Video Podcast

Stay up to date on what's happening in the digital transformation space by signing up for our newsletter, here.

Introduction

Across industries, the shift to digital solutions has exponentially increased the amount of data that’s generated - estimated at more than 2.5 quintillion bytes per day and growing. As a result, the way that organizations define, create and measure value has changed, along with the speed at which they use that data to drive innovation.

But just collecting data isn’t enough - organizations need an analytics strategy for how to leverage that data to fuel innovation and generate value.

What is analytics

There are probably plenty of different definitions of what analytics is, but the one that I keep coming back to is that analytics is “the process of turning data into insight to make better decisions.''  The reason I prefer this definition is that it makes it clear there are three important ingredients - data, insight, and decision - and if you are missing one of those, you’re not going to be able to leverage data to drive innovation and value creation in your organization.

More organizations are recognizing this and taking steps to unlock this value, but are facing a challenge making it stick.

Barriers to adoption

It used to be that the major stumbling block in leveraging data was storage and technology. It required both a financial and physical commitment to collect and store data, forcing many organizations to be judicious about what they collected and how they integrated it into their legacy systems and processes.  In fact, in one of my very first analytics roles, we only had 2 years of transactional data to use in our warehouse and had a process that rolled data off every Sunday night to manage space. It also wasn’t unusual for queries and models to run for hours, which stifled a lot of “what if” and exploratory thinking because it simply wasn’t practical to spend days on a project without a strong business case.

The introduction of cloud services like AWS have all but eliminated both of those barriers.  Data collection and storage is now a fraction of the cost it was just 10 years ago, and computing power can be dialed up/down on demand.  The other key game changer is the rise of SaaS, which makes it possible for organizations to mix-and-match different capabilities and services to create bespoke solutions for a fraction of the cost of enterprise software (or the time required).  Products like Google Analytics, Heap, Mixpanel, or Mode make it extremely easy to start unlocking value from data.

So where are today’s challenges?  I believe that they lie mainly in how organizations are approaching how they use data and analytics and integrate it within their organizations.  In my experience, there are four main causes:

  • Foundation Overbuild - this is the belief that in order to be used successfully, data needs to be extremely clean, well organized, and that legacy systems need to be integrated and consolidated before any real analysis can start. This requires a tremendous amount of work to be done with little upfront benefit. In short, there is a misalignment around data.
  • Poor Questions - this is when analytics investments and direction are guided by vague questions like "what trends do we see?" or "where is there opportunity with my customers?".  This almost always results in fishing expeditions that deliver little/no value and may have little/no chance of being implemented. In other words, there is no focus for insight.
  • Analytics Isolation - this occurs when analytics initiatives and resources are not integrated (or at least a part of) teams with decision-making responsibilities.  In these cases, little of the team’s work actually makes an impact on the overall business - model recommendations are not integrated into workflows, insights are not widely circulated, and test results ignored. In short, it didn’t drive a decision.
  • Long Path to Value - this is when the focus is on monolithic deliverables that take a long time to come to fruition.  Often, these initiatives lose momentum and mistime market opportunities, further reducing the value they can deliver.  Organizations need to be more agile to build, measure, and learn faster.

Framework for leveraging data

There are many ways to overcome these obstacles to leverage data and analytics, and your approach will be dependent on your organization’s maturity and the sense of urgency.  That being said, I’ve found that there are 4 common elements in starting to properly leverage data:

  1. Start with a use case and focused question - data is like money in that it only generates value when you use it, so it’s critical for an organization to identify a specific business challenge or question that they want to use data and analytics to solve.  

    Embedded within this is need for strong linkages back to larger organizational objectives or goals.  In my experience, a good use case should a) have a strong link to broader business objectives to ensure alignment; b) address a known pain point or opportunity; and c) have direct, definable and measurable outcomes.

  2. Evaluate and collect data based on needs, not wants - it’s tempting to cast a wide net and gather up as much data as possible, but there are two main problems with this approach. First, there’s no theoretical end to “everything” so you can never really say when you’re done.

