Proactive and suggested insights, generated by machine learning, is the difference between biased and unbiased discovery.
Users who rely on traditional reporting platform such as: IBM Cognos, Tableau or Power BI, base the premise of their question on a preconceived notion and obtain data that may have fit that notion.
As a result, major decisions were being made despite the fact that “no one knows how one sided, bias or even significant the insight is”.
The alternative, powered by Hyper Anna, relies on machine learning to scan the entire data and find patterns, key drivers and anomalies that are significant. These proactive suggested insights help Hyper Anna’s users to sharpen their questions.
An example that can illustrate this point:
A wine manufacturer wanted to know which brand had the highest turnover last week. A straight-up numbers crunch showed that brand X suffered the lowest turnover.
Conventional wisdom pointed to the brand manager. But a Hyper Anna - based analysis showed that while the overall turnover of brand X has dropped last week, it is only driven by a small number of sub-brands and concentrated in one particular distribution channel. This distribution channel did not show up in the traditional numbers crunch, however, is a factor that impacted other bands as well.
Single-factor cause-and effect thinking has become entrenched as industry wisdom because of the limitations of analysis 10-20 years ago. Now that query can be analysed for multiple factors in seconds, the expected answer may be wrong and the unexpected answer may be material.
You can view how Hyper Anna’s users interact with the platform to identify the right questions to ask and get their answers here.