Effective Strategies for SMBs to Start with Data Science

By David Sweenor, Senior Director of Product Marketing at Alteryx

Businesses are increasingly becoming aware of the advantages that data can bring to their strategies. Conversely it is becoming more and more clear that, without access to good quality data and insights, these same businesses are operating at a disadvantage – not only to large online giants, but also to their own direct competition. While many small business owners might think that data science is out of their reach, and that analysing data is supremely difficult without a data scientist, the reality is that – today – anyone can become a citizen data scientist. They just need the right tools and motivations.

Accessing data science driven insights is something that was once a pipe dream for small business owners due to not only the significant salary of data scientists, but also the availability, accessibility, and quality of data. In our digital-first society, the data we create on a day-to-day basis increases exponentially. SMBs are now at a new juncture – one with access to hugely valuable data repositories, and also the option to turn that data into actionable insights.

In short, this means that businesses are now able to make better informed decisions to stay competitive. In retail, data is being used to help make decisions like deferring planned price cuts on winter clothes ahead of a cold snap or diverting stock to stores during product shortages. In finance departments, it’s being used to turn static invoices, shipping, and billing data into actionable insights.

With this in mind, there are four key strategies which are essential for any organisation getting started with data science – areas that can easily scale up or down regardless of the size of the business or the size of the challenge:

1. Check what data and tools are available, and decide how you want to use them

All businesses – in one way or another – have data that can be used for insights that can significantly impact business decisions, whether that’s through Point-of-Sale data, or even just measuring the volume of inbound emails from customers. It’s likely that most businesses are already using some form of analytics, too… even something as simple as a spreadsheet.

Some believe that in order to pull useful insights, you need huge amounts of data; however, many projects use small datasets to deliver huge value. The key is the quality of the data – not the quantity. As the insights needed from data becomes more complex, more processes and more user-friendly tools can be added when a specific need arises.

2. Solve the small irritations

Any business leader looking at data science will have a problem at the forefront of their mind – a challenge to be met. Otherwise, it’s highly unlikely they would be exploring data science at all. on an analytics journey will undoubtedly have a problem in mind to solve. Just as the automatic telephone exchange was invented due to the irritation felt at misrouted calls, so too must your business begin the process of change by asking: “what irritates us most?”.

This can be something as simple as adding emails to an analytic process to identify which addresses are most likely to be spam, and then blocking those domains. It could even be as simple as checking sales data year on year and using that to inform staffing levels. The key here is to start small and work up to larger challenges.

3. Implement, expand, and replicate

Once the preparation has been completed, and processes have been put in place, businesses can begin to democratise the access to that data and begin transforming it into a business insight. Replicating this process and training those closest to the problem to provide quick answers to questions is one of the most significant benefits of implementing a data-driven strategy.

With a firm foundation based on assessment, preparation and small successes, we can begin to replicate and expand the work that is done with data into more complex analyses through advanced data techniques such as predictive or prescriptive analytics.

While the road to a fully realised data and analytics strategy is a long one with numerous potholes, exits and turns, the end goal is a far more effective way to make effective and valuable business decisions. Even the simplest of irritations – once fixed – can deliver a disproportionate benefit.