If you don’t consider yourself a “numbers” person, the word “data” can be intimidating. But if your decision making process isn’t based on numbers, then you’re making them on something significantly more dangerous: assumptions.
Unfortunately, assumptions are how most companies operate. According to a survey by social listening tool Mention, less than 15% of businesses have a data-driven culture. Just 17% of those surveyed said they have a high degree of data literacy, meaning that they feel comfortable reading, creating, and communicating data as information.
How to Support Data Driven Decisions
The good news? Those gaps are opportunities for you to get a leg up by building data into more of your business’s operations. Here’s how to get started on supporting data driven decisions:
1. Inventory the Types of Data You’re Collecting
Daily activities and interactions with customers generate lots of data. If you don’t know what’s already available to you, you can’t make use of it.
Some sources are obvious: If your store uses a Square POS system, you’re collecting names, types of credit cards, time of purchase, and more. Other data sources are less obvious: If you place Facebook ads, you can look deeper than conversion rates. Who’s clicking the ads, from where, and on what devices?
A quick look at the Square dataset might reveal that the majority of your customers are repeat customers. This could inspire you to start a loyalty program to reward your regulars. With the Facebook data, a type of post might rise to the top and lead you to try placing a few ads with the same post type.
This is just scratching the surface of the data you’re likely already collecting. Think about the possibilities if you were intentionally collecting other data.
2. Stay Focused
Once you consider the amount of data you’re already collecting, it’s easy to get bogged down or distracted by all the metrics available. Consider your business goals, then think through which figures you actually need to monitor.
Say you operate a coffee shop. Although ingredient costs are important, they aren’t relevant to the question of whether you should open a drive-thru-only location. What’s the average amount of time your team members spend serving a customer at the window compared to the counter? Which service channel has a higher average order amount?
Once you have a goal in mind and are collecting data, the next step is simple: making time to review.
3. Block Off Time for Review
Without review and analysis, data are just numbers that do not lead to change. ETL — short for “extract, transform, load” allows you to put those numbers into a program that will show a story.
Set aside time on your schedule once a week to review the latest changes to the metrics you’re monitoring.
Different data sets require different analysis and visualization tools. A word cloud might be appropriate for checking for trends in customers’ comments on your site. A regression analysis is more appropriate if you’re trying to tease out a correlation between two numerical variables.
4. Look at the Big Picture
Analyses can be conducted on multiple levels. Reviewing findings from multiple data sets is critical if you want to see the big picture.
Say you want to know which types of customers are most profitable. Well, you can’t just think about which ones pay you the most money. How much do those types of customers cost to service? What’s their average lifetime value?
Answering your core question requires multivariate analysis, which can get tricky. Particularly important to parse out is which variables depend on others: In the previous example, does customer lifetime value correlate negatively with per-session spending? When in doubt, ask for help.
5. Give Your Team the Keys
Once you’re collecting and analyzing data, there’s no reason to keep it from your team. Much as you might like to, you simply can’t make every decision for your company.
Invest in training. Your staff needs to know how to access your database, interpret the data, and generate reports.
Think, too, about communication. Create a common set of terms. Bring everyone up to speed about why you’re placing new emphasis on data analysis.
Finally, prioritize collaboration. Encourage team members to bring unexpected findings to your attention. Reward them for bringing data-inspired ideas — such as a new product or an untapped target market — to your attention.
6. Start Requiring Data for Decisions
The biggest challenge in becoming more data driven is cultural: When you need to answer a business question, everyone must defer to the data and make data driven decisions.
Data obsession is one secret to Amazon’s success. The ecommerce giant keeps tabs on 500 KPIs so that it always has the information it needs to make a decision. Many of Amazon’s initiatives start by spotting trends between them, such as the correlation between slower page load times and decreased visitor activity.
Develop a plan around how, exactly, you’ll consult the data. Set parameters for the quantity of data required and over what timeframe samples should be taken. If you’re a restaurant that wants to simplify its menu, for example, you can’t assume that what’s ordered for dinner on Thursdays is representative of the entire week.
Becoming a data-driven company is hard. But ask leaders at larger organizations, and they’ll tell you: It’s much easier to do when you’re small than once you’ve scaled.