Social Listening: How to Improve Your Farm Customer Insights
Two weeks ago, we explored one application of social listening platforms: social monitoring. This week, we’ll take a deeper look at another powerful way to use these tools: social listening.
Social listening is, at its core, research and analysis. It involves collecting broader data beyond brand or product-specific conversations, to uncover deeper insights. These can include customer insights, competitor analysis, audience analysis, and trends. Scalable and driven by AI, it can help you build better customer personas and refine your marketing communications.
Getting Started With Social Listening
Choose a social listening tool. Your first step involves choosing a tool, and while we could probably write an entire article focused purely on which tools exist, but there’s so many to choose from already available online. Here at WS, we’ve built some of our own tools, allowing us to better customize the ways in which we access and analyze the data we bring in.
Define your question or problem. Your next step will be in considering the questions you’re looking to answer. The more clearly defined your question, the better your raw results will be, so you’ll spend less time sorting out irrelevant results. Some examples of the types of questions you might ask are:
What does my customer really want from my product or service offering? What do they view as actual benefits?
What are farmers most excited about for the coming season?
Are there any trends involving the kinds of products or services I offer? Who is the most vocal about these trends?
What are my competitors saying when they talk about their products or services?
Do customers talk about my brand as often as they talk about my competitors?
Choose your data sources. Social listening can be the ultimate temptation, leading some marketers to think they need to listen everywhere. This most often leads to scores of irrelevant results or an overwhelming amount of data that needs to be analyzed.
Fortunately for agriculture brands, one of the best places to pull social listening data from is Twitter. We know farmers are there, engaged in conversations about their operations, connecting with brands, and each other.
Other data sources include blogs and articles on the web, forums, and other social platforms, though they tend to be more limited based on the various permissions allowed by each.
Build your queries. This step can often feel the most cumbersome. You’ll now need to list the keywords, phrases, topics, themes, and hashtags you want to listen for. You’ll also create a list of words you wish to exclude. For example, if you’re listening for the word “crop”, you’ll probably want to exclude terms that indicate the actual conversation is about Photoshop or photography “cropping”.
In addition, you can use Boolean search queries to set parameters about how different keywords and phrases appear together.. For example, you may want to listen for conversations where the user mentions “crop” AND the #CdnAg hashtag. This will exclude all other conversations that don’t have both of those keywords.
Refine your queries. As with most digital tactics, social listening is never one and done. You will need to consistently review your results, refine your query, add new keywords and phrases, or exclude keywords that you see coming up frequently.
Analyzing Social Listening Results
Once your data has been collected, it’s time to analyze the results. The methods you use in your analysis should directly reflect the question you’re trying to answer.
There are many ways you can look at the data you pull in. Here are just a few:
Sentiment Analysis. Sentiment analysis is a natural language processing technique that determines whether data is positive, negative or neutral. This analysis can help you monitor brand or product sentiment, as well as better understand customer needs.
Term Frequency Analysis. This type of analysis can help you understand the words your audience is using most often. There are drawbacks to this method. It removes context from the words by breaking sentences down into individual words. Frequently used words (like “the”, “it”, “is”) also rise to the top in frequency, and can appear “more important” than words that actually provide context and meaning to your question. With these drawbacks in mind, term frequency analysis can help you identify possible content topics you should be addressing in your marketing strategy or connections between ideas that you hadn’t previously considered.
Share of Voice Analysis. By monitoring your brand or product name and the brand or product names of your competitors, you can perform a share of voice analysis. This can help you compare how often your customers mention your brand or products against how often they mention your competitors.
Always Validate Your Insights
Like any type of data you collect, it’s important to validate your insights against other data sources. This is particularly true of agriculture-related data – there’s often not enough of it to be statistically significant, and it’s hard to project insights onto an entire audience based only on the most vocal members who tweet the most often.