As a junior, one of my roles was to cut out the newspaper clippings, paste them onto a piece of paper and fax or post them to clients. Skills honed in primary school involving scissors, glue and paper were used to maximum effect. Volume of coverage was literally measured by the inch. Occasionally we were asked to analyse the coverage, and provide perspectives on what really mattered to our clients; how it impacted their business or contributed to achieving their business goals.
Today, given the sheer volume of content available, no human being could possibly keep track of all the ins and outs of how topics are being reported on, talked about and evolving in live time.
Nearly half a million tweets are sent out every minute of the day. There is a constant risk of missing something important that is drowned out by irrelevant noise; the same noise can also drown out your own message.
More than ever, our clients need to understand how topics of relevance to them are evolving, how their business is impacted and crucially, how conversations will evolve going forward. They need to be able to understand if they should engage, and if so, when and how.
Analytics allow us to interpret huge amounts of data, but this can be overwhelming. So the key is knowing what to look for. Smart use of analytics tools and machine learning allows us to understand which variables really have an impact on our clients’ abilities to achieve their business goals. We are able to assess how topics emerge and evolve on traditional and social media, which ones resonate and how that influences behavior.
Most analytics focus on what has happened in the past. That is helpful – it is important to establish baselines, to evaluate the impact of our clients’ and competitors’ activities and gain insights that can help inform strategies. But how can we provide more certainty about what will happen in the future in today’s complex, noisy media landscape. Here, machine learning combined with traditional statistical analysis can help us read huge amounts of data and identify trends and patterns from which we can draw insights – the topics people will engage on, likely trends in those topics, when is there typically high or low levels of interest around a specific topic. In this way we can track topics and use predictive analytics to inform plans and create forward-looking content.
Machine learning can help us analyse mountains of content and digest it down to a manageable level of data that we can interpret and use to shape and improve our clients’ content plans. Predictive analytics can help us develop content plans that engage with people at the right times and amplify our engagement around our most important opportunities.
Does this mean we can soon replace communications consultants with machines? I don’t think so. We still need to be able to ask the right questions, interpret the data and shape strategy. But there is a place for people working closely with machine learning to provide smarter advice, more targeted plans and demonstrate how we are using our clients’ budgets wisely to deliver against their business goals.