. Thus, one shouldn't try and build/maintain something that is available ready-made to competitors provided first principles are met: Decoupled systems, open standards, architectural consistency, a plan for adoption and a plan for exit, etc. This automatically favors platform vendors who provide these characteristics but also leaves the door open to adding point solutions."