{"id":1122,"date":"2015-05-21T16:44:23","date_gmt":"2015-05-21T20:44:23","guid":{"rendered":"https:\/\/solutionsreview.com\/business-intelligence\/?p=1122"},"modified":"2016-10-04T14:14:45","modified_gmt":"2016-10-04T18:14:45","slug":"getting-started-with-embedded-analytics-4-things-to-know","status":"publish","type":"post","link":"https:\/\/solutionsreview.com\/business-intelligence\/getting-started-with-embedded-analytics-4-things-to-know\/","title":{"rendered":"Getting Started With Embedded Analytics: 4 Things To Know"},"content":{"rendered":"

\"oie_tcsPh5eDqQ8P\"<\/p>\n

The folks over at Sisense<\/a>\u00a0recently published an interesting blog post<\/a>: A Beginner’s Guide to Embedded Data Analytics<\/em>. It seems like there’s a lot of value here, so I thought I’d go ahead and summarize the post and provide my take. Eran Levy<\/a>, the post’s author, notes that no matter whether you are in producing automation software, SaaS products, or cloud applications, the reason why enterprise companies are becoming increasingly interested in this kind of stuff is because they are collecting\u00a0lots of data.<\/p>\n

Not only are businesses collecting lots of data, but they are starting to realize the value it adds to their organizations. Embedding powerful BI and data analytics tools within existing applications can give your product a competitive edge. Before jumping in, the author recommends you weigh the following:<\/p>\n

1. Do you buy or build?<\/strong><\/p>\n

This one is simple. You want to add an embedded BI feature to your application. Do you want to buy an existing embeddable software and integrate it or develop an analytics platform on-premise? The author makes the claim that in a perfect world the choice would be to keep everything in-house to retain full control and include exactly the kinds of capabilities that your clients will find useful.<\/p>\n

In reality though, BI is not a necessity for most organizations, and developing a platform from scratch could take a long time and cost a pretty penny. From a developmental standpoint, visualization tools don’t require an extreme amount of work, but a great BI tool is about more than just looks; it also needs to perform.<\/p>\n

So this type of platform is tough to develop and would cost a lot. Even if you were to get past those obstacles, you might still fail to achieve the level of functionality that your customers would expect from an out-of-box solution. It looks like our decision has been made for us.<\/p>\n

2. Ease of implementation<\/strong><\/p>\n

The author throws us a curveball here, saying: “Having said that, you also shouldn’t overlook the possible hidden costs and time-sinks that come with some embedded solutions.”<\/p>\n

Integration issues between your own software and the embedded analytics platform you chose have the potential to increase cost and production time. Nothing good comes as a result of those two factors. Time can become a factor, as developmental edits between you and your BI provider will need to take place.<\/p>\n

In addition, some BI software is quite complex, meaning that there will be an increased period of training before the system goes live. As a result, the cost is exaggerated, and time to market is slowed. Going the external route of buying an embedded BI feature is not the end all either, as it doesn’t guarantee faster turnaround, making it important to test a potential solution thoroughly before deciding on it.<\/p>\n

3. Defining requirements<\/strong><\/p>\n

There are a wide variety of BI solutions out there, and all of them do interesting things. In digging deeper though, one might find that the differences between them is substantial. The author notes: “The type of tool you\u2019ll require depends, among others, on the volume, variety, and velocity of the data you plan to process.” Things to consider:<\/p>\n