SaaS 101: The Drawbacks
A while back I wrote an article called “SaaS 101: The Benefits” in which I discussed some of the benefits of the SaaS model for end users and software vendors alike. Of course, where there’s a yin there’s a yang, and so this article (a long time coming, we know) will explore the other side of the coin.
For the Customer & End User:
- No direct control of the data - One of the biggest hurdles to get over is the control of the data. Specifically, what happens when things go wrong? I’m sure every company trying to sell you a product will tell you that things can’t go wrong and that they will be there to support you for years to come. It is important that you ask the difficult questions: How safe is my data? Will I be able to download it? Will it be disposed of properly and safely? Can it be sold? Can anyone else host the application and my data? Will the application source be opened so that hosting can happen in house or by another provider? Stories of companies going belly up are not uncommon, and not only for SaaS companies but for traditional software companies as well. The only difference is that when a traditional software company goes under, the most you might lose is the years of support you were expecting from the vendor. When your SaaS provider goes under, deeper implications surrounding your vital business data have to be considered.Â
- Internet connection required - I don’t know of many businesses that run without an internet connection these days. Nonetheless, it could affect your operations if you need to access an application and the internet connection is down.  A good set of companies are trying to solve this problem by allowing their applications to continue to work in a disconnected fashion for a period of time but at some point you will need to sync back up to the server. If this is a big concern for you make sure that your provider can address this need.
- Dependence on an outsider to run your business - In a big way, you are trusting an outsider to help you run your business, and if they are not keeping their end of the bargain it can really affect you. To keep it in perspective, these people are out there to stay in business and they do this for a living so arguably 95 out of 100 times they can do it a lot better than you could in house. This does not mean that you shouldn’t be aware of the implications so make sure you ask the tough questions.
- Security awareness - Another big hurdle is security. This concern is the umbrella that is home to the concerns above, as the common thread among them all is that they make you consider how “secure” you feel with SaaS. You are trusting your really valuable data to someone else. This can be a painful reality to accept but most security breaches occur because of disgruntled internal employees that end up selling or releasing the data when they are fired or when they quit, having your data managed and stored by an expert of the application is not a bad idea as long as they take it as seriously as you would. Again, it is in their best interest to do so but make sure you trust your provider the same way or more than you would trust your internal IT department and again, don’t forget to ask the tough questions!
For the Provider:
- Focus on customer satisfaction - This is one of those bad things that it’s good to have and makes great companies but we have to mention it anyways. SaaS providers need to focus on customer satisfaction month in and month out or they will lose their customers. They need to earn their customer’s business every month or they can simply leave. Contrary to on-premise deployments which are very costly and time consuming, if your customer is unhappy with the service he can up and leave at any time with very minimal cost. Some might argue that you can negotiate longer term contracts, make it hard to take their data and all other kind of shenanigans but if you ask me, it is bad practice and if you are not the best, then you better have one damn good reason why they should stick with you other than a binding contract.
- Harder development process - There are many different approaches to writing SaaS applications and they are outside the scope of this article but the bottom line is that there is a whole new set of things that you need to worry about when writing a SaaS application that you otherwise wouldn’t need if it were a traditional on-premise deployment. Things like tenant isolation, provisioning and scalability to mention a few could be a hard thing to tackle where you wouldn’t even have to think about if you were writing an on-premise application. Nowadays there are several companies working on “SaaS Platforms” including mine that make this a lot easier but none the less it’s something that you didn’t have to deal with before. For a great article on how to transition your company into SaaS read Sinclair’s article where he outlines a couple different strategies. Another thing that makes the development process harder is finding the right talent as the skill sets required are more advanced than for its on-premise counterpart but hey, who doesn’t enjoy a good challenge!?
- Compensation issues - One of the early problems for SaaS providers is how to maintain operations when there is only very little money coming in. Unlike traditional on-premise deployments where one deal could bring you $60,000 upfront and carry you for a couple of months while you close more deals, SaaS deals are MUCH smaller so initially it will be a lot harder to maintain operations unless you are properly funded so you can survive until enough money is coming in . Additionally the question of how you compensate your sales team can be a tough one to answer and can vary greatly depending on your offering. It might require you to offer higher base salaries and get creative with your commissions but without a doubt it can be one of the tougher operational questions you will encounter.  As an example let’s look at this scenario: Provider A sales an on-premise application for $2,500 a license. On a decent 20 seat deal they would bring $50,000 ($2,500 x 20) plus an additional $10,000 for support (usually 20%). Out of the $60,000 they would normally give anywhere between 1% to 4% commission to the sales rep leaving the company roughly $58,500 to run operations. In a comparable scenario where Provider B sales the same type of application as Provider A but as a SaaS application it will probably cost around $75 per seat so on the same 20 seat deal they would bring $1,500 a month taking over 3 years to reach the same $60,000 that the on-premise company received. You will need to come up with a good strategy on how to compensate your sales guys but at the same time have enough money to run the company.
- Success can be a problem - You’ve heard many times that being too successful is a great problem to have but in the case of SaaS it can literally bring you to your knees if you are not prepared. This goes back to my second point of SaaS being harder to develop. Things can grow out of control if the application is not architected properly and addresses scalability issues and your service can become unusable over time if it does not scale properly with the addition of new tenants. Make sure you don’t leave the hard decisions for later because you will run into a wall down the road.
After reading this article some of you are probably thinking “Damn, Why would anyone even think of getting into SaaS?!” But for the ones that are still not convinced check the benefits before making any decision. As with anything else you should always make a decision after you have considered all things good and bad.
What do you think? Does it make sense to jump into SaaS from end user and provider perspectives? Do the benefits outweigh the drawbacks? Or is it the other way around?
Loading …We’d love to hear your opinions, so leave your comments!
Â
Links
View Technorati Links
SaaSBlogs Tags
Tagged with: SaaS • SaaS Benefits • SaaS Drawbacks • SaaS Platform • SaaSGrid
(2 votes, average: 4.5 out of 5)
Loading ...
Hi Abe,
this insn’t a mutuall exclusive scenario… i think in some markets and some segments SaaS will win out, but never at 100% penetration