Beyond the mainstream: How to get quality insights from niche groups

Data access is a major concern in the social intelligence landscape. While many professionals rely on mainstream social networks such as Twitter, Facebook, and Instagram there are plenty of other data sources you can use to generate insights. Niche groups and online forums, in particular, can provide you with much richer insights about your customers.

During Demo Day 2022 we discussed this topic with Frank Gregory, Senior Social Media Operations & Research Analyst for USAA, and Imre Nyaka, Social Media Listening Professional at Ericsson.

Understanding data sources in social listening

Social listening tools extract conversational data from major social networks such as Twitter, Instagram, and Facebook. However, it’s important to note that social listening isn’t social media monitoring - looking at conversation volumes on these social networks alone doesn’t offer actionable insights. “There’s so much more we can do with what we have,” explained Imre.

As such, social listening isn’t just about social media. “You can pick any keyword, hashtag, or topic and see that across a forum you might not even know existed or a review site that’s more niche and you don’t know about,” added Frank.

Identifying additional data sources

Social listening tools are very keyword-focused and don’t provide the context of the complete conversation. This can limit your social listening efforts. Imre recommends naïve reading as a solution - having empathy towards people and understanding why they’re behaving a certain way. Think about your target audience and what kind of information they find if they’re looking for certain keywords. “Then you might be able to identify certain niche forums that the Google search engine is giving them,” he explains.

Sometimes you need to identify other data sources than the ones covered by social listening tools. Frank recommends partnering with your SEO and Voice of the Customer teams to identify other data sources the organization uses to track consumer conversation or consumer search. He gave an example of his work with their client in the motor oil space–Pennzoil.

They initially worked on the assumption that most conversations were happening on social. However, after running the analysis, they found the source with the second highest results (after Twitter) was forums. In particular, a forum called “Bob Is The Oil Guy,” one they’d never heard of before.

The forum was for car and motor oil enthusiasts, and they were directly comparing Pennzoil to other brands. This group had some influence and it ended up being a great source of insights. It also provided an opportunity to engage the forum in real-time by setting up “Ask me anything” sessions.

‍

Different personalities for different social networking sites

It’s important to note that different social networking sites are associated with different personalities, which will define the type of insights you can extract from each data source.

For example, Imre compared Twitter to a town square, where most of the activism and spontaneous conversations take place. Meanwhile, Reddit is more like a coffee house. It supports long-format comments centered around topics. “You can exchange much more complex ideas than on Twitter,” he explained. “These niche forums and Reddit are a place for people to gather and exchange these complex ideas and create trends, and then that will go into Twitter.”

Even in Frank’s Pennzoil example, most of the Twitter conversations revolved around the brand’s NASCAR sponsorship. Meanwhile, the forum dealt with very technical conversations.

Niche forums and sites are useful for conducting fandom research, to find fan art and stories. Within the mainstream social channels, think about every type of hashtag that might come up and look through the associated conversations. This might include some of the top hashtags or more nuanced ones that are only being used by superfans.

Overcoming data skewing

Due to the difference in the types of conversations and personalities found across different platforms, it could be easy to end up with skewed insights if you’re not careful with what to look for and where.

“When we do keyword searches and look for specific mentions, what we’re pulling in is only a small fraction of the conversation,” Frank explained. These searches only provide data on the use of that keyword, phrase, or hashtag so you’re only pulling a small, out-of-context portion of the conversation. And you really have to go deeper to avoid skewing your insights.

“Maybe that keyword is the trigger but then you’ve got to actually read through the Reddit thread to get the context in order to provide the right insight,” he elaborated. “If one Tweet has that keyword and there are 50 replies, you have to read those replies to be able to understand what’s being discussed.”

“The conversation themes are potentially going to be very different depending on the data source. But regardless of the data source, we need to find a way as an industry to be able to pull in those conversations to get those holistic insights as opposed to just the keyword.”

Looking at data access in the future

With the risk of mainstream platforms restricting data access at any time, it’s important to diversify your data sources to prepare for the future. Imre recommends monitoring niche platforms to explore new opportunities and find the next big platform. Frank recommends using forums where you can find much richer data.

‍