Melissa Davies

What’s the one thing you wish you’d known when you started in social listening?

Brand monitoring is probably the easiest entry-point to social listening because it’s something stakeholders understand and expect, but it’s often not the most insightful way to use this data. I wish I had known to push harder for some of the deeper insight use cases. 

‍

Have you had any experiences that have made you want to quit? What made you keep going?

Educating stakeholders and building trust in social listening as a source of insights is definitely a long process and can sometimes feel isolating. But the richness of social listening data and the promise of uncovering great insights keeps bringing me back. 

‍

What role does tech play in your social intelligence process? Where do people contribute?

Technology’s role in social listening is to gather, clean, organize and visualize the data; human intelligence provides the crucial overlay of interpretation that turns data into insight.

‍

Who have you seen as a mentor in your career?

I am grateful to so many people in the social intelligence community who provide informal mentoring by answering questions and sharing their great ideas and best practices – I think we are an amazingly connected, collaborative community, possibly because brand-side social listening can be a lonely job. Also, the head of Mondelez Insights & Analytics, Nick Graham, has an incredible vision for the power of unprompted/unstructured consumer data, and I’m lucky to be part of Nick’s team. 

‍

Most embarrassing mistake you made in a social listening project - what did you learn from it?

I once set up automated brand monitoring reports for distribution to a marketing team, and the first report contained a verbatim with a pornographic picture (even though it was also a legitimate brand mention). What I learned: automation can’t always be trusted!

‍

How do you see the future of social listening evolving?

Obviously the growth of AI and Gen AI will continue to drive changes in social listening, affecting accuracy, speed, data volume, and other unforeseen areas. I think we all hope that these changes will drive better, faster data automation, freeing up time for social insights professionals to focus on interpretation and recommendations. Integrating and triangulating different data sources continues to be a challenge due to the messy nature of social data, but the promise seems to be getting closer to reality. 

I also think humans’ relationship with digital/social media is evolving, with increasing focus on privacy and on the trade-offs of the amount of time we spend online. If social media use decreases as predicted, it will influence what we can measure through listening. Since I started in social listening, I have always seen people congregate into affinity groups – from Usenet groups in the early 2000s, to today’s Facebook groups and subReddits. So even as our time spent on digital/social media may shift, I predict there will still be a desire for building online connections around shared interest areas – and I hope there will still be a need to listen to these communities for consumer insights. 

‍

What’s the most useful data source? Are there any you find useless? Why?

I think usefulness/relevance of data sources is all about the question you’re trying to answer. eCommerce product reviews tend to provide detailed product feedback that we don’t see through general social media discussion. Reddit is great for understanding raw, consumer-to-consumer discussion topics. Google Trends helps with a longitudinal perspective on consumer interest in a particular topic over time. There are some sources that aren’t useful to me, but that’s more about relevance to our specific product categories and where we can find the consumer discussion we’re looking for. 

‍

How have you been able to win over ney-sayers throughout your career?

Keep evangelizing! Passion has a way of winning people over.