Email This Post Email This Post Print This Post Print This Post

Will a SaaS ERP Solution Work For You?

POSTED BY Kevin Marshall on Dec 3 under Technology

SaaS (Software as a Service) or On-Demand Software has matured considerably in the last few years. Concerns about security, backup, monitoring, scalability, and premium features have been addressed by the serious providers. Compared to an On-Premises solution (the software, hardware, and data reside at and are under the control of the organization using it), there is little doubt that startup costs are less. There are also claims that TCO (total cost of ownership) is less expensive, however these claims are typically based on uncertain assumptions and result in enthusiastic disagreement.

SaaS ERP:

The most popular SaaS applications are still those meeting a specific business need, for instance CRM (Salesforce.com), Google Apps, and payroll. However, SaaS ERP (fully integrated accounting information system, inventory control, Sales, Marketing, HR, and order tracking) software suites are starting to become more seriously looked at and adopted by the SMB market. Netsuite has been the most popular SaaS ERP offering to date. Recently, SAP has announced Business byDesign and Epicor has announced Epicor 9 indicating that the SaaS ERP space may become a serious alternative to the On-Premises model.

Three Questions:

The suppliers of SaaS ERP are quick to claim reduced startup costs, lower TCO, and feature rich modules. But there are three critical questions that must be asked before one can determine if a SaaS ERP solution is the right choice for their business.

1)  What level of customization is required? Some would argue that software is becoming so sophisticated there is no need to customize. The willingness of an organization to change their processes to match those required by a software solution may get approved in a Groupthink setting. But when it comes time to actually change, you’ll need keen leadership, change management skills, and money to confront the resistance. Additionally, the more sophisticated the software, the more expensive it will be to implement. The implementation of really sophisticated software can become so expensive to the organization; you have to wonder if the TCO promised is really going to pencil out. We all know the maxim “Implement vanilla, right out of the box” but the reality is it will sometimes cost more to conform then to modify.

2)  What level of integration is required? No software application does everything the best. Best of breed software, legacy software, and user demand require organizations to use a plethora of different software products. These products are not islands onto themselves. They have to be integrated to eliminate duplicate effort, maintain data integrity, and ensure data timeliness. SaaS integration is getting better, but integration with On-Premises software is still easer and always possible.

3)  How sensitive is the data? You have to determine what your data security requirements are and you must make data security a key component of your software selection process. Determine what data will be contained in the SaaS system and match the level of risk to the level of data sensitivity or importance. For publicly-traded companies the Sarbanes-Oxley act holds signing officers responsible for the state of the company’s internal controls, and must report any deficiencies. Evaluating and assuring your own controls is one thing-but how can you be sure about your SaaS vendor’s controls? Some providers have SAS 70 Audit Reports – you might require copies of these even if you aren’t publicly traded.

Other Opinions about SaaS ERP

A recent report from Gartner titled “SaaS Impact on ERP” by analyst Denise Ganly suggests that a suite of integrated ERP solutions is not something that can be reliably delivered right now. In fact, she says viable options for large enterprises won’t be available during the next 5 years. “Except for use in two-tier ERP deployments,” she notes, “large organizations should ignore this space.”

SaaS ERP Future

Even with the successes that SaaS ERP efforts are having in the SMB space, Gartner Dataquest predicts that SaaS will constitute just 16.7 percent of the total ERP market by 2011. That figue is nothing to scoff at but it doesn’t quite match the hype either.

Bottom Line:

A software selection project has many areas of due-diligence. If a SaaS solution cannot adequately answer the 3 questions above there is no need to look at it any further. These questions then are the very first “qualifiers” in the decision cycle. Also, don’t get caught up in the hype that SaaS ERP can be implemented with little or no participation. You can’t just turn it on! For SaaS ERP, the business must still be processes redefined, users need to be trained, other systems need to be integrated, and controls and audits have to be performed.

The concept is nothing new by the way:

A quick history lesson – over 40 years ago, way before the PC, the cost of mainframe and application software were prohibitive for all but the largest organizations. Smaller organizations wanted to use software too and the result was the creation of an idea called “time-sharing” delivered by a “service bureau”. Dial-up modems (think internet), interrupt driven time slicing monitors (think browser) and the term “multi-tenant” (same as today) were the basic elements of time-sharing.

Years later, the ASP (Application Service Provider) model came along. This was basically, the same thing as time-sharing – a multi-tenant hardware/software solution delivered to remote users by a “service provider”. Some software companies were just selling the same old tired software solutions using a new front end. They called this putting lipstick on the bulldog.

New development software like .Net, browser maturity, internet reliability, and most of all – public acceptance of all things internet – has paved the way for the latest iteration of time-sharing – Software as a Service (SaaS). Some SaaS solutions have been developed from scratch, using new web technology and shared data models.

Copyright Old Oak Consulting