Create, deploy, and maintain analytic applications that engage users and drive revenue. See a Logi demo

Embedded Analytics

5 Best Practices for Great BI Deployments

By Michelle Gardner | December 28, 2017
Share on LinkedIn Tweet about this on Twitter Share on Facebook

Business intelligence deployments are notoriously complex—but they don’t have to be. By following a few proven best practices, you can choose and deploy an analytics solution that will go the distance (and avoid a painful rip-and-replace in your near future).

>> Related: Get a Blueprint for Modern Analytics <<

1. Ask for Help

If you’re on a tight deadline (and maybe even if you’re not), look for an analytics vendor who offers third-party support, such as professional services. They can help train your people, determine your architecture, come up with a data strategy, and/or build a custom feature. All in all, this support will help you get up and running faster than other solutions.

2. Consider Existing Skill Sets

Make sure your development team can actually use your chosen BI solution. Every platform is a little different, so involve developers early and get them into the weeds as soon as possible. Do they have the skills to support an intricate development environment? If not, you may need to find a different solution or compensate for any missing skills with new team members or a third-party support team.

3. Don’t Forget Data Architecture

In BI, data is everything. Consider where your data lives and make sure it’s accessible for the new tool. Also think about what will happen to your data if you need to rip and replace again in the future. Most successful data strategies come from a partnership between you, your BI vendor, and your data consultant (if you choose to have one) or internal data expert.

A solid data strategy means your data is stored in a location that’s accessible for reporting, and that your developers understand the data model and know how to query it to build reports. In some cases, you may be able to bolt your BI solution onto your current data model. But in others—such as when you’re replacing a proprietary solution that ingests your data with it—you will need to re-architect your backend for the data layer.

4. Involve Users Early and Often

Your solution could be amazing, but you can’t expect your users to touch it if you don’t vet it with them first. Ask yourself what it will take for your users to learn the new tool. Are you building something that matches well with their skill sets? What resources, training, or documentation will you need to ensure their success?

A common complaint we hear from customers is, “Our users hate this.” But that’s rarely the real case. Once you dig in and talk to your users, you may discover they find the tool difficult to use or wish it had single sign-on—or any number of other pain points that can easily be addressed if you ask for feedback from the beginning.

5. Plan Your BI Roadmap

Your analytics timeline doesn’t stop at launch. It also includes iterating on future versions. Think far ahead and establish a well-defined process for what and when you need to deliver each of your business requirements and capabilities. The right solution should support the latest capabilities and give you the flexibility to meet challenges you don’t even know you have yet.

Look for a BI solution with plenty of customization options, so you’re not locked into a rigid experience. You also want the ability to connect to a slew of different data sources (not just the ones you have today), along with a solution that scales easily (so you can create whatever you need).

Learn more in our ebook: How to Break Up With Your BI Solution.

 

About the Author

Michelle Gardner is the Content Marketing Manager at Logi Analytics. She has over a decade of experience writing and editing content, with a specialty in software and technology.

Subscribe to the latest articles, videos, and webinars from Logi.