1) Big brands are already listening
Marketers have long recognized the limitations of established demographic-based approaches to understanding audiences but, in the absence of any useful innovation in this area, they have stuck with them despite the flaws.
Separately, there is an ever-increasing requirement for evidence around the use of language in the creative process, but this is severely hindered by an inability to cope with the vast quantities of consumer language in any kind of meaningful way.
Ever tried reading everything written about a brand online – never mind trying to maintain some semblance of objectivity? It’s not even worth attempting due to the volume, so brands fall back on anecdote. They pick out interesting statements and hold them up as universally true. Or they segment the audience – divide and conquer – and pick out one or two to dig deeper on, even then only taking samples of the language.
Both methods are clearly dangerous and miss big pieces of the full picture. But how can brands quickly and objectively absorb, filter, and decipher the meaning behind the millions and millions of words written online by audiences? Even if they could, how can they quickly apply those learnings to find new audiences? This is where Audience Language Modelling comes in.
2) Recent technological breakthroughs have made it possible
Audience Language Modelling depends heavily on recent breakthroughs in mass-scale language processing, rounding on the principle that consumer interests group together outside of established demographic buckets.
The Star Wars franchise is a very good example of this: its audience is huge and can be found to cut across different ages, genders, and geographies. By looking at the language an audience uses, and not just when discussing the brand, we can better understand their other interests and create a linguistic model from which new potential fans can be matched.
For brands, these critical learnings inform and improve the creative process and provide the means for significantly more accurate media targeting. They give a deeper and more objective understanding of what consumers think.
3) Everyone in every team benefits
Universally speaking, Audience Language Modelling enables a brand to rapidly process consumer language without requiring the engagement of expensive and time-consuming external consulting services. However, one of the most useful aspects of Audience Language Modelling is that it benefits both the creative and media teams within brands. The approach informs the creative process without adversely impacting creativity itself, and lends an evidential base to creative decisions.
Media teams are able to significantly improve their campaign targeting by modelling interests and linguistic style from the existing audience, while also finding new audiences by matching them using the same modelling approach.
We are moving away from spamming consumers with adverts for a product because they fit an age or gender-based profile to a world where consumers are served ads they are genuinely interested in. It’s only made possible by ditching the demographics and listening to what audiences say.
Rich Wilson is chief marketing officer at Relative Insight