Software-as-a-Service (SaaS) application providers are unique compared to traditional direct software companies. Most companies buy software for their internal users, but in a SaaS scenario, there’s an extra layer as their users are their customer’s users. SaaS companies serve their users indirectly as opposed to directly. They know how many accounts are on their platform, but predicting actual end user usage tends to be a challenge. This is the primary differentiator between SaaS companies and direct software providers.
In some ways, IT departments inside large enterprise organizations can also be seen in a similar light. Many of these IT groups build software that serves the customers of their customers similar to a software company. For example, many logistics companies build the software for use with their partners. While they can predict how many partner organizations will use their software, predicting end user usage within each partner continues to be a challenge.
Increasingly, SaaS product managers are being asked to provide better custom reporting and analytics capabilities in their products. While these same product managers are experts in their industry, reporting and analytics simply isn’t in their wheelhouse. It’s just not their core competency. So, they are faced with a choice to either to develop that functionality themselves or purchase it from an analytics software company…the classic build vs buy scenario all product managers will face.
Initially, SaaS providers may try to build this functionality themselves, afterall there’s plenty of charting libraries to choose from. However, those that believe they can build and maintain every a proper analytics capability are generally fooling themselves. The requirements for reporting and analytics go well beyond chart types and they are constantly changing, and SaaS companies can’t possibly be expected to develop and maintain advanced analytics capability as technology changes.
For that very reason, many turn to OEM (original equipment manufacturer) components. They license reporting and analytical capabilities from another company and integrate it into their own software. OEM is not new, in fact, even the most successful product companies use OEM components today. Think of the iPhone. Apple does not build every component themselves, but rather they get the processor, the memory, the case, and more from other companies.
As SaaS providers consider using OEM tools to provide custom analytics capabilities within their SaaS app, they have to keep the requirements unique to SaaS products in mind. These requirements include:
- Security and multi-tenancy — SaaS products are hosting many different customers, often on the same servers and databases (multi-tenancy) so it’s imperative that security is well-managed. Given that each individual customer has their own security requirements on top of the multi-tenancy security needs, managing security is much more difficult for SaaS providers than it is for direct software providers especially within industries that deal with personal information such as Healthcare.
- Deployment — The majority of SaaS providers have embraced the cloud for their infrastructure, but when OEM components are included, they also have to be aware of software versioning as it relates to their products. Many SaaS companies have product versions and OEM components have to be maintained in relation to software versions and customer expectations. It’s a complexity not found as often with internal analytics.
- Licensing — The licensing model for SaaS providers is different than it is for direct software providers. It is a distributed model that must extend to an unknown number of end-users.
- Customizations — The software needs to blend in with the SaaS app and not appear to be separate software.
- Architecture — Software solutions from SaaS providers need to be flexible and able to work on almost every type of architecture, especially as technology evolves.
- Performance and scalability — With each customer having potentially hundreds or even thousands of users, the amount of data and processing increases exponentially. Because SaaS providers are hosting and providing the service, it’s critical that they be able to scale and maintain performance of their analytics as their customer base grows, even when they cannot predict future usage among their customer base. Slows reports will hurt customer satisfaction and retention rates.
As you can see, SaaS providers have much stricter requirements even before they sell to their first customer. Bringing in a third-party OEM for embedded analytics means that the OEM partner must also meet that same standard of service. And, if the SaaS provider is asked to customize their software for their customer in any way, can they also customize the OEM embedded analytics? All of this increases the complexity for product and development leaders.
The crux of the overall challenge is finding the right company to partner with for the long term. There are many companies out there that make analytics software. However, very few have the ability to truly meet the OEM requirements of SaaS application providers. The legacy business intelligence tools are simply not built to be embedded within SaaS applications. Their many struggles include:
- Lack of customization to meet the needs.
- Lack the ability to manage licensing scalability, distribution, and security when dealing with end users of their customers.
- And, they lack focus. They are not fully attentive to SaaS application providers. As a result, they risk damaging the relationship the SaaS provider has built with their customers; relationships that are critical to the success of any SaaS provider.
Working with a company like Qrvey whose product is built specifically for OEM and can meet the specific needs of SaaS providers is a big step forward towards delivering the analytics features end users demand. Qrvey understands how to be a good partner and listens to every business, regardless of size. When making a decision, ensure it’s a company that works together to continue to advance the embedded analytics solution as the market and customer needs will change…they always do.
Read more about Embedded Analytics in my previous article: What is Embedded Analytics? Seriously.