Recently examining some data from the report by Accenture, where 150 UK and 150 US IT executives from largest 2000 companies were questioned I found that most of UK and US companies do not see the value in utilizing SaaS model to upgrade their ERP systems. Research stated that only 7 percent of UK firms, and 15 percent of US firms polled believe that current ERP systems would be replaced by new technologies any time soon.
However, there are other opinions that disagree with survey suggestion that large companies don’t see the value of implementing SaaS. A Forrester reports claims that SaaS projects are on hold and calculates that majority of large Global organizations (75 per cent) already use SOA and SaaS to some extent, and 60 percent plan to expand the reach of their using in the future.
As we can see the given data is controversial and not sufficient enough to make any predictions or conclusions. After reading several others reports I noticed that representatives of each side trying to turn the scale proposing their own version of situation and benefits/risks. So, as usual I decided to turn to Linkedin hoping to finally find the answers.
And I’d like to tell that the question got a welcoming and warm reception as many other professionals took this issue very seriously. The discussion has been started and answers ce between traditional ERP and SaaS solutions and what are the forecasts for the nearest future (Is SaaS going to replace traditional ERP systems anytime soon). Especially in the light of great success of Salesforce.com and other well-known providers of customer relationship oriented Sfell at incredible pace. There were different opinions and facts from CIO’s, CTO’s, Enterprise Architects, various Senior Consultants and Analysts, Strategists, Evangelists and other IT related high-level professionals, however, the hottest dispute raised within the Enterprise Architect Network, seems like it was very touchy subject for them.
But, I’m not going to copy and paste all answers and comments, as you can see them personally (the links will be provided at the end of article) I just would like to classify the answers of those highly-qualified professionals who responded on my call (thanks them a lot!) and give their vision on the issue.
So, consider the main pros and cons for SaaS ERP model underlined in the answers:
Pros:
1. Simplicity. In general, SaaS is simpler to deploy, from a technical perspective as you don’t need to purchase any hardware or physically install and configure the software by yourself. To start you just need to simply sign up.
2. Initial costs. While traditional ERP has high initial costs to buy, deploy and config all needed hardware and software SaaS can be launched considerably cheaper without installing hardware and hiring consultants to implement it.
2. Access. SaaS provides with the possibility of having a system totally independent of your position in the world. If you're connected to the web, you can access the system, using a notebook, for example.
3. Better infrastructure and staffing benefits. Few organizations can match the infrastructure and security investments made by SaaS vendors, by offering complete cycle of maintenance services SaaS vendor allows organizations to reduce staffing needs.
4. Free Trials - Almost all SaaS vendors give a free trial. Try out the service, keep it if you like it. Customers can assess the value of the application before paying for it.
5. Risk Free – if the company is not satisfied using the application it is easy to put an end. Just export your data and close your account. And because of this one of the main vendor’s task is to focus on customer service and needs, since it is on subscription model a vendor is judged on month-month basis.
Cons:
1. Integration. ERP applications have many complex transactional integration points with other systems and applications, both inside and outside the company and this is something that is very hard to achieve with SaaS. Each customer instance of a given ERP package will have a different set of suppliers, contractors, customers etc, each with different applications and processes, and a different array of internal systems for manufacturing, accounting, reporting, etc, all of which must be integrated with the ERP system. SaaS is best for applications with uniform use cases, limited integration requirements, and simple transactonal behaviors. ERP is none of these.
2. Security. Companies, especially large ones, are very reluctant to move into SaaS due to security concerns, they want to protect their core data and do not want to have a third party responsible for protecting that information.
3. Costs. There are several aspects should be taken into account:
- Investments. Given the significant investment that most have made for implementing, integration and customization of their ERP-systems switching is not that easy and customers will wait until the investment is depreciated.
- Running costs. Although the SaaS can be deployed relatively cheaper than on premise system an ongoing cost of cloud based offering are high, especially as the company matures.
4. Operational control. Due to the fact that ERP is their core systems, the majority of companies still want to have control over the availability of the service. If the ERP system is down, the companies main processes stop. If CRM is down, only the sales process stops. Although SaaS suppliers are delivering most often a better service than internal ICT-departments, it is still difficult to accept that you do not have any control over your most important system. And what if the SaaS-provider gets bankrupt?
5. Standartization. There is the lack of standardization of SaaS solutions. The industry is still nascent and there is no clear solution upon which a company can design and build a solution around. Given the complexity and high customizability of ERP offerings, it is a risky proposition to bet it all on an offering that is not the industry standard (e.g. PeopleSoft or SAP)
6. Restricted offering. Integrators and other companies that earn money for implementing ERP systems are not willing to implement SaaS-services. The earnings of implementing a SaaS-service is just a fraction of the earnings of a on premises implementation. By CRM-implementations we see sometimes that those earnings are just 10 % of the original 'old'-implementations. These companies are normally in the lead of selling ERP, so they have will hesitate in advising a SaaS offering.
Summarizing, I can say that there are a lot of tradeoffs between two options and although the SaaS model is relatively new all factors should be accurately prioritized and evaluated to identify the best suitable software for each organization. As it was mentioned in discussion SaaS model is seemed to be a good option for small and medium businesses which haven’t made a huge investment in on premise ERP systems. Smaller businesses that did not have access to ERP systems would be of the first to adopt SaaS ERP as their requirements would be lower. Especially, if you look at total costs of ownership of SaaS versus on premise solutions the SaaS will be considerably cheaper meaning that there will be market forces encouraging the SaaS model. As for security reasons the ability to physically isolate instances in a multi-tenant architecture using virtualisation will go a long way to address some of the security concerns, however other data governance and data legislative related questions would also need to be addressed. But as a robust and hosted CRM solutions are so successful and new generation of Rich Internet Applications will deliver a rich and high performance user experience there is an evidence that the market will migrate towards SaaS, the ERP market will follow this development. It only takes a little bit more time.
P.S. Also, I forgot to mention an existing problems with network traffic and bandwidth which can result in productivity and revenue lost on SaaS model. I agree that in some locations the quality of networks remain bad with bandwidth is too expensive, however there are different trends in the world. Moreover, most of large enterprises are already network dependant, so it can be a disadvantage for both models.
