I've reached the point in my business life where I expect—even demand—that information be available to me wherever and whenever I want it. That doesn't mean I go get it if I want it. It doesn't mean it will be delivered to me only in an email or a chat message that I'll have to remember to open.
What it does mean is that it's embedded in the place(s) I frequent—the company portal, the business applications I interact with, the customer management system I live in. Embedding isn't a special use case for analytics—it's a first-class component in every analytics strategy.
Analytic content must pull double-duty
Over the years, I've had a mental model in my mind that I can't shake. For many, BI and analytics is fully integrated with data, but it's often separated from the business process/applications world. I think of it as "separate, but equal"—two related systems that sit side by side but aren't fully connected. How many times have you "alt-tabbed" (shifted between one system and another) to get the information you need to take some action in another system? My guess, dozens of times a week... and then forgetting what it was when you go to take a step.
Subscribe to the Oracle Analytics Advantage blog and get the latest posts sent to your inbox
Analytic content needs to be able to pull double duty—that is, to stand alone as needed, but also to be embedded into the places where it will be used to drive the right outcomes. If you're looking for a way to increase consumption of analytics beyond the 35 percent of people who use it today, this is a great strategy to employ.
There are three "must haves" needed with embedded analytics.
1. It Must Be Active, Not Passive
When I think of embedding, I don't think of a static report. While that may suite some use cases, I believe I should be able to access the data and analysis I want, then take it further to explore why something is the way it is. So, any embedded analysis should let me go to town on the data.
2. It Must Be Easy to Embed, Not Hard
Any tools should make it easy to embed. That doesn't mean you have to dive into deep coding to make it work. Someone like me—who knows data and analytics—should be able to embed content so others can easily get to it. That means complexity must be shielded from the embedder. See the video below as an example of easy embedding. Make sure you listen to the narration.
3. It Must Be in Context, Not a Blank Canvas
Embedding a tool with no content is a non-starter. It has to have data in it, in the context of the activity I'm running, or in the context of my role and responsibility. If I'm lucky, it may answer my question right off the bat; but it's likely I'll want to analyze further to fully understand why something is the way it is. That's why I like augmented analytic function to be there for me to guide me in the right direction.
Think of embedding analytics holistically.
You've spent months, even years, getting the right data, preparing it, cleansing it, visualizing it, reporting it, sharing it. Go the next step and embed it. You won't regret it. And your colleagues will praise you for it—freeing them up to move faster, think smarter, and act with conviction.
To learn how you can benefit from Oracle Analytics, visit