So you need to integrate customer data

Got data? Of course you do. In today’s digital platform enabled world of marketing the challenge is often not what data you have but can and how you use it. With each new tool, custom field, and data warehouse, companies expand functionality but can also reduce their ability to deliver personalized engagement from a single view of the customer. 

Why is this important? 

According to Gartner, companies spend close to 30% of their total marketing budget on digital tools meant to deliver personalized communication at scale. While these tools generate, analyze and store rich data elements, the task of integrating the data to be stored and retrieved becomes increasingly complex. This creates several challenges including siloes, conflicts, costs, and compliance.

 Why is this so hard?

• Escalating needs: As companies grow and elevate their level of maturity, they often find the need to centralize data from diverse platforms to realize the personalization functionality promised from their expensive marketing automation and CRM platforms
• Overlapping tools: Each new marketing stack tool can increase the complexity of data exponentially, often creating a mess of API’s, webhooks, 3rd party data appends, data warehouses and Customer Data Platforms. This is compounded with redundant platforms.
• Speed to launch: in their rush to implement, developers sometimes create work-arounds, static tables that they plan to go back and fix later. Stabilizing these shortcuts is a challenge due limited documentation and shifting IT resources
• Governance: Data is frequently “owned” by the channel teams who generate it. While fine in principle, these teams bring a siloed view driven by the tools they use rather than the holistic value of the data they generate
• Industry specifics: Many industries have unique data sets or customer requirements that create additional complexity. While at Laureate University, for example, my team struggled to manage 2 email addresses (personal and student issued) in the gold standard contact schema

6 Steps to a successful integration

While each data integration effort has its own set of requirements, I've learned the hard way that successful integrations require 6 basic steps.

1.      Organizational alignment & support: Data integration is detailed work requiring long term collaboration and commitment. Recognizing the and involving the key stakeholders in the solution will ensure your efforts have an impact. This includes resources to identify, supply, and discuss each stakeholder’s data and tools and the time to accomplish the work. It also includes some decision making body with stakeholder representation.

2.      Embrace a single view of the customer approach: Applying a “gold standard” customer record approach goes beyond marketing buzz by solving a fundamental network complexity issue. This is well articulated by Hull.io who show how the number of 2-way data integrations required between tools increases exponentially with the number of tools added but stays constant when you apply a single customer record approach. 

3.      Understand the data sets: This step relies heavily on subject matter experts provided in step 1.

• Ingest all tools, tracking, and databases
• Define and organize by source: sometimes called a data catalog, this effort captures an inventory of each data set, categorizing the data type, source, format, storage, and other features

4.      Determine the requirements

• Transactional vs analytics processing: Some data sets are needed for real time transactions and personalization and require more robust integration. One example is the personal data used to prefill a form and speed checkout, another is a new lead. Other data sets are more passive and can be updated at times of lower activity. Judiciously managing the difference is critical to data network fidelity and managing data costs. 
• Define resolution rules: Data stewardship includes the need to define rules for managing conflicts, gaps, and low confidence data. Defining these rules upfront ensures consistency and reliability. 
• Compliance: GDPR, CCPR, and other regulations are now business as usual for data management

 5.      Transform and integrate: These steps help to make sure your your data clean and fully usable.

• Enrichment: Integrating 3rd party data and appends in required format and permitted use rules
• Cleansing: remove old, unreliable unusable or just plain bad data
• Segmentation: analyze and apply segments and personas
• Transformation: Where necessary, update the data formats to be accessible
• Map fields and filter data flows to tools: super important, this is where you ensure you define the sources and uses of data by platform. Best practice is to create a hitlist of core fields at the contact, company, segment, attribute and event levels as a baseline set that hits 80% of platform requirements

6.      Governance: Centralizing a team to “own” and manage the data as an asset is a best practice. Here are some recommended elements of this centralized governance team:

• Gold standard record: Manage the gold standard record, access to it and any future modifications
• Data resolution rules: Articulate and share the data resolution rules. Encourage debate and alignment
• Compliance, cleansing and retention: define the core policies and processes and support local teams in the execution of them
• API, webhook, and 3rd party append management: Define standards and best practices for integrations and either centrally manage or support teams who execute them
• Platform integration & removals: Define and share new platform guidelines with stakeholders and support the assessment of new platforms as needed

While the task of integrating diverse sets of data and tools is daunting, the rewards are immense. Companies are increasingly recognizing data as a core asset and applying resources and focus to manage it as such. 

Contact me and let’s discuss data challenges you face.