    Secondly, each piece of data you collect or want to use comes with a “tax” - money you might need to spend to set up data collection, time you need to spend to evaluate whether the data is any good (more on this later), work effort to combine and structure and possibly remediate the data.  Paying some tax is unavoidable, and you certainly do not want to skimp here, but like all other taxes, you don’t want to pay more than you have to.  Being ruthless about what you need vs would like to have will ensure that you remain focused on your use case, minimize your data “tax” and keep you moving forward as fast as possible.


  3. Answer the “next” question - no organization achieves their goal or objective in one shot, and you shouldn’t give into the temptation to get to the perfect solution or “the one” answer before you take action.  History and psychology have shown that focusing on immediate questions and generating incremental gains is the way to deliver large scale impact.

    The example I keep coming back to is the Apollo program - it started in 1961 but only landed a man on the moon in 1969. They spent 8 years building, testing, learning and then repeating the cycle again. They identified what they had to learn next to keep forward momentum, focused their energies on the next learning and remained disciplined. No doubt if they tried to focus on solving all the problems before they launched a rocket, we’d still be waiting.

  4. Think consumer-centric - in many cases, deep insights and value-driving products are built on the back of consumer-specific data.  But, consumers now have a choice about what data and information they share with you, and they’re becoming far more savvy about what they will and won’t share.  I believe they are willing to make the trade off between their information and your product, provided that it’s valuable, either solving a problem or making their life easier.  

Think about Amazon Echo or Google Home - they’re essentially listening devices that Cold War spies would have loved to use. Today, we accept them into our homes because they make managing the lights easier and faster to find things on the internet.  Thinking like a consumer as you’re identifying problems, developing modelled recommendations or developing products will help create adoption and a competitive differentiator.

Analytics and Innovation

Having an analytics and data strategy is important, but in order for it to generate value and benefits, it needs to be linked to and work within the larger organizational innovation and transformation strategy. What role does analytics play within an innovation mindset?  

I see analytics as a validator and enabler of an organization’s innovation and transformation strategy.  It does this in three ways:

  • It ensures organizations focus on creating things of value by translating common language value propositions and hypothesis into hard metrics and measures.  And this applies for both consumers and the organization as a whole - costly services or offers that consumers love aren’t sustainable financially, and products that consumers don’t use, recommend or return to need to be re-examined.
  • It supports constant improvement as organizations understand not only how they are performing now, but the mechanics of what drives that performance so they can set goals and start testing.  It identifies the areas to focus on improving and, most importantly, how we will measure that improvement.
  • It aligns with a lean startup approach by only building what is needed so organizations learn quicker and make informed decisions.  This helps ensure that analytics is integrated into the overall development process, strengthening adoption of its practices and outputs.

Simply put, an analytics and data strategy needs to be incorporated as part of an organization’s broader innovation and transformation strategy to provide an answer to that all-important question - is this working?

Where Do We Go Next

For all ways data and analytics have changed organizations in the past 5 years, there’s still more change on the horizon. While no one knows for sure what will happen, I think there are three trends everyone should be on the lookout for:

  • Without a doubt, the depth and breadth of data will continue to grow exponentially, with the increased use of social media, introduction of IoT sensors and products and shift of offline or analog processes to digital
  • Technology will increasingly be used to simplify the connections between all this data, reducing the amount of time and effort required to prepare and ready data for analysis.
  • And lastly, an increased focus on framing or “translating” - defining the right business question, identifying the right data and analytical approach, and working to ensure that data drives action - to take advantage of automation and artificial intelligence capabilities that power feedback looks and contribute to the innovation cycle.

Successfully using data and analytics to drive better decisions and support your innovation strategy is about more than hiring data scientists.  It’s about identifying the right questions you need data to answer and focusing your efforts around a defined problem, creating strong linkages with your overall business objectives and your innovation strategy, and focusing on delivering customer value to ensure adoption. It is not easy work, but it’s absolutely essential for organizations to compete and win in this digital world.