Pixel Facebook LCP""
KPI6 Logo

predictive analysis of a Reality show

Predictive analysis: the X-Factor Winners

It is obvious, it is in human nature to bet on someone. 

How many times have you made a bet on social networks with your friends or family? Or on the victory of a singer in the Sanremo or Eurovision Contest? Or on your favorite soccer-team? Many predictions, hopes, hypotheses … but in the end only one winner.

In KPI6  we are not magicians, we don’t read tarot cards or we foresee the future inside a crystal ball. We are pragmatic (but also creative) and believe in one thing: in the power of data.

Our “power” is to make the best use of AI technologies in order to read conversations on the web, measure their volumes and understand the emotions and feelings behind a given topic.

A predictive model that can be used in any field, both in corporate strategies and research, in entertainment and in public events. And it is thanks to our tools that in the last two years, we predicted the winner of X-Factor Italia in advance.

How did we do it? No agreement, no rigged maneuver, we only read what the data told us.

Would do you like to know how a Digital Consumer Intelligence platform works?

book a demo

Book a Demo to get a hands-on demonstration of KPI6. We can suggest to you how to apply a data-driven approach to your business.

How we made the final winners prediction

Over the years, X-Factor has established itself as a real social phenomenon. Since its creation in 2004, it has always fascinated the public thanks to its judges but above all to the skill and incredible voice of its participants.

It has increasingly become an essential date in music lovers’ lives. Especially in Italy, where this competition is loved by millions and millions of fans who support their idols.

Since the first episode of the auditions, since the first “boot camp”, there are already so many fans of these future singers, fan clubs, social fan pages and they wonder who will be the next star who has got the “X factor”.

In 2018 we made a forecast, obtaining an amazing result: we announce the ranking of the 4 finalists, which was then confirmed 100% by the public vote, with Anastasio as the winner. 

In 2019 we wanted to replicate the precision of our predictive model, and we compiled a new ranking taking into account what the reactions of the public to the participants in the previous months have been.

So we discovered the winner of X-Factor 2019 a day in advance, announcing the young Sofia Tornambene.

A win that was on the cards, for KPI6.

Step 1: topic-research creation for comparison

First of all, it is necessary to define the search criteria that allowed us to draw the information useful to understand in advance the result of both finals.

We created a query on Twitter in order to understand all the conversations that explicitly referred to X-Factor. But that was not enough: these conversations also had to contain an opinion on at least one of the participants in the final.

We have used all the official keywords and hashtags, references and mentions to the official X-Factor account and the names of the participants in the final.

Step 2: analysis of the conversation volumes

Sometimes the most discussed phenomenon is that which obtains greater visibility and which, consequently, could emerge more easily in a national-popular context such as a singing event.

In 2018, Anastasio, the young rapper from Sorrento, showed for several weeks an overwhelming power on social media. The situation changed in 2019 where, analyzing the number of conversations on Twitter about each finalist, divided by episode, the final winner Sofia is never to the top, compared to her opponents.

Step 3: Contents analysis before the Final

Obviously, it’s not just the volumes of online content that suggested to us how the final would have gone.

In the XF12 predictions, isolating the conversations that speak of Anastasio before the live, the result was already enough evident: sentences of exhortation to victory such as “Anastasio must win” and elements of his single “universal judgment” occupied the first positions of this chart.

Also according to the Emotion Analysis, the most evident emotions in the conversations that Twitter users addressed to the rapper were a deep admiration and emotion.

With XF13 we have added external data to the ones of our Digital Consumer Intelligence platform, useful above all in the formulation of our prediction: the Spotify ranking of the singles of the finalists.

By adding these data to those already in our possession, it was possible to understand a further trend of sentiment towards singers. Again, Sofia Tornambene’s single “A domani per sempre” (Sony Music) ranked second in terms of streaming number, compared to La Sierra.

So what made us bet on this sixteen-year-old as the winner of X – factor 13?

The sentiment that users feel towards her. It is precisely from the sentiment analysis of the conversations on social media that we notice a clear detachment of Sofia from other singers, with a large percentage of positive sentiment (81% vs 19%).

Predictive machine learning

As we said, we managed to predict the XFactor winner for two years in a row.

This analysis has highlighted what is the potential of a big data approach that involves the interpretation of social and web. We used a TV show as case-study, but let’s imagine what a brand could do if it could literally see in the future:

  • Anticipate competitor’s moves;
  • Hypothesize the results of the launch of a new product on the basis of similar previous episodes;
  • Monitor the perception of a brand day by day, hour by hour.
  • Understand the evolution of the level of trust of its consumers
  • Address crisis management and understand when is the right time to restart and rethink its own strategies

Thanks to our AI technologies, today it is possible. Just be guided by data and modulate the winning strategy for your business. An approach that can provide accurate predictions of other important future events, both political and of entertainment.

We are planning new research to find out the result of the major public events that are usually heavily influenced by public opinion.

Because … in data we trust!

Would do you like to know how a Digital Consumer Intelligence platform works?

book a demo

Book a Demo to get a hands-on demonstration of KPI6. We can suggest to you how to apply a data-driven approach to your business.