Big data and data science are terms that are on every business’ radar these days, but like many industry buzzwords, it’s good to pause and consider the actual value your business can derive from data.

Even as someone who has worked in data science for the last eight years, I don’t believe businesses should always invest heavily in data. Doing so only makes sense if data is solving problems that are critical to your success as a business. Here are three common scenarios where this is the case:

#1 Data is a part of your core product offering

The top reason to invest in data is when the data is itself a core part of your product’s value proposition to the customer. Take for example Google Maps, a product whose primary function is to provide a convenient interface to GIS data.

For such a product, you need to build the right infrastructure for storing, curating, processing, and querying the data. You need to hire the right people who can build this infrastructure to scale and who can work with data at scale.

Every business’ goal is to scale their user or customer base, and with a product built upon fast data computation, you will need to be aware of how you can support a rapidly growing store of data.

#2 Data makes existing products smarter

With the rise of the data natives, we have an entire generation of consumers who just expect their products to be smarter and more intuitive.

If you’re not catering to the modern consumer, someone else will. 

Building smarter products often relies on collecting extensive usage data about how consumers interact with your product, and then harnessing this data via machine learnt models and algorithms to further improve the user experience.

For instance, if you are building a reading app for smartphones, the most basic, first version you build might be a simple app that allows users to write reviews for books they have read. You might however, want to quickly follow this up with a recommendations layer that suggests new books to users based on reading patterns of other similar users. The algorithms that generate these recommendations become more and more accurate as the app collects data about how users interact with the recommendations. This requires instrumenting your app to collect exactly the right usage data that can trigger this virtuous feedback loop. Off-the-shelf solutions for recommendation systems do exist, but can be rigid and difficult to adapt if there is anything custom about your app.

So think about whether the intelligence value-add that comes from data will be critical to the future success of your product. If so, invest in data scientists and data engineers early on. They will play a critical role in making key decisions around data collection and product design needed for creating intelligent features that your users will not only love, but expect from your product.

#3 Data helps inform future product directions

Maybe data doesn’t play a front and central role in your product. Even so, investing in data makes sense when you want to take a data-driven approach to improving your product.

For instance, a product manager at Microsoft Office might study usage data in order to figure out how frequently the copy-paste command is used and how the command is commonly accessed. This in turn would inform whether the copy-paste command needs to be made more or less prominent, and whether access to the command needs to be made easier. Studying common access patterns might also help inform the product manager of common workflows that Microsoft Office could better help support.

Many organizations also employ decision scientists to analyze key business metrics such as growth, engagement, and drivers of these metrics to inform future strategic business directions. A decision scientist might need answer questions such as: What would be the impact of a change in the pricing model of a mobile app on growth and engagement?

Data-driven product management and decision making also relies on extensive instrumentation and availability of usage data in order to build product roadmaps and make critical decisions. This makes most sense once you already have a minimum viable product and a sizable user base. But keep in mind, engaging a data scientist or engineer early on will be important for designing the right data collection strategies. Invest in data if you are committed to making key product and business decisions based on data, and not simply on intuition.  

Is data relevant for your business?

The ease of gathering data in an always connected world coupled with the modern consumer’s expectations for customization and intelligent features means software companies need to keep their data needs top of mind. Consider these three scenarios and prepare your team early on to stay competitive in a present and future where data is everywhere. 

Want to hear more about our perspectives on data science? Read our interview with SalesforceIQ's Vitaly Gordon on the Future of Data Science here and here. 

Shubha Nabar is a Director of Data Science at SalesforceIQ, where her team is creating intelligent tools for Salesforce’s Customer Success platform. Previously, Shubha obtained a PhD in Computer Science from Stanford University and worked as a Senior Data Scientist at Linkedin.