A Practical Guide to Data-Driven Decisions for Businesses

If you own or work for a small business, you might be feeling the pressure to become a “data-centered” business and to make “data-driven” decisions. If you’re a bit foggy on what that means for you or your business in practice, you aren’t alone. Many small businesses understand that making smart use of the insights gained from digital metrics can help them run their business more successfully, but the majority have no idea how to create such a process. In this article, we will lay down some basics of what it means to be “data-driven” in a small business context and provide some practical first steps to creating your own data-driven process.

Data-Driven Decisions

First off, what does it even mean to be data-driven? When referring to a decision as data-driven, it means that it is being made based off of data as opposed to a gut feeling or experience. It might feel like giving up control of the decision, but data-based algorithms consistently outperform human intuition. In fact, the process for data-driven decision making isn’t so different from the way our brains make decisions. We make decisions for the future by understanding and analyzing what has happened in the past. Data is simply a record of past events that holds incredibly complex and minute detail, so that when we analyze it we can more accurately understand what is going on and therefore make better decisions.

Data-driven decision making should make your strategies more accurate and therefore successful, but there are some misconceptions about the advantages of using data to make decisions. Using data won’t necessarily make decision making faster, easier, cheaper, or more straightforward — that will still boil down to the efficiency of your process, and you might find that it is more expensive to attain the data you need to make a good decision. However, the long-term savings from making a better decision, and making it the first time without the same trial and error, can mean that you win out in the long run. and When starting out it is important to view the whole process as a grand experiment, and not to scrap the data-driven technique because of a failure. Data is simply a record, it is neither wrong nor right – but it should come as no surprise that interpreting data is a skill and will only be improved over time and with some errors.

But the real question remains – what does it look like to implement a data-driven decision-making process in real life? Unfortunately, that answer varies as widely as the types of decisions you are trying to make, but below you will find a basic process outline to get you started, and some resources to help you along the way.

A Practical Process

1. A Starting Place

Most small business owners are adept at making business decisions, but every business can locate areas that need improvement. When you are working all hours of the day to run your business, it can be overwhelming to incorporate data into every decision you make – so choose one place where you need help. Are you not getting as many online sales as you need? Are you having trouble hiring people who fit your company and stick around? Do you suspect that a product or service you offer isn’t everything your customers want? Data can be applied to any situation, and can help bring order out of what feels like chaos. Pinpoint one area of improvement to work with first, collect any information you have that shows the current state of affairs, and go from there. When starting out, try to start small. If the data that would help you is complex, very large, or would involve analyzation skills you don’t have, then go back and choose a simpler project to start with – or it might be a time to search out a specialist.

If you aren’t sure of where to start, take a look through this article of data-driven tools.

2. Record

One of the biggest benefits to making data-driven decisions is that you can trace exactly which factors led to your success or shortcomings. To do that, you have to record your process – the more thorough you are, the better you will be able to track progress. Include the problem you were trying to solve and any information you have about the state of things before you make any changes – this information is your baseline and it allows you to determine if any changes are successful. In all things, the more information the better, you just need to figure out how to organize the information from your process in a way that is most useful to you and your team.

Helpful Tools:

  • Google Docs or Sheets
  • Excel, Evernote
  • Data visualization services will also record and track your progress over time.

3. The Information You Need

Once you have located an area where data might help, you need to figure out what information will help you make a better decision, and where to find it. You might be able to collect some data internally, while other types of data might only be available through a tool or service. If you are looking at information from your website, you might reach out to the company that handles your site. If you want answers about hiring, maybe a tool with more resources would fit your needs. Maybe the data would come from a survey you create in-house.

Helpful Tools:

  • Types internally generated data (surveys, reviews, customer satisfaction ratings, product returns)
  • Tools for finding Web Data (Google Analytics, Google Adwords, Klipfolio, Basecamp, Webtrends)
  • Other Services (Market research consulting companies, syndicated research white papers or raw data)

4. Analyzing Your Data

This is the part that gets most people spooked, but it really shouldn’t – even if math was never your thing. Most of the math involved in basic analysis is simple: finding averages, means, percentages, mapping trends – all things we did in middle school. Data of a qualitative nature (meaning data that is not in number form – think written reviews or survey responses) might not even require math to be analyzed at all. Instead of thinking you can’t do it, try writing it out in terms of what information you have and what answers you need, then determine if you can get your answers from what you have. That being said, some businesses, especially as they grow, will find that they are dealing with large and complex data sets that can’t be analyzed easily using traditional methods (these unruly data sets are what we refer to as Big Data). When you can no longer gain valuable insight from your data because it’s become too complex, it might be time to consider hiring a data analyst.

Helpful Data Visualization Tools:

5. Measure Success or Failure

Once you have analyzed your data and made some changes based on the data you collected and analyzed, figure out which affects of your decision will best measure success. If you have baseline data to measure success against then do that, but if you had no pre-recorded data, then you should use this opportunity to set a baseline and record it, so that future success can be measured. At this point you can also decide which factors to continue to track, which will give you continuous feedback about that section of your business. It is also a great time to weigh the time and effort it took to implement the data-driven process against the gains and determine if it was in fact worth it. If a process generates too little data, it might not be cost effective to implement a data-driven process yet.

Data visualization tools allow you to set parameters for success or failure, so you can always measure your progress.

Learning to use data to your advantage will result in greater success than relying on intuition alone. There is no doubt that the process can be intimidating because of the overwhelming number of possibilities, but there are tools to help and many tools are free or built into services you are already using. Google analytics, tracking software, surveys, product reviews, and many others are all examples of data-generating tools easily available to every business, no matter how small. A data-driven small business probably will not need to process large databases of information or hire a specialist, as most likely they will never deal with Big Data, but that certainly does not bar them from using data to their advantage. At the heart of the whole process is an attitude or company culture that values and seeks out a way to make decisions that is efficient, repeatable, and measurable, and gets excited about the possibilities that data offers.