👋

Hey ,

You can feel like a headless chicken if you keep running without a destination in mind. A study at Dominican University (contrary to popular myth that it was done at Harvard or Yale) found a similar conclusion: folks who wrote their goals accomplished 50% more than those who did not write their goals. 

Data is the new oil, so the saying goes. Raw crude oil gushed from the earth, it is not that valuable unless it is refined into different types of products - petrol, jet fuel or plastic. Like crude oil, raw data can only be useful once we organize, analyze and present it into a coherent story.

Yet, processing and analyzing just for the sake of it, will turn you into a headless chicken. What is required is to define a goal, ask a question or define a pain point in concrete terms at the outset. You don’t want to be part of the statistics where more than 80 percent of data analytics projects did not deliver value (Gartner).

As I outlined in the last post, there are five simple steps on any data journey 1) define your goals; 2) gather and/or acquire data; 3) organize and cleanse the data; 4) analyze and 5) visualize the results.  

In this post, I would like to explore in more detail on how you can set goals for your data project. Because setting the right goals in the beginning of your journey increases the odds that your big (and small) data projects are successful.

How to set up the goal post?

There are several methods to setting up project goals. The first one is Start With Why. It is a framework created by Simon Sinek. Next is setting goals the SMART way. I know it sounds cliche but bear with me. There is an interesting fork to the plain-vanilla SMART framework.

The other is known as OKR - Objectives and Key Results. This methodology is widely used in tech circles.  

Start With Why

Simon Sinek, gave an inspirational TED talk and wrote a book with the same title, Start With Why about a decade ago. The central thesis he is trying to convey is that each company would do well by asking the existential question “Why do we do what we do?” The answer to this question might bring about more clarity and focus to how a business can stand out from similar competitors by communicating its uniqueness.

The core of Sinek’s theory is threefold - Why, What and How - as encapsulated in this Golden Circle framework.

Source: SmartInsights

By stretching our imagination a little bit, we can repurpose this Start with Why method to setting goals for our data project. Starting with why means articulating the problem statements, pain points or goals we would like to achieve from the project. Thereafter, we can specify the outcomes or results we would like to achieve (the what) and set a plan for achieving the results (the how). 

If you keep asking Why, you will get closer to the reason you’re working on the data project. Or you can also do the same for your life too :)

The Clever way to be SMART

If you think the Start with Why framework is too psychedelic to folks around you, perhaps it might be smart to turn your attention to the evergreen SMART goal setting framework.

The acronym SMART means: simple, measurable, achievable, relevant and time-bound. The goal of this framework is to help you define a goal with greater clarity to ensure alignment with everyone in the team, right from the top all the way to your peers. And it does make you look, sound and smell smarter in the process.

Let’s quickly dive into each of the SMART components in greater detail.

Source: Adobe

• Specific: Precise, simplistically-written and easy to understand, even for someone with basic knowledge of the project.
• Measurable: The ability to use metrics to determine the success of a given project.
• Achievable: The goal should challenge you slightly but still be reachable with consideration to the skills and abilities of you and your team.
• Relevant: The goal is important to you, your team and your company.
• Time-bound: A timeframe for completion must be imposed.

Done well, it takes a fuzzy and generic goal into a more meaningful goal to ensure greater chance of success.

Source: Northpass 

The folks at i-EM (Intelligence in Energy Management) had a clever way of rephrasing the SMART principles into a collection of questions. They added a few more elements to the SMART framework, yet essentially capturing the essence of the goal setting process.

Imagine you’re a power plant manager. You want to measure your plant’s performance. To start, you ask these questions: 

1) What do I want to accomplish? I want to improve my plant’s performance

2) For what reasons? So it can produce more and make more money

You will then go deeper by asking follow-up questions to further refine your goals.

Source: Intelligence in Energy Management

Just like there are many ways to skin a cat, there are also different methods you can take on the time-tested SMART principle. Irrespective of the approaches, their ultimate objective is to help you define your goals in greater clarity for everyone involved.

Get set to OKRs

The last system that I would like to introduce today that might help define your data journey goals and the results you want to achieve is called OKRs or Objectives and Key Results. The framework is popular among tech companies since it was invented at Intel and later popularized at Google.

In essence, OKRs guide you to set the Objectives or goals of the data project and the Key Results outlines how you aim to achieve the stated goals. 

Using this system, you are to choose the most meaningful goal - the Objective - associated with a data project that you want to achieve. You will then determine a handful of Key Results or activities that will contribute towards the goal. Key Results are typically aggressive, measurable and usually limited to 3-5 for each Objective.

Jan Zawadzki, a Data Scientist at Volkswagen Group gives a good example in applying OKRs to a real life example in the automotive industry. A team wants to develop a driver-assistance function that alerts truck drivers to the presence of pedestrians in urban areas. They agree that a 98% detection rate is a proper stretch-goal for the first quarter of the project, supported by several Key Results.

Source: Towards Data Science 

John Doer, the legendary investor, is one of the biggest proponents of this method. He even wrote a book about it called Measure What Matters: How Google, Bono and the Gates Foundation Rock the World with OKRs. Companies like Google, Amazon and Spotify adopted OKRs in their goal-settings process.

OKR may seem simple, yet it is powerful enough to ensure you’re taking the data project in the right direction.

Get on With It

It can be overwhelming if you are just starting on your data journey, however big or small. There are a myriad of databases, tools and methodologies you can choose from. Do we use Python or R? Which database is more optimal - Hadoop, Oracle or MySQL?  

Honestly speaking, these are the least important considerations at the beginning of any data project. It is more crucial to define and set the overarching goals on the data project, that fulfill your business objective.

Pick a method that I described above (or any other methods you’re comfortable with) and just go for it. Start small and focus on solving a problem with immediate value and work your way up to answer more complex questions or goals. Remember, however small your initial scope is, ensure there is a goal for the data project, lest you will be running like a headless chicken among a sea of datasets.

Be great,

Reez Nordin

If you like what you read, do share this link with folks in your network for them to sign up at

https://sendfox.com/spicyvc