This is the first post of a series on how companies can get smarter by using data. It begins with an audacious goal to inspire companies to move beyond the typical approach of building models and reports and allows all of us to explore the idea of the intelligence-driven company.

Tl;dr

Company decision making has evolved from gut-driven to data-driven. Intelligence-driven companies are the next evolution where decisions and actions move from humans to machines.

“Data this, data that.” So much data!

The pressure for companies to become more data savvy is both omnipresent and overwhelming. “Data is the new oil”, as the zeitgeist goes. The current competitive advantage is to become a data-driven company. Becoming data-driven is a major and difficult evolution for many companies. Despite the current attention on data-driven benefits, it also helps to look at where things are already heading. Today’s competitive advantage will not guarantee a win tomorrow. Technology and trends iterate much more rapidly, and the business decision/action cycle will only get faster.

There is already a new paradigm underway that I call the intelligence-driven company. Much of this change is ushered in by the rapid advances in teaching machines to be smart – artificial intelligence, machine learning, deep learning – as well as cheap and ubiquitous cloud computing, sensors and automation. It’s a new era where intelligent and automated actions replace data-driven human decisions, and where these decisions instantaneously affect countless downstream data points. The intelligence-driven company will be able to adapt and thrive in ways unimaginable by its data-driven peers.

Where we’ve been…where we’re going

There is a maturity of decision making that companies have historically followed. Sometimes companies follow a combination of these, other times not. Roughly speaking, the evolution of the decision making process can be seen by the introduction of new trends over time:

Old school: The Gut-driven Company

A gut-driven company is any company that operates in near or total absence of information. It’s pretty much how humans have run their businesses since day one. To this day, the gut-driven style still persists in countless businesses around the world. Millions of small businesses from the local Bangkok street market to the Brooklyn corner dry cleaner likely operate without actually looking at data, but rather decide where to take the company based on gut feel and intuition.

A cousin of the gut-driven company is the HiPPO-driven company, in which the highest ranking/highest paid person in the company guides the decision making process. Subordinates fall in line behind the HiPPO, and trust that the HiPPO’s instincts are sufficient to move the company forward.

On the continuum, gut-driven is as simple as it gets to running a business. Thankfully, humans are remarkably good at pattern recognition with small amounts of data, as well as making quick decisions. And the gut-driven approach is still widely used, and can be successful against competitors who operate at the same level of sophistication. But when gut-driven meets data-driven, competition can get very lopsided. In today’s business climate, evolving is key to surviving and thriving. Future posts will cover in great detail how a company can evolve beyond being gut-driven toward data-driven and beyond.

Current phase: The Data-driven Company

Data-driven companies are the next phase of sophistication. In its purest form, a data-driven company allows anyone in the company to draw from a single source of truth to make informed, real time and actionable decisions. This data-driven decision making is woven into the fabric of the company. A data-driven company is a formidable competitor against less sophisticated rivals.

Data-driven companies will see a lot of benefits.

• “23 times more likely to acquire customers, six times as likely to retain those customers, and 19 times as likely to be profitable as a result.” (Forbes)
• “Organizations are continuing to make investments in analytics to meet the growing demands of the user community for more robust and usable analytic solutions. In analyzing Nucleus ROI case studies on analytics, we found the average returns from analytics have been increasing, reaching $13.01 for every dollar spent in 2014 from just $10.66 in 2011.” (Nucleus Research)
• Data keeps growing…and growing…Making sense of the growing data is vital. Globally – ”According to IDC Research, digital data will grow at a compound annual growth rate (CAGR) of 42% through 2020. In the 2010-2020 decade, the world’s data will grow by 50X; i.e., from about 1ZB in 2010 to about 50ZB in 2020.” (EETimes)

Becoming a data-driven company is very hard work. Christopher Penn does a great job at breaking down the evolution of a data driven company.

