Machine learning is changing monitoring as we know it – are you ready?

Machine learning has officially been around since 1959, ever since Arthur Samuel started using it at IBM. But it wasn’t until recently that we really began to understand just how much it could help solve some of today’s toughest challenges – including those in communications.

A few weeks ago I had the pleasure of speaking on a panel about the power of real-time analytics in integrated communications at PRSA Health Academy. We touched on machine learning and, having recently led the development and launch of Galileo6, Spectrum’s proprietary marketing intelligence platform, alongside my talented team and our partners at Tickr, it’s a topic I’m pretty well-versed in – but I’m always looking to learning new things.

That said, here’s what I took away from our panel discussion:

Change is not coming. Change is already here. In the past fifteen years, the amount of available data has increased exponentially. In 2016, Google was receiving more than 2.3 million search queries EVERY MINUTE. And Siri was answering almost 100,000 requests. This year there are over a billion social media posts every 2 days, and I don’t even want to hazard a guess as to Alexa’s 2018 workload… but I guarantee you, it’s substantial.

So the challenge for communicators is in making the data work for their organizations. The key is in improved statistical and computational methods, not just storage or even computational capacity. The world has changed and the old way of working is obsolete. If we aren’t using real-time data to make better communications plans, identify issues sooner or uncover deeper audience insights, you’re going to be left behind.

Machine learning can deliver faster insights, with your help. Monitoring is an essential function for communications and it can be a key place to glean insights. The problem we are facing is that, historically, monitoring has always focused on aggregating mentions. And in today’s environment we have a lot of mentions – but very few of them actually matter. What monitoring should be focused on is relevance: pulling out the mentions that are most important, based on your criteria for what’s important, and bringing them to your attention as soon as possible.

And this is what machine learning is really good at! Learning from the inputs you give it, getting better over time and ultimately helping you sort through data faster.

When your time isn’t spent sifting through data and trying to parse out what’s important and what’s not, it frees you up to focus on the real issues – pulling out insights and adjusting strategy to better meet goals. Which leads me to my last big takeaway…

Change is hard. Just because you have access to more data doesn’t mean that you know how to use it overnight. And just because you have tools that leverage machine learning or other ways of making data useful doesn’t mean that you are out of a job. But it does mean that you and your teams have some work to do to change the way you work, to make the most of the tools and technology that are available.

Machine learning is going to help communicators do our jobs better. There’s no doubt about that. But the real question is – are you ready to roll up your sleeves and do the work?


This article was originally published on LinkedIn by Rob Oquendo, Spectrum EVP of Digital and Creative. Rob leads the The LAB, our group of digital and creative specialists, in leveraging digital media as an integral element in global strategic communications programs.

Like what you see?
let’s talk