BLOG

ML-powered Automated Insights are here to stay

PUBLISHED ON

Tweet about this on TwitterShare on FacebookGoogle+Share on LinkedIn

It’s undeniable that Machine Learning has taken the digital world by storm. But along with its many promises, there has been much hype, making it hard to tell whether particular “AI-powered” labels actually represent true breakthroughs that will have a real impact on our day-to-day life.

We certainly believe that Sweetspot’s Predictions, Automated Insights, and Smart API Calls, our ML-powered features, have earned this label in their own right. While Predictions have been quietly serving our customers and partners from the very outset (2012!), the last two are only now beginning to unleash their true power. I nevertheless believe they all deserve a proper shout-out.

So, let’s take a look at how ML-powered features have revolutionized the way data consumers receive and interpret their reports:

Predictions

By applying predictive models such as the Autoregressive integrated moving average (ARIMA), Sweetspot is able to forecast a KPI’s future performance on the basis of historical data. This is not only useful towards ascertaining whether a particular goal is likely to be met, but it can also help us define the said goal under a more realistic frame.

Sweetspot Prediction Line

Image 1. Sweetspot Built-in Prediction

Automated Insights

KPIs speak for themselves with Sweetspot’s latest feature. By combining historical values, predictive algorithms, and defined objectives, Sweetspot can turn raw data into self-generated insights that appear alongside each KPI – expressed in plain English.

Image 2. Sweetspot Built-in Automated Insights

As you can see in the example above, Automated Insights draw conclusions from historical data, but they also take into account predefined goals (when available) to form a hypothesis on the basis of:

  • Historical values
  • Predictions of future behavior
  • The target objective (the amount we wish to reach, total elapsed time, and time remaining to achieve said goal)
Sweetspot's built in workflow

Image 3. Sweetspot Automated Insights Workflow

Sweetspot continually evaluates Predictions and Automated Insights in order to refine the process. This allows us to deliver an optimal reporting experience that marketers and analysts can confidently leverage to make data-backed decisions.

Smart API Calls

Our last feature in the list is less flashy and obvious, mostly going unnoticed by dashboard consumers. It is, however, a key feature used to improve our internal data integration processes.

Those well versed in data integration projects will be well aware of the many limitations presented by the vast majority of third-party APIs, including data caps or wait times.

To give you an example: imagine a particular data source requires a consolidation period where information for a given month won’t be available until the 5th day of the following month. Sweetspot can detect this behavior, learn from it, and calculate the best possible date to start requesting information from the said data source. We can thus avoid unnecessary (often costly for the customer) API requests resulting in unavailable data.

Looking Ahead

Sweetspot continues to explore new, practical uses of machine learning to improve day-to-day tasks that can help marketers and analysts better understand their data to make wiser decisions.

Do you have a suggestion or a reporting headache you’d like to solve? Let us know, and we’ll be happy to explore new opportunities!

Tweet about this on TwitterShare on FacebookGoogle+Share on LinkedIn

Alfredo Mezquita

Managing Director at Sweetspot. Bachelor in Software Engineering and new technologies enthusiast. I love sports, travelling and cooking.


Add a comment

Try Sweetspot today!

Not Another Dashboard.