data science budget planning represented by a calculator and a piggy bank with glasses on.

Planning Your 2018 Data Science Budget? Here Are Some Logical Projects

It’s that time of year again!  With budgeting for 2018 at the forefront of your agenda, you may be wondering where to head with your data science efforts.  Investments in big data are often expensive, but when planned correctly you can manage costs and also ensure that your new systems and employees generate data-driven profits.  We at Dataspace wanted to share a few things our clients are pursuing or considering to improve their data management teams and systems to bring big insights to their 2018 strategy.

Our list is arranged from least to greatest according to the maturity of your data science environment. So, regardless of today’s capabilities, there’s a next step that you can take.

We use Excel, but it’s starting to feel inadequate

Spreadsheets were one of the earliest BI tools. So, for those of you who think you’ve been bypassed by the BI revolution, think again. However, if you don’t yet have a more modern BI tool, you’re probably working with messy, disparate spreadsheets that require manual data entry.  Maybe you’re frustrated by the complexity, the inability to integrate the various streams of data collection, or the lack of centralized control over who sees what and who can change the analyses.

The first step to amending this situation is to adopt a modern, BI tool such as Qlik Sense, IBM Cognos or Tableau which can link to the detail data in source systems and join it to data from the same Excel spreadsheets you currently use.  You will gain an integrated view for senior management and the ability for business users to correlate pieces of data from across the business. And, with their cloud offerings, it’s easier and cheaper to get into these kinds of tools than ever before.

We are currently working with a client that performs data analysis by running SQL queries and then manually porting the data into Excel spreadsheets.  The organization’s new CEO is frustrated by the unintegrated nature of the analyses she receives. She has no way to compare trends across her different business units or to drill into those trends. Other executives are frustrated by the amount of time it takes to create and refresh analyses and also by the static nature of the visualizations.  We are helping them implement Qlik dashboards to remedy these pain points.

We have a modern BI tool, but our data systems need help

Already have a BI tool in place?  The next step is thinking about the data itself and the source systems from which it originates.  It is time to consider building a data warehouse where you can integrate all of your data and make it easy for all employees to access and use tor reporting.

Another one of our clients adopted Qlik Sense about six months ago. They’ve started rolling out applications and are seeing some significant successes. Now they’re taking the next step – building a data warehouse to allow easier access to more clean, reliable data.

Need help taking this step? Dataspace is experienced in helping clients plan and take the steps to build data warehouses.  

We Have a Data Warehouse

So you’ve built and know how to manage your data warehouse.  From this stage, we recommend two moves.

Step 1: Build your data science capabilities – develop or hire resources who know how to interface with the business, ask the critical “what if” questions and then use their technical skills to integrate data and reveal insights.

Step 2:  Provide these individuals with data.  This data is often unstructured, in the form of web logs, textual data, images, voice recordings or what have you.  This often entails moving to new data storage technologies such as Hadoop.  Hadoop is highly scalable and flexible in the sense that you can run it across many servers and process huge amounts of both structured and unstructured data.

We have Hadoop or other Big Data technologies

If you are already succeeding with Hadoop or different data storage and analysis platforms, the next step is to move towards using true predictive analytics.

Push your data science staff to adopt R, SAS, Python, Tensorflow and other tools, languages, and methodologies.  Artificial intelligence, predictive analytics, and machine learning will allow you to develop applications like product recommendation engines and pricing optimization engines, that rely on advanced analytic techniques to inform future decisions. Further, think outside of your organization and consider applications tailored to your strategy goals and the needs of your customers.

We’ve started working with predictive analytics

If you are feeling good about your current data strategy, congratulations!  You are ahead of the curve so keep up the good work.  Please also keep in mind that Dataspace is able to assist with providing temporary resources for any projects you have requiring extra hands!  

Are you budgeting to make your next big step in the big data world? Do you already know your direction but need top-quality, expertly-vetted staff to execute your plan?

Either way, we can help! For more information, contact us.

Scroll to Top