Leveraging Conversation Mining for UX Writing Research
As a ’90s kid from America, when I think of mining I usually picture the seven dwarves from Snow White going to work in the diamond mines, extracting perfectly cut shiny gems from the walls and pouring them into mining cars. (It’s just that easy, right?!) As an adult, I know that mining is a little bit more dangerous, dirty, and involved than that. Thankfully, not all mining involves pickaxes, helmets, or heavy machinery. Or dwarves.
The term ‘conversation mining’, coined by the UX Writing Hub in 2018*, is a simple concept really — build a product that speaks to users in their own language through strategic research of real-life conversations. Strong UX writing at its core allows the user to easily accomplish a goal with as little stress as possible. So it makes sense that in order to communicate effectively to the user how they should navigate through the product, we as UX writers need to learn to use language they understand. How do we do this? By getting creative with our research techniques
What is Conversation Mining?
Conversation mining is a vital piece of UX writing research that involves delving into the online communities where a product’s target audience resides.
The intention of conversation mining is to put ourselves in the places where our target audience is freely expressing themselves in order to glean from them the different words, phrases, terminology, inflections, and more that allow us to speak to them at their level. After all, a digital experience is, in fact, a conversation between the product’s voice and its users, and we should have a common vocabulary.
For example, if you were working for a subscription site like Cat Lady Box that sells to women 25–34 who are cat owners, you would want to get into the mind of that target audience. When you begin to explore the posts that people have made on social media with the hashtag #catladyboxday, you find many posts that also have the hashtag #ififitsisits.
So, on the site instead of writing “It’s a cat lady’s dream come true! Subscribe today.” you can try using, “If I fits, I sits! Subscribe today.”
Where to mine
The first thing we need to do is figure out where to go to collect the data. Social media is a great resource for conversation mining, but it’s not the only one.
- Facebook groups
You can find a Facebook group on just about every topic, no matter how obscure. Once you join a group, you can search within that group by topic or keyword to find something relevant to your specific needs, or just linger long enough to notice the trends. See which comments garner a larger number of responses, both positive and negative.
- Online reviews
Sometimes a great source of data can be the feedback from customers. Here is where we can learn a lot about not just language but specific pain points or goals of our users. Feedback can help influence changes in the UX/UI of the product as a whole.
Anyone who’s familiar with Reddit knows that its users hold nothing back. You’re sure to get some unfiltered opinions and comments by searching for specific subreddits relevant to your audience. For example, if you’re designing a product that sells winter sports equipment, you can find a subreddit for snowboarding.
- Instagram, TikTok, Twitter, etc.
A great way to utilize the gram is to find out who your target market’s influencers are and go find them on social media. Working for a cosmetics company? DIY make-up tutorials are huge on TikTok. Follow some of the top influencers and observe the dialogue. Hashtags can also not only help you find your way, but you can incorporate those popular hashtags into your UX.
- Call the center or help request records
In addition to online reviews, a great source for mining might also be your company’s help requesting records or transcripts from the call center. Be sure to reach out to your customer service team to see if you can access these records.
Online forums still exist and they can be a wealth of knowledge for specific audiences.
- Personal blogs
Authorities in the space or even your very opinionated normal person can be a helpful source of inspiration as well.
Tips and best practices
There’s no real strict guideline on how to conduct your conversation mining research, but here are some methods and techniques that can help guide you.
- Go where they are
You don’t need to hit every single category above in order to conduct effective conversation mining. Identify which ones are the most relevant to your audience and your goals and stick to those.
- Organize in Airtable
While you can use any system you like to keep your data organized and sharable, we recommend Airtable. It allows you to categorize and sort each data point by mood, for example as positive, negative, or neutral.
- Explore user’s goals and pain points
As you’ve probably gathered, conversation mining isn’t just a great tool for collecting verbiage, it’s also a great way to have your finger on the pulse of your audience in order to gauge how they think or feel about your product or their related goals. Keep track of these opinions in your spreadsheet. This can allow you to see opportunities to improve their experience without them even needing to ask for it
- Build or refine your user personas
You can get an idea for where to look online for your audience by referring to your user personas if you’ve already built them, or your findings can influence the creation or refinement of those personas. (Or both! Yay, iteration.)
Speaking of iteration, conversation mining can be done at any point in the process, not just in the initial research phase. If it’s an hour before the deadline and you’re struggling with a piece of microcopy that’s not really hitting the mark, go mining for inspiration.
- See what’s popular
Good UX writing isn’t always about finding something snappy or different. It’s about identifying what resonates. A good way to measure conversation that lands is by observing how big a reaction a comment gets, or how many retweets or shares of a post there are.
As you can probably tell, conversation mining is one of those things that isn’t an exact science. It’s important to get creative with it and use your imagination depending on your project’s unique attributes. And just like with most things, the more you practice it the better you get.