Contagious Media

What it is

In 2001, while a graduate student, Peretti emailed Nike asking the company to customise a trainer with the word sweatshop. When they refused he pressed further, requesting a colour snapshot of the ten-year-old Vietnamese girl who made his shoes. The exchange spread on its own, forwarded from inbox to inbox, and it reached far more people than anything Peretti had set out to reach.

The interesting part is what he did next. Rather than treat the episode as a fluke, he went to Eyebeam's research programme and ran it as a subject: build things designed to spread, watch which ones do, and work out why. That is the whole of contagious media as a concept. Virality becomes an engineering problem with a test bench.

Out of that work came viral lift, a measure of how far a piece of content travels beyond the people who were directly shown it. Tim Wu reports an instance of a post reaching many times its direct audience. That figure is a company metric reported in a trade history and this publication has not traced it to a primary source, so it belongs in an illustration rather than in an argument.

One timeline correction travels with this note. Peretti did not spend a decade at Eyebeam. He was there in the early 2000s, co-founded The Huffington Post in 2005, founded BuzzFeed in 2006 and moved to it full time in 2011.

In effect

The practitioner version has an academic counterpart, and the two disagree in a way worth keeping. Jonah Berger's work on sharing asks what makes people pass something on, and answers with properties of the content and the social situation around it. Peretti's question is narrower and more operational: what can be built, and what can be measured, so that the next thing spreads further than the last.

The disagreement surfaces at the point of commercial value. Berger's position is that spread is worthless to a brand unless the brand is load-bearing in the retelling, which is why an advertisement everybody remembers and nobody attributes has failed. Peretti's metrics do not distinguish those cases, because reach is reach. Evian's Roller Babies is the standard example that separates the two accounts.

For a reader, the practical consequence is that the industry which produces the things filling a feed is not guessing. It is running a measured optimisation, with the target set as onward transmission, and the emotional register of what spreads is a consequence of that target rather than an accident of taste.

What it does not say

It does not say that virality can be guaranteed. The method raises the hit rate across many attempts. It does not make any particular piece spread, and the laboratory framing is about volume and iteration rather than certainty.

It does not say that spread and effect are the same thing. Something can travel enormously and change nothing, and the counter-argument from the marketing side is precisely that.

It does not rest on evidence in the sense this publication usually means. Peretti's contribution is a working practice developed in a design programme and then scaled inside a media company, and the figures that circulate with it are company numbers. Claims about why content spreads should be taken to the research literature on sharing rather than to this concept.

---