Embedding analytics into your application can be challenging. Gathering the information your organization needs to effectively assess embedded analytics vendors can be a time-consuming step in the evaluation process. In this blog, we’ll talk about the questions you should ask to successfully evaluate embedded analytics products and maximize project success.
Make a List of Ideal Features
Once you’ve decided to improve your application’s analytics capabilities the first step you should take is to make a list of reporting and dashboarding features your users will want from your application both in the short and long term. This means talking to customers and other stakeholders. By laying out an ideal features list you will get a better understanding of what to look for from vendors, though no single vendor may tick all of the boxes you lay out.
Technology and Architecture
The next step is to define technical project criteria to take stock of what you will need from an embedded analytics vendor to be the right fit technologically.
Here’s a list of sample questions you should ask internally:
- What type of programming languages are we comfortable working with?
- What type of environment are we comfortable installing our software in?
- How do we want to host our analytics in the cloud? On-premises? Or a mixture of both cloud and on-premises?
- How do we want to distribute this new analytics capability to my users? Will this be a core part of a product update, an add-on, or will it be a tiered approach where we roll out some capabilities but charge an additional amount for specific features? These questions can often affect how you decide to architect your integration.
- Is my data easy to access?
- How can I integrate security permissions from my current application into this new component part?
Here’s a list of sample questions you should ask vendors:
- What technology stack is it built on? Is it built on Java, .NET, PHP, et cetera?
- How do you white label the solution for a seamless end-user experience?
- What controls are customizable via API?
- How do you integrate on the front end and/or the backend?
- If your application is multi-tenant, does the vendor support multi-tenancy?
- Do they support Single Sign-On?
- What types of security do they support?
Development and Report Development Style
The embedding process can vary greatly from vendor to vendor, understanding what time investment is needed from your development team is a great way to estimate total project cost. Many vendors offer scoping calls or initial POC calls to give you a better understanding of how long a POC may take and how much time it may take to fully implement.
Report development style is also crucial to look at when evaluating. By this what we mean is the embedded analytics solution you’re looking at more self-service focused, developer focused, or a combination of both? Understanding how reports are created in this new system is very important for understanding how this product will integrate with yours and more importantly how your users will interact with your application in the future.
Licensing models can also be a driving force when looking at whether to build or buy an analytics solution. You should look for flexibility within how a vendor’s licensing models structure. Whether that be a perpetual license model, where you pay an upfront fee and ongoing maintenance fees. Other options include term-based models where you pay a yearly fee over a set term number of years or royalty models where you pay a percentage of sales. Understanding the different licensing models will give you a better understanding of how a 3rd party product will fit within your product’s pricing structure and how you will ultimately sell additional analytics capabilities to your customers.
The Right Partner
Finding the right partner can be the hardest part of evaluating an embedded analytics solution. It’s important to understand what you’re looking for in a partner and to constantly challenge your presumptions because you’re going to be working with this vendor for a number of years. Finding a partner that has the right technology stack, the right options, development style, resources is only part of the equation. An organization’s ability to meet your support demands is also essential, whether that be the pre-sales help to create a POC, on-boarding you after purchase to make sure you start off on the right foot, or providing help down the line when it’s needed. We have customers who have been with JReport for 10 plus years and have one of the highest retention rates in the industry because we consistently focus on embedding since day one and our support system is second to none. The right partner should also have the right outlook for their future development. A vendor that has the wrong strategic direction or worse yet no strategic direction can be disastrous for the future of your product’s analytics.
Now that you’ve gathered all the information you need, create a set of evaluation criteria to define what a successful POC may look like. This improves the likelihood effectively evaluating a product from installation to final proof of concept. This is the most crucial step you’ll take when you’re evaluating embedded analytics products because it will be the ultimate guiding force for the remainder of your project.