Of course, everything that is presented like the new Holy Grail, we should take with a pinch of salt. So that's what we did when Big Data was presented as the new solution to everything. But the power and importance of data and analysis can't be denied of course. Let's have a look ...
Every big company these days is a tech company in its core. Amazon, Netflix, Spotify, Apple, they all are companies that deal in data, in recommendations, in digital distribution. They all have a mindset of helping you best with the use of data analysis and algorithms. Yes, Netflix brings you movies, but without the underlying tech architecture we would look at buffering videos and would get bad recommendations. And of course Spotify is a music platform, but their recommendation quality is unprecedented.
In a digital age everything is data. And when everything is digital and everything is in a database we can track it, we can trace it, we can count it, we can analyze it.
For example, read how sophisticated Discover Weekly by Spotify works.
To create Discover Weekly, there are three main types of recommendation models that Spotify employs:
- Collaborative Filtering models, which work by analyzing your behavior and others’ behavior.
- Natural Language Processing (NLP) models, which work by analyzing text.
- Audio models, which work by analyzing the raw audio tracks themselves.
What does that mean?
- Spotify knows what I play and like and compares that to others.
- Spotify crawls the web and knows who writes about what and sees relations.
- Spotify analyses music files and recognises metal, house, hip-hop and if the tunes are angry or uplifting.
By combining those three elements, Spotify is better than any other service in coming up with really good personalized episodes of Discover Weekly.
When you look at data you look at the past, but can data from the past predict the future? That's what Spotify does as well of course. It lets you listen to music you haven't heard before because it thinks you will like it.
But there is more. Hitwizard was designed by Dutch company Goldmund Wyldebeast & Wunderliebe and is based on the Spotify API. It is a self-learning computer algorithm that makes predictions on the likelihood of a track appearing in the Spotify Top 200, based on airplay and other unique characteristics.
And Shazam is very good at that as well. As you know, by starting Shazam on your mobile, it starts listening to the track you hear in a bar, a club or on television. And when people do that, that is a strong signal you really like that track. Shazam knows a lot about upcoming hits too.
So what does that mean for a club?
In general clubs are no Amazon, Netflix or Spotify of course. They are no tech companies to the core. But that doesn't mean they can't use data to streamline their processes. Or they can't combine data to learn and optimize their business. Digitize every bit you can so you can analyze and optimize. But there is already a lot of data available too:
- Data from the website. Who visits which page, what are people searching for, etcetera.
- Data from the ticket provider. Who buys what and when, where do they come from, etcetera.
- Data from the bar. What do people drink and how much of it?
- Data from the artist's (social) media. How big are they where?
Etcetera, etcetera. All valuable data to optimize, but to predict as well.
Paradiso is going to experiment with predicting ticket sales. If we combine all the available data and use Artificial Intelligence, how close can we get to the actual amount of tickets being sold? That would help the programmers of Paradiso a lot. We only just started, we'll keep you posted on this one!
For now: digitize every bit of your business!