How you can set up a data-driven and scientific process to make better and faster business decisions
In today’s business landscape, making smart decisions is both harder, and more important than ever. You’re constantly bombarded with an avalanche of data from seemingly every angle, and gut decision-making can only take you so far. This is where data-driven decision-making can change your business – and all it takes is a few changes and some good old sweat!
Today I’ll be going over a few small steps you can take – without hiring a dedicated data scientist – to improve your business decision-making and get the results you want.
This might sound like a no-brainer to anyone who’s ever had a strategy meeting (all of us), but it’s mission critical to start your projects with a goal, and most importantly, a metric to benchmark and measure that goal.
This means you need not just a goal, but a SMART goal. These follow five very simple guidelines:
Specific - Your goals must be as precise as possible. Although this sounds easy, we all very frequently fall into the trap of making overly broad goals. How many times have you heard your boss say (or if you are the boss, how many times have you said) something along the lines of "we need to increase revenue!". How can we improve upon this vague and broad goal? By adding important details like the intended target, or specific part of the business where you want revenues to grow. This goal is more specific: "We need to increase Ontario revenue from our flatbed business by at least 10%"
Measurable - Your goals need to be specific, but also quantifiable. This means that you must be able to measure it with numbers. Simple enough for goals relating to revenue, profit, or volume - but what if the goal is something a bit more abstract? In that case, you'll need to find a cause-and-effect relationship that links it to a measurable outcome. For example, if you wanted to "increase employee productivity", you could would need to find a metric that accurately reflects "productivity". Careful though, as this is one of the assumptions that you'll have to keep in mind when you're evaluating the results of your experiments.
Attainable - Nothing feels worse than setting an impossible goal. That's why the best ones are challenging - hard enough to motivate managers and employees alike to perform better - but not impossible, which only serves to demoralize your team. This one's pretty self-explanatory, but it does require a lot of qualitative, intuition-based understanding of your team and organization.
Relevant - Although this one again seems like common sense, your goal must be relevant to the overall direction and success of the organization and team. Why? Because people don't like working hard for things they perceive to be pointless!
Time-Based - Lastly, is probably the most important aspect of a SMART goal. It must have a deadline. And not just a "we'll evaluate it after a few weeks" deadline, a firm date and timeline for when your team will evaluate your success in reaching this goal. With this, it's important to teach your team that goals aren't about winning and losing - they're about learning. At FreightPath, we'd much rather fail to reach a goal but uncover new insights about our business, then succeed without understanding why. Because when you understand why you succeeded or failed, you can begin to control future outcomes more and more
By investing the time and effort into creating SMART goals for your team or organization, you will set yourself up to be responsible for your own success, with a framework for success and benchmarks so you can see what is and isn’t working.
You can’t implement a data strategy if you don’t know where your data is, and in what condition it’s in. After you've set your goals, make sure you take time to understand where your data is stored in your organization, how it’s organized, and what tools you need to work with it.
For example, imagine that you’re looking to decrease freight spend by 10% this month – where is all your transportation data? How accessible is this data? Is it stored in a centralized cloud location? On paper? On excel spreadsheets? Figure it out so you can begin a plan to clean it up.
If you don't understand how to work with your transportation data, or are confused about how to centralize it, feel free to contact us - we're here to help you.
This is one of the most under-appreciated aspects of running a data-driven business. Oftentimes, the trends you need to see and analyze can be understood instantaneously with a good graph, oftentimes with very little statistics work required.
As a decision-maker, sometimes it’s more important to recognize the overall trend than specific numbers – a graph can tell you instantly and intuitively that sales dropped exponentially, in a way that a table of numbers just can’t. This doesn’t have to be expensive or complicated either – FreightPath has great built in analytics support for a variety of logistics KPIs, for example, and Excel has a great built-in graphing platform if your data is still in CSV’s.
Like any science, good data science relies on the scientific process – making a hypothesis, testing a small change, and observing the results. This is key! Just because you notice a trend, you shouldn’t jump to change your entire strategy overnight.
Decide based on your data, insights, and trends a single change you can make in your strategy to impact your benchmarks or KPIs (Key Performance Indicators), then follow through and look at the results! Be honest with yourself – did the change work? How much did it change? What could you have done better? After doing this, come up with a better change; rinse and repeat. This is the essence of good data-driven decisions – test, observe, evaluate, repeat.
With these small tips in place, your organization should be well on its way towards becoming more data-oriented and making better-informed decisions. Remember, the first step is always the hardest one, so don’t be afraid to make mistakes!