BeyondCore Review Summary.
BeyondCore is something of a glimpse into the future of business analytics. The current vogue in analytics is for all things visual – graphs, maps, charts, dashboards and so on. While this is a significant move forward from the paginated reports we have all come to love and hate, data visualization is not the end game. In fact the visual analysis of data has a number of dangers associated with it, and not least the fact that interpretation is entirely subjective. Give a chart to ten people and you will probably get ten different interpretations – some seeing outliers, others see a trend, and other may see nothing at all of significance. It is also quite time consuming to explore data in an attempt to find out why something is happening the way it is. What we really need is a business analytics platform that is capable of automatically finding significance in our data and telling us why it is significant. Enter BeyondCore.
The four dominant modes of analysis are covered by BeyondCore – descriptive (traditional BI), discovery (looking for unknown facts), predictive (finding consistent patterns that can be used in future activities), and prescriptive (actions that can be taken to improve performance). What is particularly significant about BeyondCore is that all the complexity is hidden from the business user. Users simply focus on the data that are of interest and BeyondCore does the rest – revealing insights, predictive patterns and remedial actions. Nearly a decade of development has been invested into this platform, so that complex analysis can be accessed by business users via an easy to use interface.
Making business decisions deserves nothing less than a platform that will indicate whether a fact is significant. Simply staring at a chart will not do this, and human biases are just too unreliable. If your business needs to move beyond subjective analysis to something that is more rigorous, then BeyondCore will provide that rigor in a way that is accessible for all types of business user.
A good example of an automatically generated data story is shown below:
Under the Hood
Machine learning and regression analysis operate in the background looking for meaningful patterns that stand the test of statistical significance. Each graph is annotated with key insights and focuses the user’s attention on the crucial details. Additional graphs that the user should see are recommended, to better understand related patterns and the overall context. Users can still specify any graph they wish to see, and BeyondCore treats it as a starting point for further guided exploration. Based on the user’s interaction, BeyondCore generates mathematical models to prescribe specific actions that users can take and predict the impact on their business metric. BeyondCore’s models can be exported as R code for further analysis or refinement by data scientists.
Testing for Soundness automatically checks if a pattern is statistically significant and compare its relative importance to all other patterns being considered, and diagnostic analysis evaluates all possible hidden causes (confounding effects) for visually obvious patterns. Component Driver Analysis breaks out the different statistical effects of each of the variable combinations on the outcome variables.
Prescriptive analysis and forecasting is supported through several features:
- Predictive Analysis automatically generates models based on patterns in the data.
- Fine Grained Forecasting automatically builds a bottoms-up forecasting model based on your data.
- What-If Analysis allows users to evaluate alternative outcomes by simply changing the prediction scenarios.
- Prescriptive Analysis recommends actions to optimize a selected business metric and predicts the outcome that will be achieved.
Collaboration features including Story Sharing which allows collaborators to securely share, edit, and comment on stories to find additional insights together. It also allows users to download and publish stories in Powerpoint, Word, or interactive HTML format for broader audiences.
Finally BeyondCore offers a Microsoft Office add-on which is essentially a cut down version of the full product.
Data
BeyondCore supports One-Click Data Cleansing automatically highlighting data quality issues; and recommends and performs remediation actions. It also supports Data Transformation allowing quick and powerful mapping, formatting, and transformation of data.
Most data sources are supported including SQL & NoSQL data sources, including SAP HANA, Oracle, Hadoop/Hive, mySQL, Excel and CSV.
Management
BeyondCore can be deployed in-house or in the cloud as a service. The architecture is highly scalable and is provided with tools for auditing and monitoring to ensure data security and corporate governance.