Big Data for Small Business

Suzanne Burns
By:
On: 17 Aug 2017
Share this post

big data small business

Suzanne Burns is a Data Scientist at Three. With a background in both small start-ups and large enterprise, she knows how to make big data work for both.

The so-called “big data wave” is rolling, and if you run a small business you’re no doubt wondering whether you should dip a toe in. You may be reluctant because you think the “big” implies it’s not for small businesses, or you think it will demand big investment and time resources you don’t have. Let’s take a look at the advantages and challenges of big data for small business to help you decide what to do next.

Big Data is Within Reach

Gathering or accessing data doesn’t have to be costly or resource-hungry anymore, and the insights it can offer are equally valuable whatever the size of your business. In fact, in terms of getting hold of data, it’s never been easier or more cost-effective to do so – putting it easily within reach of any business.

The bigger difficulty is not getting hold of the data but getting a grip on it. There’s so much of it available that if you’re not careful, it can prove as useless as no data at all. The key is to know what data you want, how to interpret it, and then how to use it to drive decision-making.

Where to Find Your Data

Data has never been more accessible; Google, YouTube and Facebook for example, provide free data analytics on your web and social media content. Their reports can provide rich insights into your audience and how they are interacting with your business online.

To reach more meaningful conclusions from your data, you should compare the results from your own audience against the wider population. For example, a value of 60% of your target group owning an Apple product loses significance when you see that 60% of the population own an Apple product. The key is to spot the differences and take action to change non-buyers into buyers.

Taking it up a notch, the rise in popularity of Application Programming Interfaces (APIs) among companies like Salesforce and Twitter has made it possible to access third-party data. APIs allow third-party software to connect and communicate to your own, making a world of data available to you. To avoid being swamped by information, you will need the data cleaned and in a readable and usable format. In the past this would require a lot of hand-written SQL code, however user-friendly Extract-Transform and Load (ETL) tools such as Pentaho Data Integration (PDI) will now do that for you.

What Should You Do with Your Data?

To obtain real value from your data, you first need to know what questions you want answered.

First decide the key performance indicators and metrics that are relevant to your objectives. With these as your yardstick, you’ll be able to assess from your data whether your actions are having the desired effects.

Secondly, you have to know how to visualise and model your data in a clear and unbiased manner to answer your questions. It’s easy with software such as Tableau that connects to tables in your database and offers a quick and easy way to build bar-charts, tables, time-series graphs, etc. Once the views are built, your employees can access, read and drill down into the numbers using Tableau Reader.

Using Data to Predict the Future

On top of descriptive analytics, you can build prediction models to help forecast the future. A prediction model will calculate the probability of your customers taking a future action. The type of prediction models to apply and test range in complexity from well-known Regression methods to advanced Machine-Learning methods.

When building predictive models it is a good approach to gain descriptive insights first before you launch into applying a statistical model. These insights will help guide you in what to keep or remove as inputs into the model. That said, there are now tools like RapidMiner and Microsoft AI that have simplified statistical modelling. The results of these insights and model outputs can give a business the ability to make better informed decisions – decisions that will have a positive and cost-effective effect on your business performance.

Big data is not the preserve of big businesses. By getting the basics right, it can make a big difference to small businesses too.