For data-driven companies, enterprise data is considered the new gold. Every organisation dreams of using data to make more informed business decisions. But the reality is that insight-driven organisations still represent a minority of businesses. Many collect data but fail to have a viable strategy or a comprehensive view of the data they should collect.
To find out how businesses can use data to make better decisions, we’ll look at some insights from Markus Anderle, Vice President, Global Sales Intelligence at Salesforce. Markus leads a team of data scientists, data engineers, and business delivery teams. He also works in collaboration with Sales Operations.
Enterprise data refers to data that a business generates, collects, or purchases, which has the potential to be leveraged into actionable insights. Enterprise data is shared across an organisation, enabling it to be more predictive and make smarter decisions.
Nowadays, every company wants to be data-driven and use analytics when making important decisions, but that hasn’t always been the case. Not long ago, the enterprise model relied on senior management to guide the process. These days, they have plenty of help. By removing the bias of past experiences, ‘smart’ enterprise data can empower businesses to draw a clearer map to the future.
There are four pillars upon which successful, data-driven companies sit:
Let’s take a look at how businesses can construct these four pillars to build a firm foundation for success.
An overarching, inclusive approach is key for successful data management. This means including all the data you can think of, from employees and customers to operations. Companies can’t take a narrow view; they need a broad, holistic approach.
All data should be considered important. For example, the number of employees that ride a bike to work may seem irrelevant, but when put under the right lens, it can shed light on a workforce’s habits, lifestyle, productivity, and more.
Data should also be collected with artificial intelligence (AI) in mind. Using data with AI enables data-driven companies to make more informed operational decisions. If a business wanted to introduce robotic process automation, for example, then its AI and machine learning programs would need data to create algorithms and learn patterns. This is what future outcomes are predicated on.
As Markus Anderle says, “Snap the world – take a picture of your entire data and as it changes you need to snap it again, to be able to understand what decisions taken a month or three years ago have had an impact, and look at the outcome one month later. This is typically something people don’t do; when they collect data they don’t have this in mind. It’s either not collected at all or it’s collected in a way that is changing constantly.”
There are a wide variety of tools and technologies to help you become a data-driven company. These range from expert-level tools to out-of-the-box technologies that provide smart data insights quickly and effectively.
The Salesforce model is to have data scientists, data engineers, and business delivery teams attached to business units (Marketing, Sales, Customer Success, and so forth). This allows us to maintain a close relationship with the business and make quick, data-driven decisions.
These experts don’t use “out-of-the-box” technologies. Instead, they use low-level tools, programming language, open-source resources, and the ecosystem surrounding machine learning and AI. This is the most flexible approach.
Many data-driven companies don’t have teams of data scientists and engineers and instead utilise simple tools that can help them make better decisions right away. These out-of-the-box technologies – like Einstein and Pardot – can empower companies to keep up with their larger competitors, even without robust IT departments and data scientists.
Technologies require talent, but how do businesses know exactly what to look for? Do they need data scientists, data engineers, data architects, or business analysts? It can be tricky for organisations to pin down what talent will help them the most, especially if they have limited resources. A good place to start is with:
A data engineer/architect. Marcus recommends hiring someone that can:
Build a good model to capture the right data
Understand the different use cases for the company
Can inform the company on how to efficiently collect data
“Imagine this as a pyramid. At the top of the pyramid is the predictive modelling AI and at the bottom of the pyramid, you have data. You need to have a good foundation, and this foundation is data.”
A business analyst. Business analysts are also important. They have knowledge in using SQL for data queries, can use Python, and can write little programs that look at correlations in the data. Many also know some entry-level data science, which is a plus if you can only hire one or two people.
Organisations shouldn’t be intimidated when it comes to AI and machine learning. “The beauty of this area is that it is very accessible and anyone can learn it,” Markus says.
Training and reskilling programmes are becoming increasingly popular in the current climate. Businesses can easily put together a strategy to train a single person or an entire team. Data-driven companies should offer a comprehensive curriculum with the proper resources. They also need to pick the right people for the job, as this is a ‘people transformation’, after all.
“The retooling is a multi-year process, but enterprises most likely have folks like business analysts or people that show affinity or are interested in writing programs or taking a deeper look at data,” Markus says. “Build on those members of the company; give them the tools to be creative; encourage them to train. Invest financial resources in yearly trainings, and provide the time for that person to upskill, since it is so accessible.”
One essential factor in becoming a data-driven company is creating the right company culture. Leadership needs to roll out roadmaps and strategies to lead digital transformation at all levels of the company.
It’s crucial to close the gap between the decision makers and the people that understand the algorithms and their potential. This means closing the gap in terms of expectations, essential technologies, and talent.
Markus Anderle puts it like this: “Sometimes with AI, people think you can solve everything and make your company super effective but they are not seeing potential stumbling blocks: the data is not there; you don’t have the talent; you don’t have time or the buy-in. There are always some things to overcome.”
It's important to have a strategy that reviews:
How data is collected
What talents a business has and doesn’t have
What areas will see a profit
Where it makes sense and where it doesn’t
A data-driven company needs a multi-year roadmap in place, and they can then plan for small successes in areas that look promising. This is important and it is something that’s often overlooked.
We are producing enormous amounts of data. And the focus on data engineering and data architecture will only grow more pronounced. As data gets smarter and smarter, out-of-the-box tools may make the data science role obsolete. But in its place will be a new role – one that leverages data science and uses the array of innovative new technologies to make businesses faster, more agile, and smarter than ever before.
New technologies like quantum computing will become essential to manufacturing, engineering, and the banking industry. Businesses will need to keep up with these technologies to stay ahead. And if they do, the new frontier is filled with golden opportunities.
To see how Einstein can help your business leverage data, check out the Einstein Data Discovery demo.