Email
Password
Remember meForgot password?
    Log in with Twitter

article imageQ&A: CIOs looking to save money turn to SaaS analytics Special

By Tim Sandle     Apr 4, 2020 in Business
With such a high cost to implement, many CIOs and IT teams make the mistake of “setting and forgetting” their analytics programs, says Craig Kelly, VP of Analytics at Syntax, who looks at the risks.
The costs to implement and deploy a successful data analytics program can often reach upwards of $1 million between an estimated $200,000 in infrastructure and software implementation fees plus the average $800,000 needed to support the developers monitoring the system.
By “setting and forgetting” their analytics programs, CSIOs could unintentionally render an entire program useless if it becomes buggy or you miss a required update from the provider. To minimize these risks, Craig Kelly, VP of Analytics at Syntax, a managed cloud service provider, recommends businesses consider SaaS analytics applications rather than the traditional analytics software, because they’re programmed to work autonomously to download updates or conduct ongoing system maintenance/testing for you. Kelly explains more to Digital Journal.
Digital Journal: What is the value of data analytics for business?
Craig Kelly: The ultimate value of data analytics is to provide actionable intelligence to discussion-makers and, ideally, throughout the whole enterprise.
DJ: What are the costs involved in setting up data analytics?
Kelly: The Total Cost of Ownership (TCO) of data analytics includes the storage of all required software (physical servers, VMs, cloud servers or hosted serverless architecture), licensing for the applications, including database licensing, ETL (Extract – Transform – Load) applications, and front-end dashboarding and reporting software. There is also the cost of outside consultants to consider, as well as the costs required to cover the internal resources required for the system’s implementation.
Upon successful deployment of your data analytics program, costs will include maintenance or subscription fees for the applications involved, possible upgrades in the future (whether or not these are necessary will depend on your system’s architecture), as well as the ongoing maintenance and administration of the applications. Analytic usage will continue to expand typically, requiring extensions to the application.
DJ: Why does data analytics sometimes fail to deliver?
Kelly:Analytics can fall short for a number of reasons:
The applications perform poorly or provide data that lacks integrity. If users log into a dashboard that takes longer than a few seconds to update as they navigate the pages, they can quickly become agitated and stop using the system. Similarly, nothing will kill user adoption faster than poor data integrity. Users need to trust that what they are looking for is accurate.
Incomplete Data. Often, the KPIs displayed only tap into a subset of available data sources, resulting in an incomplete picture. For example, sales analytics built on ERP data can seem great, but it won’t provide data that answers how sales are performing without relevant CRM data. Good analytics implementations typically tap into multiple sources.
Metrics that aren’t actionable. If the data doesn’t tie to what’s really important to the business, users won’t have the incentive to use it. Similar to rewards that social media users seek when frequently checking their streams, analytics users need to have data that is engaging and related to their goals; otherwise, there is little value.
Lack of forward-looking analytics. It is great to see trend lines of the business’ performance and compare this against your stated goals and forecasts. But what really makes analytics a game-changer is forward-looking, predictive analytics. With constant innovations in Machine Learning (ML) and Artificial Intelligence (AI), this is currently gaining a lot more traction in data analytics. Lack of resources to maintain and administer the back end. Oftentimes, customers do not have the required expertise to maintain their data analytics platforms. This is especially true if the customer had only a single person who administered the platform before leaving the organization. You want to avoid knowledge silos like this for this exact reason.
DJ: How much work does data analytics require in terms of keeping it up-to-date?
Kelly:The amount of work needed to keep your data analytics program up-to-date can vary greatly depending on the size of the analytics’ footprint and the architecture used. Typically. there are batch data loads running at various frequencies that require upkeep.
The front-end of analytics also requires regular attention when adding new users, incorporating various forms of security and continue building out new, relevant KPIs and dashboard content as the business changes over time.
DJ: What is the Syntax solution?
Kelly:Syntax has three different solutions we currently offer to our customers. The three solutions are:
Oracle Business Intelligence (OBIEE) and Hyperion Solutions – Within this practice area, we specialize in and provide both project implementations and ongoing managed services to support existing environments. We support both on-prem and cloud-based architectures for these solutions.
EmeraldVision – This solution supports specifically SaaS-based analytics for which customers pay a simple monthly subscription for their users to consume the analytics. This leverages a serverless architecture for our customers, and we maintain the back-end data loads to limit the internal expertise needed by our clients for ongoing administration. And because it’s a SaaS architecture, customers don’t have to worry about future upgrades because their environment is regularly enhanced. They also don’t need to worry about the costs required by maintaining the infrastructure itself.
Analytics on AWS – We leverage various AWS building blocks, such as S3, AWS Glue, Redshift, Athena and QuickSight, to provide our customers with flexible data lake and data warehouse solutions. All of these solutions leverage a cloud-based serverless architecture and provide superior performance, elasticity and capabilities for our customers, and additional functionality around machine learning and AI.
All of our solutions incorporate data warehouses with built-in analytics capabilities and primarily focus on customers with various ERP systems, such as SAP, JD Edwards, and EBS. Our data analytics solutions regularly supplement our customers’ ERP data by tapping into additional sources like CRM (Salesforce, Oracle Sales Cloud), marketing data and any other third-party on-prem or cloud-based application data.
More about Saas, csio, Software, predictive analytics
 
Latest News
Top News