1. Data-resistant – The company is reluctant to let go of gut-driven habits. The biggest hint you’re in this stage is when you hear, “We’ve always done it this way”. Another indicator is politics or arguments around data integrity and accuracy, but no incentive to fix the underlying data issues. In this case, data issues are used as a red herring to divert attention from more important matters.In my experience, moving out of the data-resistant stage is typically the hardest part. This often isn’t a data issue; it’s a human and cultural issue. It sometimes takes a lot of work to convince key stakeholders that embracing a data-centric approach is worthwhile. Culture and gut-driven habits are hard – and sometimes impossible – things to turn around.But, if you’re lucky enough to get the company on board with moving toward being a data-driven company, the next steps are often easier.
2. Data-aware – Data is already likely available. The question is how to make sense of the data? At this stage, the company will take baby steps to identify the data it already has, and decide how to create value from data.
3. Data-guided – Data is a tactical asset. Companies begin extracting value from data to answer questions such as “What happened?” and “What customer?”. At this stage, the reliance on gut-driven decision making begins to wane. The company will start investing in resources such as data storage, data quality, data governance, data warehousing, and business intelligence (BI). Data is moved out of silos. Analysis and reporting begins to be centralized and owned by a data team, usually consisting of a data analyst and a data engineer. A data scientist may be hired now, or in a future stage when solving predictive problems becomes more important for the business.
4. Data-savvy – Data is a strategic asset. Questions begin to grow toward, “Why”-based, such as “Why did customer churn increase last month?”. There is a single source of truth for data. Data is used to drive insights, and there’s a move to become more action-driven.
5. Data-driven – Data drives action. Actionable insights driven by a deep understanding of “What” and “Why”. Data powers the company’s planning, discussion and major decisions, and this slices across company hierarchy. Everyone is empowered to make actionable, data driven decisions from a single source of truth.

The evolution to data-driven can take several years. And as you can see, the benefits are extremely powerful. Be patient and put in the work every single day to push forward.

Next phase: The Intelligence-driven Company

The next evolution beyond a data driven company harnesses the power of machine intelligence and moves beyond simple actionable-insights based reporting. I call this next phase the intelligence-driven company. The intelligence-driven company uses the power of machine learning to produce intelligent, automated actions. Instead of reading a report and taking action, intelligent software agents simply take the action for you. The human in the loop may choose to review the actions taken. But the action and report are essentially decoupled. Depending on the task, no longer does a human need to view a report, deliberate a decision, and take action. The action is already taken.

The challenge with the data-driven approach is the friction involving the time, energy, and emotion (we’re humans, and emotion is inescapable) in the feedback loop of analyzing a report and taking action. Intelligence-driven automation eliminates this feedback loop. Time-to-action is drastically reduced; time-to-value increases proportionately. Also, human emotion is removed from the feedback loop. What is left is increased productivity through intelligent automation.

How does intelligent automation differ from rules-based automation that has existed for decades? Rules based automation is heuristic based, normally relying on if-then statements or codified rules engineered by humans. The upside with is that a rules-based approach may capture the majority of ways to answer a question. An example is “Is this number less than $1 million?” A simple if-statement can give a yes or no answer to this question. The downside is that it is incredibly hard to account for all edge-cases with a rules-based approach. For example, the question, “What product should I recommend to this customer?” cannot be likely answered with simple rules.

Intelligent automation offers an alternative to rules-based approaches by looking for patterns across bodies of data. To answer the question, “What product should I recommend to this customer?”, there must be sufficient examples of customer and product interactions so the algorithm can determine what product to recommend when a particular customer is seen. This is a benign example, and has been utilized across ecommerce for decades. What becomes more interesting is when this intelligence-driven approach becomes commonplace across all aspects of the company – marketing, sales, finance, supply chain, HR, and more. And what happens when companies start communicating with each other via intelligent automations? Pretty exciting stuff indeed.

Of course, pure unattended automation raises a number of risks. These risks have been raised for decades, as you can see in this 2009 IEEE article, “Are we automating ourselves into a corner?”. This paradox of automation is certainly a concern. On one hand, intelligent automation offers a panacea for human and information related friction in the workplace. On the other hand, if left unchecked, a once intelligent algorithm may turn stupid and take faulty actions. The results could be slightly malign to disastrous. For intelligence-driven initiatives, I expect a human will need to be in the loop, at least at first. This human will have the domain and technical expertise to understand how to diagnose and fix a problem in an intelligent machine misbehaves. Because of the risks involved with intelligent automation, I suggest a hybrid approach that blends data-driven with intelligence-driven actions.

Conclusion

Business is at the cusp of a new revolution where intelligent automation will augment or supplant the old ways of doing things. The promise of the intelligence-driven company is something that needs to be on most companies radar. A few companies are likely already in a position to start capitalizing on this next wave. Given the exponential nature of technological advancement, expect the next 10 to 20 years in business to provide more exciting evolution than the last several hundred years.