How Digital can help achieve personalization at scale?

Personalization at scale

Studies indicate that personalization at scale has the potential to create $3 trillion in new value.This value can be realized by building relevant technologies, addressing organizational disconnects, forging trust with customers and protecting the data.
Marketing has changed and true value of marketing lies in delivering experiences that are world class to the customers and at the same time deliver value to the business. To achieve this, personalization at scale is crucial because personalization can deliver tailored recommendations, content and experiences across all channels/devices along the customer journey.
Implementing and integrating the right technology is complex and requires coordination.Successful companies achieve this by tackling technology and business challenges at tandem. This is achieved by aligning the CMO and CIO organizations closely.

How to drive personalization at scale?

Personalization at scale requires orchestration of Data, Decisioning, Design and Distribution. Along with this, right technology needs to be there to unlock the potential of personalization.

Personalization at scale requirement 1 – Data

• As most of the data in the organization are in silos, integrating customer data platform, data management platform and identity resolution platform is required to integrate the data and make it available across each channels to achieve personalization.
• Data needs to be centralized so that activity in one channel can immediately support engagement in another channel at real time. To achieve this, enterprises need three critical data management systems.

Personalization at scale requirement 2 – Customer Data Platform

• In many companies, customer data is distributed across disparate systems that are typically managed by different stakeholders.Customer data platform integrates data in flexible unified model to develop a customer identity that can be used consistently across channels.
• The data in Customer Data Platform is stored in cloud in order to handle petabytes of data that can be made available at low latency.
• The Customer Data platform should be customer friendly so that it can be used by non technical marketers. In addition it should have an analytical bench where data scientists can deploy AI models to create 1 to 1 signals that can be used for targeting and personalization.The Customer data platform must be connected to Enterprise data systems but at the same time should be a stand alone asset.The marketing organization should define the requirements for Customer data platform.
• The customer data platform can be built inhouse, off the shelf software or a hybrid model.The hybrid approach combines the inhouse data lake with external SaaS based Customer data platform.

Personalization at scale requirement 3 – Identity resolution platform

• The customer data platform relies on the first party data where as Identity resolution platform expand the pool of addressable customers and prospects.
• This is achieved by matching email ids of anonymously visitors to web browser cookies, unique ids of mobile and other multiple devices.
• Identity resolution platforms manage customer, device and location graphs across digital and non digital channels.
• Identity resolution platforms need to comply with regulations like GDPR in Europe.

Personalization at scale requirement 4 – Data Management platform

• A Data management platform receives signal data from Customer Data platform and activate the same in digital channels. A DMP also receives third party data to create additional micro segments.
• For example, a CDP platform have customer segment that have made recent purchases and visited the website recently.The DMP can sync this customer segment against third party data to identify sub segments within this group with similar profiles like millennials, adults and old people. The DMP identify prospects with similar behavior of customer segments and then target them with personalized messages/offers across digital channels.

In order to address issues of data trapped in silos, CMO and CIO organizations need to work closely to identify use cases where data is needed, defining how and where data will be combined, identifying systems that will use the data along with setting up a data governance and cross functional data governance council.

Personalization at scale requirement 5 – Decisioning

• Decisioning logic in current enterprises is not centralized and resides in individual black box systems which results in a disjointed customer experience.
• An integrated decisioning engine using AI and machine learning models needs to be developed in order to deliver a consistent customer experience. These models are built within the CDP to allow companies to predict the next marketing action to deliver to customers based on individual and microsegment behavior.
• The Decisioning engine should balance conflicting market rules in order to create consistent customer experiences.This is achieved by using Machine Learning/AI models which identify patterns not previously detected by static models.
• No single platform can act as a centralized Decisioning engine.Hence some companies build their own decisioning engine which results in high technical debt that can freeze a business.So marketing platforms are investing heavily to improve their AI based Decisioning capabilities.
• Marketing and Technology teams should work closely to identify areas for investments for inhouse solutions or off the shelf software. They can experiment with new Decisioning tools in the market to test their capabilities. These tools are constantly evolving and are offered at low costs.

Personalization at scale requirement 6 – Design

• Personalization at scale demands huge volume of content and experimentation because the number of different ways to address customers increase exponentially. This creates a need to design and create huge volume of content.
• In order to achieve this, companies need to move away from developing bespoke content and create a content factory. All content – art, videos, email, banners and even content for non digital channels need to be broken down to modular components for mixing and matching in dynamically populated templates to be delivered in multiple form factors. The master content is stored inside a Digital Asset Management repository and tagged for easy retrieval and tracking.
• The modular content is organized around a taxonomy that is based on multiple factors like customer segments, offer type, product categories and campaign type.This is needed to automate the offer selection.
• Once content and offer modules have been tagged, the decisioning engine can test on multiple content variations to deliver the best results.

Personalization at scale requirement 7 – Distribution

• All the digital channels should be integrated to coordinate communications and react to customers actions in real time.
• This is achieved by connecting data, decisioning and design elements to the marketing technology systems that deliver customer experiences.This allows companies to react at the time of opportunity.
• For example when a customer is shopping for a product, the data and decisioning systems identify which products to serve based on the customer profile.This triggers the content and offer repository to assemble the relevant content modules and serve them to marketing technology systems. The marketing technology platform then delivers the right offer to the customer in the relevant channel.
• The system tracks how customers react to the offer provided and this data is passed to CDP in real time so that it can learn which offer works and which does not.

Conclusion

• The technology to unlock the value of personalization at scale is available but one needs to address people issues because people are the gatekeepers of data.
• It is people who decide how to collaborate with their peers. They are the agents of change who makes personalization not only possible but also delightful to the customers.