Work through your build vs. buy vs. on-premise decision by taking an honest assessment of your needs and answering the following questions:
Do you need a solution for which no software currently exists?
If you need a solution for a very specific business or organizational function for which you are sure no software has been created, the obvious option is to build, or customize to a great extent. No precedent has been set, and building enables a distinct competitive advantage.
Do you need integration options?
If you need your solution to integrate with other business software, buying a solution that has either already addressed that issue, or provides tools such as APIs that make it an easier puzzle to solve can save a lot of headaches.
Do you have the resources available for the job?
Sometimes it all comes down to the amount of resources that you can dedicate to the task at hand. Knowledge, vision, and skills are only worth something if they can realistically be spared to tackle a project like building contextual software. Keep in mind that building in-house will most likely take key talent away from important projects directly tied to revenue.
What is the relative complexity and purpose of the project?
In that same vein, resources are usually routed to the highest priority projects first and foremost. With a more complex but relatively low priority project, you could be left well short of what you need to effectively build a solution.
What is your timeline for deployment?
If your solution is one that needs to be defined and implemented on a very short timeline, buying is the answer. With the support to deploy quickly, your purchased solution is not just in your users hands in less time, but will come ready with complete documentation allowing them to learn the software quickly.
SaaS provides better protection than On-Premise by blocking threats in the cloud, away from your network. SaaS eliminates the risk of having a single point of failure, and shifts the risk from the client to the vendor, with SLAs that guarantee uptime. Trusted cloud providers leveraging economies of scale (such as Amazon Web Services) invest more in security certifications, services and data privacy than any single client company would or should.