Overcoming 5 top data infrastructure missteps

We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 – 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!


Despite increased investments in data infrastructure platforms and applications, many organizations still make significant missteps. As a result, they fall far short of goals for improved efficiency, increased productivity and greater profitability.

There are a few key reasons. They include a lack of agility, a failure to adapt to change, and the inability to scale outcomes. In response to these problems, Michael Katz, CEO of MParticle, shared his thoughts on why organizations fall victim to these challenges, and what they can do to overcome them.

Founded in 2013, MParticle is credited with largely creating the customer data platform space. The firm provides customer data infrastructure that helps teams unify their marketing stack and improve customer engagement. Katz says MParticle does not deliver the campaigns, “we make them better by improving data quality and continuously enriching customer context across multiple channels and partners.”

The following are five of the top missteps that Katz sees customers making and what is required for successful execution.

Data infrastructure misstep #1: Not starting with a clear first-party data strategy.

“The key to avoiding it is improving collaboration among data creators and data consumers and aligning the goals of the business and established workflows to data collection requirements,” Katz said.

Data infrastructure misstep #2: Assuming a steady state in the business rather than solving for data entropy.

“The key to avoiding it is first to define your risks and identify your vulnerabilities to those risks,” Katz explained. “Next is to implement a data architecture with an integrated observability framework that can help the business become more resilient. Lastly, the organization needs to establish workflows for escalation.”

Data infrastructure misstep #3: Creating an ELT architecture without clear business goals and requirements up front.

Katz stressed that “The key to avoiding it is committing to a collaborative data design process and establishing well-defined workflows around data governance and consumption.”

Data infrastructure misstep #4: To rely on excessive dbt transformations as a quick fix to poor governance and data debt.

“The key to avoiding it is to treat your data as a product,” he said.

Data infrastructure misstep #5: Relying on generalized data pipelines to solve complex customer engagement use cases.

“The key to avoiding it is becoming educated on go-to-market requirements, marketing-related nuances, and choosing a more opinionated/purpose-built pipeline such as customer data infrastructure,” Katz said.

In addition, Katz emphasizes that no data platform or pipeline alone will solve anyone’s challenges.

“Success requires collaboration among teams, a cultural willingness to embrace change, proper incentives, and the right workflows,” he explained. “More specifically, it starts with creating a first-party data strategy that aligns data collection requirements with business goals and outcomes. It requires a commitment to actively investing in protecting and improving data quality and establishing a governance framework up front.”

Originally appeared on: TheSpuzz

Scoophot
Logo