There seems to be a lot of talk about Software-as-a-Service (SaaS) applications coming from vendors and consultants these days. I’d imagine they’re as great as people say for some companies, but what are the SaaS data integration challenges that IT groups face between SaaS apps and in-house systems? If we go to the cloud – and there’s been talk around that we might explore it – what should we look out for?
One of the hottest areas of the software market during the past few years has been the delivery of new software products through Software-as-a-Service offerings. The SaaS model can take many different forms and views, from a hosted application that end users log into and use via their Web browsers to specific discrete application processing services and technologies that are hidden behind the scenes of a desktop application (e.g., data integration tools or address standardization and customer demographics services). The value proposition from vendors within this space is usually focused on a lower total cost of ownership, including less expensive maintenance. The actual processing occurs on the vendor’s technology infrastructure.
The concept of outsourcing computer processing has been around since the advent of commercially available computer processing. In the old days, the idea was to make processing resources available to organizations that couldn’t afford them. Today, the idea is to leverage specialized processing capabilities to augment a company’s existing infrastructure investment. As such, the value of SaaS isn’t related only to cost but also to time to delivery. In a number of product areas, it’s simply quicker to deploy a SaaS application than it is to install, set up and configure a traditional packaged software product.
Some of the things that we often encourage folks to consider when looking at SaaS technologies are:
- Moving data to/from the SaaS vendor. There are SaaS providers that support everything from accounting to analytics to extract, transform and load (ETL) capabilities. In most instances, in order for a SaaS product to function, it will require regular data feeds from a company’s operational systems – and in turn will be required to provide data back to in-house systems to support other downstream business processes. It’s important that issues such as packaging, formatting and latency are well understood.
- Content and processing security. How will system access privileges be implemented and managed? While there’s very little to worry about regarding the protection of your data, understanding how those details are addressed and managed is important. You’ll inevitably need to add users or modify their access to the system; you’ll also want to support selective security for individual data content and processing functions. All of the SaaS vendors understand these issues – it’s important for you to understand the details.
- Application processing and data customization. This item is focused more toward the application providers (e.g., CRM and accounting software vendors) than the discrete processing services and technology providers. It’s important to understand how flexible the SaaS software must be to support your specific processing needs. If you expect to outsource business processing functions to a SaaS vendor, you’ll need to be specific in defining the processes you want it to handle. Most SaaS applications are highly flexible – the trade-off will be upfront configuration and programming time.
- Response time and availability. This item rarely gets much attention in the SaaS world. Most SaaS vendors brag about their scalable offerings, their flexible infrastructures and the massive amounts of horsepower they have available. But the important item to focus on is response time and availability. If you’re in the position of outsourcing core operational processes to a SaaS vendor, you’ll want to make sure that you can always run your company.
Dig Deeper on Enterprise data integration (EDI) software
Related Q&A from Evan Levy
Learn the difference between change data capture (CDC) and data federation. Find out how companies can use both data integration technologies to ... Continue Reading
Find out if real-time data integration applications has more data quality issues than other approaches. Also, see if real-time or near-real-time ... Continue Reading
Get a definition of operational data integration, plus learn how it differs from analytical data integration and find out if you should implement ... Continue Reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.