For about six weeks in 2015, live streaming from a phone was the most exciting thing on the internet, and two apps were racing to own it.
Neither of them owns it now, and the reason is not that either lost. It is that live streaming stopped being a product and became a button.
The race, in brief
Meerkat arrived first and arrived loudly. It streamed live video from a phone and announced it to your Twitter followers, which is how it grew, because Twitter’s network was doing the distribution.
Periscope was acquired by Twitter before it launched publicly. So the platform whose network Meerkat depended on now owned a competitor.
Then Meerkat lost access to that distribution. The mechanism it had used to reach people was restricted.
Meerkat subsequently pivoted away from live streaming entirely into a different product, and Periscope ran for several years as Twitter’s live streaming app before being shut down, with live streaming folded into the main platform.
We are not going to give you the dates or the acquisition value, because none was confirmed against a primary source for this article and the figures for the acquisition in particular are widely repeated and rarely attributed.
The two accounts, and both are partly right
“Twitter killed Meerkat” is the popular account. The restriction of the distribution channel was decisive, and a small app that grows by piggybacking on somebody else’s network is an app that can be turned off by that network.
“Live streaming was never going to be a standalone product” is the corrective account, and it explains why Periscope also ended despite winning that fight.
Here is the thing that reconciles them, and it is the useful part.
Live streaming has no content library and no archive value. A live stream’s whole proposition is that it is happening now. When it is over, there is nothing to come back to, or nothing anyone comes back for.
Which means a live streaming app has no accumulated asset. It cannot build a catalog. Every day it starts from nothing and has to give people a reason to open it right now.
A feed can show you what happened while you were away. A live product cannot.
So the natural home of live streaming is inside something that has a reason to be opened anyway. A social feed, a video platform, a messaging app, a shopping app. That is where it went, and it went there everywhere at roughly the same time.
Live streaming did not fail
This is the correction the whole page exists for.
More live streaming happens now than in 2015 by an enormous margin. It is a standard feature of every major platform and an entire commercial category in some markets.
What failed was the idea that it was a destination. The behavior won and the apps lost, which is the same shape as several stories in this cluster.
It is exactly what happened in what happened to Vine: the short vertical video format did not die, it was absorbed by platforms with more distribution, and the company that proved it got nothing.
What survived
An archive of broadcasts was made available at Periscope’s shutdown, which is a better outcome than a simple deletion and is worth noting because it is unusual in this cluster.
We state no detail about its scope, because none was verified.
What did not survive is the ephemerality being the point. Live video is now recorded, replayed, clipped and distributed, which is a sensible product decision and is also the opposite of what these apps originally were.
Meerkat’s team went on to build something else entirely, which is a common and underdiscussed ending: the product dies, the people do not, and the next thing they build is frequently unrelated.
The pattern worth naming
A product built on top of another company’s distribution is a product that company can end.
Not by malice. By deciding what its own API and its own network are for, which is an ordinary business decision it is entitled to make.
That is the durable lesson from Meerkat specifically, and it is worth more than the story: anything whose growth depends on somebody else’s platform has a dependency it does not control, and the moment that platform has a competing product, the dependency becomes a decision somebody else makes about you.
The same structure appears in what happened to Google Reader from a different angle, where an entire ecosystem of third-party products was built on a service that was then discontinued.
What we are not going to publish
The acquisition value. Widely repeated, not verified for this page.
Dates for the acquisition, the pivot or the shutdown, none of which was confirmed against a primary source.
User, viewer or stream counts for either app.
A claim about intent behind the distribution restriction. The commercial logic is legible from the outside and does not require one.
FAQ
What happened to Periscope?
It was acquired by Twitter before its public launch, ran for several years as Twitter’s live streaming app, and was later shut down with live streaming folded into the main platform. An archive of broadcasts was made available at shutdown.
Why did Meerkat fail?
It grew by announcing streams to a user’s Twitter followers, and lost access to that distribution after Twitter acquired a competitor. It then pivoted away from live streaming into a different product.
Did live streaming fail?
No. Far more live streaming happens now than in 2015. What failed was the idea that it was a destination product rather than a feature inside something people already open.
Why can a live streaming app not survive on its own?
It accumulates no library. A stream’s value is that it is happening now, so there is nothing to come back to, and the app has to give people a reason to open it from scratch every day.
Are Periscope broadcasts still available?
An archive was made available at the shutdown. This page states no detail about its scope because none was verified at the time of writing.
This page publishes no acquisition value, no user, viewer or stream figures, and no dates for the acquisition, the pivot or the shutdown, because none was confirmed against a primary source at the time of writing. The sequence is described as a sequence instead.
