A catchphrase I heard a few years ago—which didn’t catch fire, but still smolders (especially in Europe)—is temporal messaging and its close cousin, temporal rhythms. Simply put, when do consumers send messages to one another? What patterns do their messaging behaviors follow? And how can we use this data?
Why temporal messaging matters
Messaging is just one online behavior, but peak messaging times and patterns are vital to know and here’s why:
- Messaging is a very good proxy for other online activities. People who are online messaging are also shopping, watching streaming media, etc.
- Messaging is becoming device agnostic, and people can send or accept messages on a laptop, tablet, smartphone, etc.
- People message one another with word-of-mouth product recommendations.
- People message one another when it is convenient to do so, so are likely to share product recommendations within social media.
When is it best to reach them online? Well, it seems people fall into patterns and rhythms of messaging one another: temporal rhythms.
The best study of temporal messaging was performed in 2006 by HP Labs. They studied student use of Facebook messaging, and at the time, only college students had access to Facebook (and 90% of them used it). The title was nearly as long as the study itself—Rhythms of Social Interaction: Messaging Within a Massive Online Network. The researchers tracked 362 million messages exchanged by 4.2 million users over 26 months—an absolute mountain of data by 2006 standards.
It surprised the researchers just how consistent and predictable student behavior was. For example:
- Facebook messaging peaked during class time and study hours. That made sense—students used computers to socialize, study and exchange information about classes.
- Facebook messaging tanked on Friday and Saturday nights, as well it should. Presumably, students were socializing face-to-face during those periods.
- Weekend “rhythms” ran from mid-Friday to mid-Sunday, while weekday behavior ran the five days in between.
- Seasonal messaging behavior was pretty consistent, including over summer and winter breaks.
- About six in 10 messages were between students at their own school, the remainder to students at other schools. So, the word-of-mouth “reach” was very significant, both on-campus and on other campuses.
Behold the temporal database
Those findings were surprising and groundbreaking—eight years ago. But HP Labs promoted something very innovative at the time, which was the temporal database. It was (at the time) the latest dimension of online behavior and the least utilized. We knew already just who bought what, where and why they bought it, but now we could measure when—when they talked about it, when they researched it, when they purchased it and when brands could reach them online.
Temporal data would include:
- Peak online hours for whichever demographic you choose, be it college students, parents of college students, Hispanic consumers, single women between 18 and 34, etc. When are they online most? Least?
- Peak online hours for specific activities, like consuming streaming media, playing massive-multiplayer online games, online shopping.
- Peak hours for brick-and-mortar activity, like when consumers (choose the demographic) are in stores with their smartphones or mini-tablets.
This all sounds familiar
But, you say, we do know all of that. Yes, we do—or, at least, we know much of it. That does not mean taking advantage of it is a piece-of-cake. For temporal data to work, there must be sufficient data to make the necessary determinations on communications and applicable layered segmentation. Then, the trick is using it wisely, for example, to target someone in your demographic sweet spot enough times to intrigue rather than annoy. To target them in the hours before their typical brick-and-mortar shop, or when they’re typically relaxing with their devices at home. All temporal data.
I thought the term temporal database would catch fire—it didn’t. Instead, the phrase big data did, and temporal data was wrapped into it.
To summarize what temporal messaging is, it is the construct of hitting the right audience with the right message through the right channel—at the right time. The end goal is to achieve the best possible response.
Sure it does. It just wasn’t called that.
Do you track temporal messaging? How do you use temporal data to improve your reach?