AWS Smart Business Blog

Reinventing Small and Medium Businesses with Better Data Insights

If you had the ability to synthesize and analyze all of your business data, what would it allow you to achieve? Between business apps, social media, and internet-connected devices, you likely have enough data inputs to fuel your small or medium business’s (SMBs) decisions. One common misconception is that wide-ranging data analysis is only possible at digital-native startups or enterprise companies, but smart businesses of any size can reinvent themselves in numerous ways. Here are a few examples we see across numerous industries:

  • Analyzing data to see what works and what doesn’t for business investments
  • Predicting industry and economic trends that will have an impact on business growth
  • Improving customer retention by analyzing buyers’ preferences and behaviors
  • Tracking performance metrics like results from sales and marketing campaigns that can assist in more effective business promotion

The ability of an SMB to analyze and use data-driven insights can set them apart from their competitors. An article on points out 93 percent of all data is “dark data,” which is defined as “information assets that companies collect, process or store but fail to put to use.” At Amazon Web Services, we hear from customers that barriers to entry for data analytics are far-ranging from difficulty in hiring analytics experts to knowing where to start with creating a data pipeline. In a tight labor market, customers know its growing importance but often find it too daunting. Fortunately, it doesn’t need to be that way: especially if SMBs don’t consider themselves data experts.

Learning data analytics starts with understanding these common data formats

You likely have access to data from numerous sources—such as software applications, business processes, and third-party markets. It comes in all formats, including:

  • Structured: Data stored in a predefined format, such as spreadsheets
  • Unstructured: Various types of data stored in their native formats, such as text documents
  • Referential: Data used to classify or categorize other data, often third-party data that is static or slowly changing over time, such as public demographic data
  • Historical: Data about past events and circumstances, such as log files and financial reports
  • Transactional: Financial information captured from e-commerce or in-store transactions

Transforming this data to make it analytics-ready is challenging. As we mentioned earlier, customers tell us they have trouble hiring the right people to analyze the data they’re collecting. In addition, new hires often lack the industry knowledge and invaluable context required to gain data insights. For SMBs, a cost-effective, easy, and scalable data transformation solution is the key to making use of this information. With business outcomes in mind, SMBs can narrow down the field of focus and make use of the data that is important for their organization.

There are several options available to store and organize a company’s data, and a data lake is one of the most sought-after and functional options.

What is a data lake and what are specific use cases for SMBs?

The concept of a data lake is to create a central source where companies can store their data regardless of format. If your business collects any of the five types of data formats shared above it, it can be traditionally difficult to integrate into one view. Even though a data lake is not a new concept, companies are still trying to figure out when it is right for their organization. Data lakes can benefit SMBs in these three ways:

  • Ability to analyze data more effectively: Build custom databases by combining data from all of an organization’s different sources to create richer datasets
  • Diverse and wide data availability: Data can be combined to create unified datasets for use by different teams
  • Improved data efficiency and agility: Data can be stored in native format, which reduces the processing time to move and transform the data

New to digitization or looking to add more cloud capabilities to your SMB? Explore solutions by industry, benefit, use case, and more on AWS Smart Business

How the cloud solves these data analytics challenges

With the goal of tapping into the unrealized potential that these data challenges present, AWS developed a suite of services that allow customers to get started immediately with low- and no-code solutions, unlocking the power of analytics to any size firm with any level of technical expertise. This means you do not need to be a data expert to understand your numerous data sources.

A common use case that we see with our customers is putting together data from a Customer Relationship Management (CRM) system and revenue from an accounting system. Let’s explore a sample architecture that shows exactly how AWS lowers the barrier to entry for creating and using a data lake solution.

Diagram depicting the AWS solutions recommended for data insights

At the heart of a modern data architecture is Amazon Simple Storage Service (Amazon S3), an object storage service that acts as a central repository for all types of data. Using Amazon S3 as a data lake acts as a foundation for SMBs by unlocking a multitude of AWS services that integrate with it. First, export the data from your CRM and financial systems, then place both files into an Amazon S3 bucket. It doesn’t matter the format—the data will be prepared and cleaned in the next step.

Data residing in Amazon S3 opens the door for an uncomplicated, straightforward approach to analytical insights. Data in Amazon S3 can be any combination unstructured and structured data.

The next step is to combine and transform that data into a format that can be queried with basic SQL statements. This can be the first major challenge for a resource-restricted SMB. Fortunately, AWS Glue Databrew offers a visual interface to clean and convert the data into readable formats that can then be used by other AWS services. Users can combine customer and financial data, remove sensitive data points, convert the two files into the same file format, and even remove duplicates and invalid data all without writing any code.

Screenshot of AWS Glue DataBrew user interface showing low- and no- code solutions

Figure 1: AWS Glue DataBrew in action – a visual way to edit, transform, and load data

Once the data is ready, a Glue Crawler will automatically discover the data structures and place that metadata into an AWS Glue Data Catalog. With just these three steps, the previously unrelated files in an Amazon S3 Bucket have been transformed into a data lake that is ready to be queried without any specialized skillsets or complicated infrastructure concerns. You do not need data scientists to use this tool.

Amazon Athena then allows you to query that transformed data in place, meaning there is no need to create and maintain a data warehouse to take advantage of the new datasets. In other words, it allows you to join the data from both the CRM and financial files in a single database. With Amazon Athena, there is no upfront cost, maintenance, or need for complicated data extraction jobs, and it will scale alongside your business needs. Instead, you only pay for the queries that you run.

These queries can be immediately useful, providing returns and insights on collected data. However, decision makers often need a more visual approach to the data. With Amazon QuickSight, line of business employees can deliver actionable insights in an easily interpreted fashion, even allowing users to ask natural language questions to dive deep without any technical knowledge. Instead of writing code, a user can simply ask, “Which customer segment grew the most year-over-year?” and have insights delivered in near real-time with a dashboard like the one below.

Amazon QuickSign dashboard depicting visual diagrams of sales data in pie chart, line graph, and other formats

Figure 2: Amazon QuickSight provides suggested questions written in plain English with no coding required

Next steps

The great advantage of this setup is putting the keys to analytical insight in the hands of the people who know it best. They can combine data-driven insights with their expertise in decision making that results in a clear competitive advantage. With a fully serverless approach, a low- or no-code implementation, and the ability to provide beautiful visualizations, businesses of all sizes can begin their data journey and transform its collection into valuable analytical insights.

These services are available today and ready for businesses of all sizes to start using. If you want to make your business intelligence tools work harder for you, learn how you can gain more insights or contact an SMB cloud expert.

Dimple Dhar

Dimple Dhar

Dimple Dhar is a Sr. Solutions Architect with over 20 years of architecture, design, and development experience in multiple industries. She supports SMB customers at AWS with her strong engineering and analytical skills, as well as her ability to understand complex business processes. Dimple holds a degree from University of Jammu and is located in California (US).

Phil Knapp

Phil Knapp

Phil Knapp is a Sr. Solutions Architect who uses his experience as a client manager and his history in the IT and services industry to support SMB customers at AWS. He is skilled in account management, performance improvement, CRM, and cloud technologies. He is located in Illinois (US).