Narrative Science is now part of Salesforce!

Learn More

Blog 5 Actions to Take Now to Build a Better Data Culture

When done well, leveraging data is the ultimate competitive advantage — beyond the technology you can buy and beyond the products that you can sell. However, becoming a data-driven organization requires investments in both culture and technology to shift how people make decisions. Day-to-day, Bruno Aziza, Head of Data & Analytics for Google Cloud, advises companies who are looking to make this shift. 

Here are 5 actions that he recommends organizations take now to begin building a sustainable data culture.

Start Establishing Freedom & Framework

Let’s be honest. Data adoption is a really hard problem to solve. 

This idea of making sure everybody has access to data, but also making sure that that access is governed — it comes down to freedom and framework. Everybody should have access to the data they need in order to make their job a better job and innovate for their company. But at the same time, not everybody should have visibility on all the data.

Data has grown tremendously and not just in volume. Now you have data that you’re pulling from on-prem systems across multiple clouds, created by applications that we didn’t have 10 years ago. It’s a problem that just keeps on giving. Every step that we make towards fixing this freedom within the framework is two steps backwards because the complexity is increasing. 

It’s the ultimate challenge for the Chief Data Officer — creating freedom so everyone can use data, but creating a framework so that data does not become a liability. The last thing we want is for the Chief Data Officer to be the Chief Governance Officer. Those individuals need to have the infrastructure and the tools in order to enforce policies centrally and make sure that everything runs effectively so they can focus on innovation. This is where the cloud can really help.

Stop Hyper-focusing on Your Competition

Adopting a data-driven culture should be about using data as a means to engage and empower your employees i.e. your engine for growth.  It should also be about creating better experiences for your customers. Not simply, “This is what our competitors are doing — now let’s plug and play”. The factors and metrics that should really motivate this type of change are customer churn, customer experience, etc. Am I able to provide services faster? Am I able to detect new opportunities for me to offer better services and recommendations? 

Start at the Top

A sustainable data-driven culture is built from the top-down. Bruno identifies 3 red flags that may indicate it’s time for a change in strategy.

  1. If you don’t have an executive at the top of the organization that is mandating that data is the way the entire organization makes decisions. 
  2. If you don’t have a Chief Data Officer who’s executing and ensuring culture transformation.
  3. If you’re not living it with an organizational footprint and have an org chart that is actually representative of how every function is going to take advantage of data.

Now, who should the CDO report to?

The truth is, there’s no single answer to this question. However, a great starting point is to consider what department is the center of gravity for your organization. Which department is driving the direction of the company? Sales? Marketing? Finance?

Your CDO and data team should probably go under that.

At the end of the day, the appropriate choice is dependent on the company, the department that primarily drives the business, and the mindset of executive leadership.

Stop Falling Victim to Vendor Speak

We like to label things. It’s human nature. And vendors aren’t an exception.

It’s easy for vendors to oversimplify customer’s use cases. They may even invent new terms or concepts that sound great but don’t solve the real “job to be done”. 

This often happens in conversations about data warehouses vs. data lakes. While data lakes and data warehouses are both widely used for storing big data, they are not interchangeable terms. They have two different use cases and personas.

Data warehouses are typically better suited for structured information where performance and the quickness of the answer is what matters to the end user. A data lake is more suited for unstructured information and exploration. Data scientists work more with data lakes, while data or business analysts work more with data warehouses.

In summary, don’t let vendors label your use case. Instead, define it yourself. 

Start Centralizing Data & Decentralizing Analytics 

Why should you centralize data?

The first reason is governance. You want to have a central way for your company to absorb and leverage data in a way that is governed and secure. The second is data protection. You can 1) virtually store anything for very cheap 2) run ML models at a very low scale and 3) return results on it very quickly. Ultimately, centralizing your data will help you return better results from your models.

Take Spotify for example. When you have tens of millions of people listening to music every minute of the day, you have access to an extraordinary amount of data.  As the service continues to acquire data points, it uses that information to train the algorithms and machines and discover insights that contribute to a more personalized experience of listeners. By having that data centralized, ML models are able to run at scale for the service of creating value for the end user. 

By decentralizing analytics, data analysis is not limited to the responsibility of a single data team. Instead, it is accessible and in a perfect world, embraced by everyone in the company as a means for better decision-making.

Start The Move to a Cloud Architecture

Organizations of all sizes and types are embracing digital transformation as an essential, strategic business objective. Moving to the cloud can be a key enabler of digital transformation, especially when considering its many benefits — increased agility, nearly unlimited scalability, more efficient use of resources, end-user self-service, etc. Cloud architecture can be the key to unlocking issues around adoption, governance, and security. This move can enable the Chief Data Officer to focus more on innovation and less on the nuts and bolts of the operation.

To hear more tips from Bruno, check out his episode of Leading with Data here.

Subscribe to the Series

Listen to Leading with Data to hear stories & strategies from changemakers in the data & analytics space.