Solve Your Company's Data Problem by Adopting a New Mindset

By Melissa Todisco, September 9, 2019

A few weeks ago, I was in a customer meeting with a large consumer goods company. They were excited at the idea of building a simulation of their market, but were unsure if they had all the data needed to answer all their questions. The meeting brought together people across multiple departments – strategy, insights and digital - to discuss:


• What data do we need?
• How much data do we need?
• How granular does it need to be?
• How far back does it need to go?
• I don’t have individual-level data, is that ok?
• Can we automate the process?


At the beginning of the meeting, I could sense there was a feeling of overwhelm. They were concerned about the amount of effort it would take to gather data because it was spread out all over the company. Throughout the meeting, we talked about our mindset around data and everyone left the room with a sense of relief and a clear path forward.  


Here are 4 practices that we encourage our customers to use to overcome their data problem: 

Shift From a Scarcity to an Abundance Mindset

Most analyses start with not having enough data. That’s why you are doing the analysis. While you personally may not have access to the data today, that doesn’t mean you can’t get access to the data. We live in a world where everything is tracked. There is always a way to find it, buy it or start collecting it. We recently developed a data readiness assessment to help people evaluate whether they have the required data to answer specific business questions. While most respondents had most of the data, the majority were missing at least one piece of data. For example, maybe you don’t have access to market research data today, but it isn’t that hard to field a survey of your target market to learn about awareness and perceptions for your brand and your competitors. Or maybe you personally do not have access to marketing investments. That may be easy data to collect after a conversation with the right contact in your organization.

Focus on the Business Need Before the Data

It is easy to become quickly discouraged when you start an analytics project thinking about all the effort. However, if your project is grounded in a business need that will deliver value across the organization, the spirit of the project shifts. When I was at DataX San Francisco this past year, Pallav Agrawal, Director of Data Science at Levi’s shared this framework as a great way to elicit questions from the business to prioritize analytics projects:

Take an Agile Approach to Analytics

Agile is a concept that started in the world of software development, but has expanded into analytics and marketing as a work philosophy designed to deliver projects on-time, on-budget that make a higher impact. The core principles of an Agile mindset include:

In Ken Collier’s book Agile Analytics: A Value-Driven Approach to Business Intelligence and Data Warehousing, he talks about how you can bring far greater innovation, value, and quality to any data warehousing, business intelligence or analytics project.

Practice MVP or Rather MVM (Minimum Viable Model)

Minimum Viable Product was a concept popularized by Eric Ries in the Lean Startup that stresses the importance of learning in product development. Eric Ries defined an MVP as that version of a new product which allows a team to collect the maximum amount of validated learning with the least effort. Modeling is also an iterative process that often requires input from a business decision-maker. However, a business decision-maker often doesn’t want to become involved until there is some kind of answer to evaluate. Practicing MVP (or MVM), allows you to get feedback early on to make your model better and more robust with time.

Despite the fact that big data has been on the rise for the last decade, many large organizations are still trying to figure out the most effective way to bring together different datasets in order to create a what-if scenario planning capability for business decisions.

It isn’t that the data doesn’t exist – but rather there is a process that requires collaboration to get to the finish line. Because we live in a world of convenience where Amazon can have something at your doorstep the next day (or even the same day), it can be difficult to motivate all internal stakeholders to jump on board to a project that will be an iterative process. That is why only 9% of organizations have achieved analytics transformation and data scientists commission such a premium salary.