Online-social conversations offer brands both challenges and opportunities. Ignore at your own risk, as online chatter can tear down a brand or make it altogether irrelevant. Harness the power of social media trend detection, however, and you are fast on the way to gaining competitive advantages.

But if ‘social media trend detection’ sounds too good to be true, you might be right. While exciting and potentially lucrative, deriving trends from social media conversations is new and still maturing.

For that reason, let’s focus our time on busting some myths and proposing some best practices.

Like we said, social trend prediction is new and challenging. Let’s debunk three myths and put you on the right track.

All Trends Follow a Bell-Shaped Path

Some trends do follow a bell-shaped path – they emerge, increase, peak and then fade. But many trends don’t follow that path. Veganism has been a trend since the 1970s, steadily increasing in its influence. Other trends, like interior design, swing back and forth, going in and out of style.

Our research at Ipsos suggests four types of trend lifecycles

All You Need Is Technology

Beware of monolithic technology that claims instant, full-proof trend prediction. Technology facilitates trend prediction, yes. But meaningful trend prediction – the kind that contextualizes and tailors trends according to your business – requires cultural analysis and human intelligence.

All You Need is ‘Social’ Data

This one may surprise you. Although trend prediction primarily requires social (conversational) data, it is incomplete if it does not consider behavioral data as well. This makes sense if you think about it. The ‘DIY’ trend may garner high conversation volume, but are everyday people putting their money where their mouth is, so to speak. We can validate the strength and influence of a trend by considering search data (intent), or sales and market share data.

Of course, there are other myths or misconceptions out there, but these are the main ones to avoid.

Blending Machine and Human Intelligence

As alluded to, proper social media trend prediction requires a mix of machine learning and human analysis.

Applied to Social Trend Prediction

• Machine Intelligence – Unsupervised, computational machine-learning models are an incredibly powerful way to model and decode the meaning behind emerging expressions. Algorithms, now readily available to us, not only cluster and size the topic landscape within a social data corpus, but also surface their contextual surroundings.
• Human Intelligence – Beyond mere intuition, human intelligence should incorporate the cultural experience and analytical and behavioral frameworks. Human analysis opens clear innovation lanes by mapping trends to underlying consumer motivations, usage occasions, and emotional and functional needs.

View Trends Holistically

Lastly, we encourage brands and organizations to look at trends holistically. In other words, what we usually view as ‘trends’ are actually interconnected ‘micro-trends’ that roll-up to a larger trend. For example, a rising interest in ‘Turmeric’ may more accurately represent a micro-trend that folds into a larger macro-trend – in this case, natural well-being.

The above concepts derive from our Research colleagues at Ipsos’ Social Intelligence Analytics, which Synthesio proudly powers with our technology. Our very own Chief Research Officer, Sandro Kaulartz, has popularized and perfected the above best practices, writing more about the topics here.

Let’s explain how the Synthesio offering works.

Machine Intelligence by Synthesio

With bottom-up topic modeling, Synthesio automatically categorizes and groups all of the potential conversation themes and trends within a data-set. In doing so, Synthesio democratizes the power of next-gen AI to non-technical business users. In a matter of minutes, you can analyze and detect various social media trends. Here, we look at the conversation themes surrounding CBD.

Human Intelligence and Advanced Analytics

Taking our own advice, we then pair machine intelligence with human intelligence and advanced analytics.

Our Social Intelligence Researchers run advanced semantic analysis to clarify the conversation themes driving specific trends. Back to the ‘turmeric’ example, our advanced analytics show the relationship between ‘turmeric’ and other properties such as ‘free radicals’ and ‘antioxidants’.

However, a clear innovation lane is not always apparent. We require human analysis to uncover consumer motivations and functional and emotional needs. Here, our team of 300 cultural experts lean on their expertise to provide actionable recommendations based on your business strategy.

With big promises comes the need for rigor and best practices. Social media trend prediction will mature as long as we bust the myths and stick to reality, together.