While many of us still share hot takes and cat photos via public social media accounts, a growing number of conversations are moving to private places. Mark Zuckerberg himself noted the trend in a March 2021 post, saying, “Private messaging, ephemeral stories and small groups are by far the fastest growing areas of online communication.” Consider Discord, an instant messaging and digital distribution platform. The company is said to be nearing a $15 billion valuation as its user base has soared.
And it’s not just personal communication. Many of Part and Sum’s e-commerce clients have noticed customer interactions taking place in direct messages (DMs) or chat instead of public comment threads. So what can you do if your business relies on knowing exactly how people are communicating these days?
Holler, the leader in conversational media (think stickers and GIFs) came to Part and Sum with this exact quandary. In order to build new applications on the company’s natural language processing platform, Holler needed a look into these increasingly locked communication behaviors. Specifically, Holler wanted to understand cross-app sharing, where a link or screenshot travels from an app to a group chat and back again.
Asking for a report on private online habits rarely delivers the truth—nobody wants to admit how much time they really spend swiping through Tinder. So, how could we convince strangers that allowing us to read—and capture—their DMs was not only beneficial to them, but also privacy-safe?
Getting to why
We recruited a diverse group of 30 people who use peer-to-peer messaging apps, including Slack, Teams, Discord, Tinder and Bumble. Each participant was informed of the study’s purpose, how their information would be used and who would have access to it.
A Part and Sum strategist facilitated Zoom interviews, with participants joining on desktop and mobile. On desktop we could see each other clearly, and on mobile subjects screen-shared so we could watch them navigate and scroll through conversations. Having obtained permission in advance, we recorded our sessions.
We began the project interested in the types of content people share, and that angle shaped our initial questions. But the good stuff came when we stepped back and let people tell their own stories and dive down the rabbit holes of their conversations. We soon discovered it’s far more valuable to understand why people share—that is, what sparks that need in the first place, and what actions follow?
In order to make sense of such specific behavioral data, it’s important to review as you go and identify potential topics for follow-up. During weekly meetings, strategists briefed Holler on our findings and together we brainstormed new hypotheses to test.
Although it takes time (and client buy-in), this iterative approach is the best way to refine lines of questioning and arrive at more detailed conclusions. And ultimately, this process generated four key personas that provided the basis for prototyping.
One of the most important things we learned is that people want sharing to be seamless. The initial spark (“I gotta share this!”) dies if effort is involved, or if it’s hard to find shareable content in the first place.
With this in mind, Holler quickly developed prototypes of different in-message functionalities to put in front of study participants. Based on their feedback, we were able to outline concrete ideas for contexts like shopping, food and gifts. More importantly, however, we assembled a list of principles to guide future development, even as conversation platforms continue to evolve.
Final takeaways
Armed with our findings, Holler published a groundbreaking study about the use of media in communication and closed a $36 million funding round. But the lessons we learned during this unique study can be applied to many insight-gathering activities:
- Don’t just ask, observe. The adage, “Actions speak louder than words” is especially true when it comes to consumer behavior.
- When in doubt, ask “Why?” You may think you know what you want to know, but open-ended questions can take you in surprising (and valuable) directions.
- Research and review simultaneously. Don’t wait until all the data has been gathered to start reviewing it. Taking an iterative approach—gather, assess, react—will prevent important ideas from getting lost and help you identify dead ends before you go down them.
- Get concrete. Creating something to react to, even if it’s just a sketch or mock-up, helps people focus on details they might not have considered in a hypothetical exercise.
