Replatforming is all the rage nowadays as the right choice and application of marketing technology progressively correlates with better commercial performance.
Whereas companies largely bought all-in-one solutions or clouds to cement operations, purchase patterns are shifting now in order to become more versatile.
Why is that?
WHAT IS REPLATFORMING?
As marketing technology adviser I have the opportunity to peek into many kitchens when marketing or e-commerce prepares for a change of gear.
The shelf life of sites or Apps and underlying applications seems to get shorter and shorter. That’s where the idea replatforming comes in. This is the process by which an e-commerce site will move from one platform to another. Or for that matter, the adoption of marketing automation systems to streamline and automate the demand generation and lead nurturing.
It has been well documented that the greatest challenge in selecting a new platform is simply understanding the differences in functionality between them. So, a big challenge for businesses that want to keep up with the latest new trend or technique is: What to choose? How to organize?
ANYTIME, ANYWHERE REQUIRES MORE THAN JUST TECH
It seems everyone is in agreement in regard to today’s challenges:
- When I talk to marketing people, it is about customer experience and connecting to individual customers, anytime, anywhere
- When I talk to e-commerce managers, they want to sell and deliver anytime, anywhere
- When I talk to intelligence specialists, it is about modeling 360° customer behavior, anytime, anywhere
- When I talk to the CIO, it is about data integrity, security and compliance, anytime, anywhere.
- While companies prepare for the shift from the mass to the individual, the biggest misperception is that by just putting the technology in place, you get the benefits.
People, process, data and technology have to travel together. A lot of executives expect instant miracles but forego to depart from a shared vision and to manage its execution, by way of a strategic replatforming strategy.
A great, replatforming example for the ‘anytime, anywhere’ reality has been developed by Dutch bank ING Bank. By implementing a centralized advanced analytics and campaign management program that creates personalized offers in real time and deliver them through multiple channels, ING has gradually increased response and conversion rates while lowering costs of customer acquisition.
While it has taken ING Bank more than 5 years to make the fundamental technical and organizational changes needed – undoubtedly at huge cost, the state of today’s technology offering enables virtually any organization to emulate their approach at acceptable cost, provided they align people, process, data and technology.
The biggest challenge in relation to strategic replatforming is to break up the silos and get the communication flowing. Too often, new digital channels and digital specialties have led to new hierarchies, new silos and communication gaps, and more complexity. Great things happen when marketers, intelligence, channel and technology people get together and learn to speak the same language.
ARCHITECTING THE NEW STACK
Marketing technology stacks can be broken up into ‘need spaces’ or functional tasks such as customer relationship management (CRM) or web content management (CMS).
With marketing now directly responsible for revenue growth -on top of customer experience and traditional branding and communication functions, the right choice of tools used across the marketing and commerce spectrum is to grow in importance.
In that context, two fundamental developments change the buildup of the marketing technology stack for good: Microservices and machine learning.
- Event-based architectures and microservices are beginning to see widespread adoption in marketing and e-commerce in pursuit of the ‘anytime, anywhere’ customer. Large applications are broken down into small, autonomous functional marvels that solve a concrete problem. Examples can be found in such areas as tag management and conversion optimization, facilitated by APIs (Application Programming Interfaces) to enable easy integration between otherwise distinct applications.
- Data has become the lifeblood of personalization, replacing traditional intelligence with advanced analytics. Newest analytics platforms such as Hadoop, Spark, and tools like Julia, Python and R. are versatile and open, with higher processing capabilities that can handle real-time, unstructured data.
Looking backward, businesses have bought content management and commerce systems, and marketing automation platforms, on-premise or SaaS. These applications are mostly about business logic, making workflows efficient. Replatforming in these categories will likely have to befit the current stack buildup.
Looking forward, as companies need to better engage with prospects and customers in a very personal way, companies will resort to a new class of data-centric tools that empower the heart of marketing and e-commerce operations by way of algorithms.
Machine learning, which relies on automated pattern recognition will play a pivotal role in these new solutions. The core idea is to mine customer data for affinities and propensities to spend in order to target prospective customers in individually relevant ways.
Data-first tools will initially appear with focused use cases. They integrate as analytical data co-processors with incumbent systems for data access. A tool like Sagent (disclosure: I am an investor in this company) that provide automated, sophisticated analytics for ‘persuasion profiling’, is a good example.
This may prove to be just the first step in a transformation towards open, ecosystems that will radically change how the marketing technology stack is built, and maintained. It might as well render current end-to-end marketing clouds and analytics solutions from vendors as Oracle, Adobe, and SAS obsolete.
THE CRUX OF DATA-CENTRICITY
It’s getting easier and easier to bring data together, and the tools to analyze that data are getting more sophisticated, and also easier to use. The objective is clear: It is to become data-centric such that marketing and e-commerce are driven by real-time evidence-based insights, for the anytime, anywhere era.
Current thinking is to put all of your data into one system where you can do something with it. You need to have it pull from a group of different systems where it’s generated and put it into ‘formatted systems’ like expensive data warehouses or CDP’s (Customer Data Platforms).
The way forward is to stop doing this and use a ‘data lake’ in conjunction with advanced analytics tools. Microsoft’s Cortana evidences an interesting option in that regard. Then, when you act on it, you have to be able to easily push it back out and have access to it in the channels you use to service your customer.
This is not about replacing what you have. It is about strategically leapfrogging to accommodate new data and new analytical workloads in order to produce new insights on your customers across a set of characteristics, including:
- Demographics, preferences and life moments
- The behavioral pathways and attitudes during shopping
- Channel preferences
- Product affinity
- Response to offers
To succeed in an environment of unprecedented change marketers and e-commerce managers need to put data and algorithms at the center, and use them to drive campaigns and other marketing and commerce activities.
With it comes the notion that we have to accept and trust that algorithms run the business. This will be for most marketers and e-commerce managers a true paradigm shift.