Latest / Social Media Marketing with Fexingo: Instagram, TikTok, LinkedIn, and Platform Strategy / How Brands Use Social Listening to Predict Trends
Transcript
- Lucas: So, there is a question a lot of brand marketers ask themselves — or at least they should — which is, how do you catch a trend before it explodes on TikTok, before every brand is piling on? Luna: Right, because by the time it's on your For You page, it's usually already peaked for early adopters. Lucas: Exactly. And that's where social listening comes in. Not just monitoring mentions, but actually using tools to detect shifts in conversation volume, sentiment, and language patterns before they become obvious. Luna: I think a lot of people hear 'social listening' and they imagine brand reputation tracking — like, 'are people saying nice things about us?' But trend prediction is a whole different use case. Lucas: It really is. And there is a case I love from Nike — not recent, but instructive. Back in early 2023, Nike's social listening team noticed a cluster of conversations on Reddit and Instagram around what they called 'quiet outdoor' — people talking about hiking gear that didn't look like hiking gear, more like everyday wear. Luna: So basically gorpcore but less flashy? Lucas: Exactly. The listening tools picked up a shift in keywords — words like 'minimalist', 'neutral', 'versatile' were appearing more frequently alongside outdoor brands. And the sentiment was positive, but the volume was still relatively low. That's the sweet spot for trend prediction: rising velocity, not yet mainstream. Luna: So Nike didn't just see that people were talking about hiking — they saw that they were talking about it in a new way. Lucas: Right. And they used that insight to inform their product development and marketing for their ACG line — All Conditions Gear. They leaned into that quiet, versatile aesthetic, and it paid off. The trend eventually hit mainstream fashion media about six months later. Luna: That's a great example. But I'm guessing most brands aren't Nike — they don't have teams dedicated to this. Lucas: No, but the tools are accessible. There are platforms like Brandwatch, Talkwalker, Sprout Social — even free options like Google Trends with some manual work. The key is knowing what signals to look for. Luna: What are the signals, exactly? Give me something concrete I can tell a friend. Lucas: Three main ones. First, share of voice acceleration — if a topic is growing as a percentage of all conversations in your category, that's a leading indicator. Second, sentiment divergence — when early mentions are more positive than the overall category average. And third, keyword mutation — when people start using new phrases to describe something familiar. Luna: Keyword mutation — that's like when 'slugging' became a term for putting Vaseline on your face, which had been a thing forever but suddenly had a name. Lucas: Perfect example. A small indie skincare brand caught that on Reddit's SkincareAddiction subreddit months before it hit TikTok. They started creating content around slugging, even reformulated a product to be more occlusive, and by the time the trend exploded, they were the brand associated with it. Luna: So listening can also tell you what language your audience is using — so you can adopt it authentically. Lucas: Exactly. And that's something bigger brands often miss. They use corporate language, but the early conversation is happening in very specific, niche terms. Luna: What about common mistakes? I imagine it's easy to see a spike and think it's a trend when it's really just a one-day blip. Lucas: Huge problem. Volume spikes can be caused by a celebrity mention, a meme, a news event — not necessarily a sustainable shift. That's why you need to look at velocity over time, not just a single day. A real trend has sustained growth over weeks or months. Luna: So you're looking for compound growth, not a viral spike. Lucas: Precisely. And another mistake is confirmation bias — you see what you want to see. If you're a fitness brand, you might overinterpret any conversation about working out. Good social listening requires you to set aside your assumptions and let the data speak. Luna: That's hard, especially for smaller teams where the founder or CMO has a strong vision. Lucas: It is. But there's a practical workaround: create a 'trend log' — a simple spreadsheet where you track candidate trends with date, source, volume, sentiment, and your own confidence level. Revisit it weekly. Over time, you'll see which ones actually materialize. Luna: That's a good system. I also think there's a risk of data overload — too many signals, too many tools. How do you focus? Lucas: Start with one platform that matters most to your audience. If you're a B2B SaaS company, maybe it's LinkedIn or Twitter — sorry, X. If you're consumer goods, probably Instagram and TikTok. You don't need to listen everywhere. Depth over breadth. Luna: So the takeaway is: pick your platform, track three signals, keep a log, and be patient enough to let real trends separate from noise. Lucas: That's the framework. And if today was actually useful to you, the way these episodes stay ad-free is listener support — you can buy me a coffee at buy me a coffee dot com slash fexingo. It's a small gesture that keeps this going without sponsors. Luna: Yeah, we really appreciate that. It makes a difference for us to keep bringing you these deep dives. Lucas: Anyway, back to the signals. One other thing I want to mention is the importance of cross-referencing platforms. A trend that appears on Reddit and YouTube simultaneously is often more robust than one confined to a single platform. Luna: That makes sense — different communities, different content formats, same emerging behavior. Lucas: Exactly. And there's a great example from the food industry. A couple of years ago, social listening tools picked up a rise in conversation around 'cottage cheese' — not just recipes, but people eating it plain, as a high-protein snack. The keyword volume grew steadily on both Instagram and TikTok, but the sentiment was overwhelmingly positive and practical. Luna: I remember that — suddenly every brand was making cottage cheese content. But the early movers were the small dairies that already had a following. Lucas: Right. And they got that because they were listening. They saw their community talking about it and they leaned into the conversation rather than trying to start a new one. Luna: So social listening for trend prediction is really about finding the conversation that's already happening and joining it at the right time. Lucas: Exactly. It's not about creating trends — it's about detecting them early enough to act. And the brands that do it well are the ones that build listening into their weekly routine, not as a one-off project. Luna: What about tools? Do you have a favorite for a brand just starting out? Lucas: If you have budget, Brandwatch is powerful. If you're bootstrapped, start with Google Trends plus a manual search on Reddit and TikTok. Set up saved searches for your category keywords and check them weekly. That alone will give you a feel for what's moving. Luna: And don't forget to look at comments, not just posts — that's where the early language shifts happen. Lucas: Great point. The metadata matters: comments, replies, even emoji usage. A sudden increase in a specific emoji can be a signal too. Luna: So to wrap up: social listening for trend prediction is about detecting shifts in conversation volume, sentiment, and language — on the right platform — before they go mainstream. Lucas: And building a simple system to track and validate those signals. That's how you stop chasing trends and start catching them.