When a marketplace features many comparable suppliers, consumer preferences get harder to understand, presumably because while rational calculations among distinct offerings are converged upon, less-rational calculations among equivalents show diversities rivaling that of individual humans themselves (indeed because of those humans: less-rational calculations often sum subconscious associations, little-understood aesthetic preferences, impressions born of experience, and other widely variable elements). The saturated US "personal sharing network product" market may be something like this, especially for teenagers (to whom Facebook's compelling differentiator --that it has everyone-- does not sound too good).
That said, we can assume that this (specific) market of US teenagers prefers Snapchat to Facebook, Instagram, or Messenger for some combination of these kinds of reasons:
- Snapchat may have differentiating features or differentiating implementations of features (e.g. streaks; their specific frames; Map; etc.). I think there is some truth to this, but little. Most of Snapchat's features seem likelier to help with retention than adoption; some are inaccessible to new users.
- Snapchat may have a differentiating absence of features, too (no profile; no feed crowding into your sense of self or presence or whatever; no reactions; etc.) I think there may be some truth to this. Feeds are very public. Teens, if I recall it correctly, are private in various ways.
- Snapchat may have network advantages: if US teen usage has been declining at FB for several years, new teens continually compare heavily-used and -populated Snapchat with a FB that lacks e.g. their elder siblings, their older friends, their peers (and has their older family, their teachers, maybe their own baby photos!) Unlike Facebook, however, Snapchat's network has not evidently expanded beyond the contexts teen users want; that is, older siblings and even parents may be there, but (perhaps due to the aforementioned absence of feeds and profiles) it matters less. Or perhaps their day of reckoning is coming!
- Snapchat may have marketing/branding-based differentiation, which can happen in some markets. Because marketing of this sort is extremely costly, it formerly appeared that this would be so expensive as to guarantee Facebook eventual victory. But Snapchat might for whatever reasons --perhaps some of the above, and especially the potential network advantage-- simply have a strong, persistent brand edge over Facebook, rather like Coke over Pepsi. Of course: Coke now faces its own brand threats.
It's interesting to note that it may be reasonable to think about marketing in a different way for network-based products: once at scale, no marketing at any cost will make a larger impression than your existing user base. Any advertising spend, any effort imaginable, might pale in comparison to the daily impact of your existing users on how other users, new users, and potential users think about your product. Given how intensely network effects can shift information flow, it may even be the case that changes to your product cannot have more influence than your user base! Facebook and Instagram now have many of Snapchat's features, and Instagram in particular has had a lot of success with them, but the influence of previous users, their use-cases, their norms, and their histories persists, both in behavior and in historical interfaces like feeds, profiles, and the like.
- For teens, what's good about Facebook for us --everyone we know is on it; we get the most feedback and see the most stuff there-- is not desirable at all. They don't know all these old people! They're building new social networks themselves, tentatively, and they don't need interloper nodes.
- Thus: FB and IG have no network advantages over Snapchat. To the extent network matters at all, Snapchat may have more of their immediate demographic neighbors anyway: more of the good/relevant, less of the bad/irrelevant. Teens, as people building new social graphs in their lives, may have different networks preferences than adults.
- In addition, Snapchat may beat Facebook and Instagram for teens as a product, in part because they're both optimized for a user at a different stage of life. One common observation: Facebook feels designed for people with social connections in many places; it makes it easy for e.g. kids at college to keep up with people back home, etc. It may be hard for Facebook not to optimize over time for this kind of network preference, as doing so will generally yield good results, which may put them at odds with teens and their concentrated, novel, shifting graphs.
- And at last: in a market where no competitor has clearly advantageous differentiation --neither the features nor the networks confer anything clearly in this case-- brand can matter enormously. Teens may indeed be especially likely to seek an "other" brand, to differentiate themselves against older cohorts. (It's then a question whether Snapchat should prefer to somehow be a "perpetual teen" preference or to extend their features, e.g. with their Memories functionality, to support older and more "normative" use cases, which may eventually make them vulnerable to "the next Snapchat").
Whether Snapchat's brand reputation among teens reflects the facts of their product and network, the continuing "marketing effects" of their existing users, deliberate decisions made by the company in its marketing, good luck, or some combination, I couldn't say. It's also hard to say whether any of these companies can achieve serious differentiating advantage; whether Snapchat itself will "scale into" Facebook's problems or whether other new competitors will reveal a kind of "trendiness" to the social media space which makes all such companies susceptible to disruption.