Data analysis is a part of the Business Intelligence. It can be used for many purposes: confirmation of a hypothesis, forecasting, or discovering new features in data. We are all surrounded by information and being able to see the bigger picture can help us make an informed decision. In the past, many business decisions used to be made based on a hunch, because no one actually could get data from thousands of people at once. Now, with the internet, you don’t even have to make surveys to get your market research. The information is out there and all you need to do is make an analysis and draw the conclusions.

 

There are a couple most popular data analysis techniques: A/B testing, Data Mining and Clustering. All of these techniques can be used in various ways, giving a lot of possibilities to improve your business.

The A/B testing means you split your targeted group into two subgroups, and separately test the results of your actions on both of them. As a result, you know which action gave you higher conversion rate. For example, you have a website and you would like to check if your newsletter sign-in button should be red or green. You can very easily test it by showing a website with the red button to one group of potential clients, and the green one to another. At the end, you know how many visitors signed in, and you can determine whether the green or the red button gave you better conversion rate (more subscribers). You can do the same with email campaigns. Just split the recipients into two groups and prepare different contents of our email, then analyse how many people responded to each type of email.

Nowadays, you can find many tools enabling you to make A/B testing. The popular ones are Visual Website Optimizer for changes in your website design, and for A/B email campaigns you can use MailChimp.

The start up co-founder Dan McGrady described his case study with data analysis, which resulted in a 72% increase in conversion rate.

Data mining can be described by many methods, like association (market basket analysis), classification or clustering. The market basket analysis is like its name suggests: the analysis of a shopping lists of clients in order to find products that are most commonly purchased together. Such analysis can help you arrange your shopping shelves, so that the majority of customers will find the products they buy together and be more satisfied with your service, since you have made their buying process faster. You, the business owner, can also put some additional products in between these bought together, increasing the probability that customers will buy them.

The same analysis can be used for both real and online shops. For the real shops, managers can reorder the products on the shelves. For the online shops, owners may implement a good recommendation system.

Clustering is also a very relevant and popular method. It simply splits your clients or targeted audience into groups (clusters), based on their similarity. Imagine Twitter users that are clustered based on the topics they discuss. This can help you reach the desired leads by monitoring proper tags and engaging with top influencers within the topic’s cluster. Moreover, you can monitor your competition, who is in the same cluster as you are. One new Twitter tool that does this for you, and helps you infer conclusions, is called Twittie.

If you cluster your clients based on the profit they bring to your company then, knowing the clusters, you may analyse deeper the purchasing behaviour and demographics of each group, and prepare marketing campaigns suited to their needs.

The techniques I described are just a few out of many and they all can very easily increase your revenue and help you make the right marketing and strategic decisions. Therefore, don’t waste your time with a hunch. Start analysing with the tools available, and grow your company in the right direction.