Why Nothing Ever Seems to Go Viral

As Hit Makers author Derek Thompson explained in an interview with Forbes, for something to go viral means the product or idea in question “spread through many generations of intimate sharing.” People who liked it shared it with other people, who in their turn shared it with other people, who in turn shared it with other people, again and again in short sequence, rapidly multiplying the number who saw it (and shared it) in a manner analogous to the spread of an epidemic from Patient Zero.

Of course, for anyone looking for an audience — and who isn’t? — the prospect of such success is enormously attractive. But it also turns out that, as Thompson himself remarks (and as many a frustrated marketer has doubtless come to suspect), it almost never happens.

When I consider how social media works this seems to me the more obvious. Take Twitter, for example, which makes certain data conveniently available to everyone with an account — among that data, the number of “impressions” they made in a given period, which is to say, the number of times the item appears on screen in front of a follower, whether or not they look at.

Based on the numbers I have seen (not just my own) someone who Tweets something can expect impressions from it equivalent to 1–3 percent of their followers. Someone with a thousand followers, for example, might expect to score ten to thirty impressions per Tweet.

Not an impressive figure, is it?

And remember, impressions are just that, a thing far short of the deeper concept of “engagement,” about which Twitter also provides figures to its users. This is the Twitter user interacting with the Tweet some way — if only by clicking on it.

As it happens, the rates of engagement-per-impression are not very high. An engagement rate of 0.9 percent — in other words, a less than one in a hundred chance of an impression leading to engagement — is considered a good, solid rate.

As this includes such things as accidental clicks, or multiple engagements by the same person (who might click on it, and like it, and Retweet it), one may safely assume that the odds of any one person really taking an interest in any single Tweet presented them is lower.

So 0.9 percent, of 1–3 percent, and you might get one engagement for every ten thousand followers you have, with the more substantive engagements, like likes and Retweets, apt to be a fraction of that. One in several tens of thousands.

Ninety-nine percent of Twitter users have 3,000 followers or less. Far too few for anyone to expect that their own followers will engage any one Tweet in a meaningful way, let alone Retweet the Tweet. Apparently 0.1 percent have 25,000 followers or more, enough to have a reasonable chance of one of their Tweets getting a Retweet. Still, only 0.1 percent of their followers has a similarly sized following (25 people). Given that only one in tens of thousands might Retweet, the odds that one of their followers with a large following of their own will Retweet any of their Retweets are very slim indeed.

The result is that the chances of any one Tweet getting a Retweet, and still more a second Retweet, are minuscule. The chances of one Retweet leading to another leading to another Retweet, again and again, in line with the “going viral” model, are a minuscule fraction of that.

The figures of which I write here are averages. Of course, averages contain wide divergences. Some will do much better — but others will do much worse. And it is worth remembering that the bigger one’s following, the lower the engagement rate they are likely to have, for various reasons, offsetting at least somewhat the benefit of their bigger following from the standpoint of the Tweet’s wider propagation.

On top of that some argue that the continuing fragmentation of the media — social media most certainly included — is making anything’s going viral even less likely than before. But it was never all that plausible a thing to hope for in the first place.

Originally published at https://raritania.blogspot.com.