TicToc (틱톡)

» see tictoc.net as it lived, on the wayback machine

~ the obit ~

Start with the name, because it will otherwise sit in the way. This TicToc was a Korean mobile messenger that launched at the end of July 2011. A different company put out a short-video app with almost exactly the same name in China a few months after this one had already closed. The two are unrelated, and the one that owns the sound of those two syllables today arrived second.

Madsmart was incorporated in March 2011 and the app was live by the end of that July. Kim Chang-ha wrote it, having studied quantum and nuclear engineering at KAIST and then spent years building search, at Neowiz and then at Firstsnow, which NHN bought. Bon Angels put in 350 million won of seed money, and its partner Chang Byung-gyu said in print that if the classification had to be made, Kim belonged under genius, and might turn out to be a Mark Zuckerberg. By January 2012 the company was twelve people in an office near Seolleung station.

The product moved extremely fast. By the company's own count it reached ten million downloads inside five months, faster than KakaoTalk had got there. Twelve million by the middle of January 2012. Rankey's monthly active user figures put TicToc second in Korea in April 2012 at 3.92 million, ahead of Daum's MyPeople and a very long way behind a KakaoTalk already at 18.84 million.

~ the rap sheet ~

~ twelve servers ~

The reason it was fast is the only unusual engineering decision in the whole story. TicToc did not keep your messages. Message contents were never written down on the server; they were relayed and dropped. Kim told a reporter in January 2012 that this was why the whole service ran on about twelve machines while handling over six hundred million messages a day, and that rivals needed something like dozens of times more hardware to do the same work. The App Store tagline was a pun that translates to the toc arriving before the tic is out.

That decision aged into something the industry would spend the next decade marketing. By the end the Play Store description had stopped selling speed and started selling the same fact as privacy: chat history is deleted immediately after a message is sent, and user chat history is not stored on our server. The architecture never changed. Only the argument for it did.

~ the turkish campaign ~

This one actually came here, which puts it in a small category. TicToc ran a Turkish version of its site on its own domain with a Turkish notice board, and listed Turkish alongside Korean and English as a supported language. The Archive's URL index still holds the media buy: landing pages tagged to Milliyet and to Izlesene, and an event page built around the marketplace n11. Somebody in Seoul was paying for Turkish advertising in 2012 and 2013 to get this app onto Turkish phones.

It did not take. The messenger that mattered here went from MSN straight to WhatsApp with no Korean stop anywhere in between, and the name does not ring for me at all. The ad campaign is better evidence of TicToc's time in this country than anybody's memory of it.

~ bought, then stalled ~

On 2 April 2012 SK Planet announced it had bought one hundred percent of Madsmart's shares for cash, from the founder and the angel investors together. It was careful to say this was not an absorption: Madsmart would run as an independent subsidiary with its management and staff intact. A year later, on 23 April 2013, SK Planet resolved to absorb it anyway, effective 1 June. Madsmart had existed as a company for a little over two years. The price has never been published anywhere I can reach, so the figure that circulates is not going in this file.

The decline is visible in the same measurement series that recorded the peak. TicToc went from 3.92 million monthly users in April 2012 to 3.01 million in October. Newstomato's year-end piece put the blame squarely on the acquirer's operating strategy, described a big-company decision structure with low execution, and said the product had been caught between SK Telecom's carrier business on one side and SK Communications' NateOn on the other. An inews24 piece two weeks later used the phrase invisible checks between affiliates. The founder's argument for why twelve people could beat the conglomerates had been that big companies decide slowly and a small team can put everything into one product. That was eight months before the sale to one.

~ the notice in the chat window ~

There is no surviving shutdown page, because there was never really a shutdown page. The schedule went out on 5 April 2016 as a notice inside the app's chat window, and the English and Turkish notice boards on the website were already months and years stale by then. The schedule had three steps. On 5 April the announcement. On 12 April new downloads and new sign-ups were blocked. On 13 May, a Friday, the service ended.

You can watch the middle step happen from outside. The Google Play listing was captured alive every week for nearly three years and returns 200 for the last time on 6 April 2016. On 13 April it returns a 403 saying the app is unavailable in your country, which is what a delisting looks like from another continent. SK Planet never said why. The company had kept the thing alive for three years after swallowing the startup that made it, which is longer than most acquirers bother with, and then it ended it in a message that only existed inside the product.

The address is a parking lander now, 114 bytes of JavaScript redirecting to a page where the domain is for sale. Both app listings return nothing. The logo was a cyan square with a white speech bubble in it, and the last capture of the wordmark on the site was taken on 13 May 2016, the day the service went off.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »