We’ve been delivering news over chat since 2014. Here’s what we’ve learned.

Everyone in the news industry is trying to decide what to do on chat platforms. We’ve run experiments since 2014 at Nyt.fi

Jussi Pullinen
6 min readMay 27, 2016

Want to deliver news over chat? Well, we’ve been doing that since 2014 and here’s what we’ve learned.

It was the autumn of 2014 and we were hitting some roadblocks.

We’re a Finnish media brand targeting a millennial audience and we were noticing our youngest readers were less and less active on Facebook, Twitter and even Instagram. The reason? They were using Whatsapp — LOTS of Whatsapp (over 70% of Finns under 25 use chat almost constantly, according to our research).

As young people were sharing less and less on these public or semi-public platforms, they spent more and more time on chat, sharing things with small closed circles of friends. But what role could our content play in a media environment like this?

To experiment, we started a Whatsapp newsletter.

It was mothing special, just a news digest, a few of our own stories and some curation from around the internet. We expected we’d get a few hundred early adopters to join and maybe we’d learn something from that. So we put out news on our site and social media that we were starting an experiment on Whatsapp, not expecting much.

We were overwhelmed. We got thousands of subscribers in a matter of days. Admittedly, being a part of Helsingin Sanomat, the biggest daily newspaper in Finland, helped. What surprised us, though, was how well our younger readers (not a core audience of Helsingin Sanomat) reacted to the service.

The reaction was overwhelmingly positive.

So we kept sending out the newsletter, and it evolved into sort of a daily take on what had happened “in the real world” and what was trending on the internet.

Here’s an example, nothing fancy, just plain text:

What also surprised us was the intensity of the relationship with the readers.

They were chatting directly to us, sending pics, messages, asking for advice on where to go out and get something to eat or drink (we cover the restaurant scene in metropolitan Helsinki area). We were added to private chat groups, mostly for laughs and sometimes for advice on bars, music or something else important.

So we talked back. If the users sent a selfie, we’d send one back. If they asked for advice, we tried to help. We tried running a photo competition with a chance of getting tickets to a local festival and got loads and loads of photos back. All in all, it felt very personal and very natural to be a media brand and to chat.

What kind of content worked?

We met with some of the subscribers and ran some polls as well as looking at general analytics data.

Based on this, one of the key elements of success was our editorial voice, which we’d honed for a year online before starting the experiment. People liked our enthusiastic, even chirpy, take on news and the world in general. They liked having a positive knowledgeable friend in their Whatsapp.

Secondly, they liked the digest format. Delivered in the afternoon after classes and/or work had ended, this was, according to them, a relaxing and an entertaining way to get briefed on the events of the day. Again, editorial voice matters.

Third, they loved the local stuff. People who were from the Helsinki region really liked getting tips on new restaurants or bars or info on events on the town via chat. They wouldn’t necessarily go and look that info up but this was a way to get it conveniently.

Four, people didn’t like clicking through to our site, even though they liked the format. So driving traffic was not going to be a viable business model.

Did it scale, you ask? Alas, no.

After a week or so we had 3000 subscribers (Helsinki metro area has about 1,1 million people) and soon 5000, mostly young, and we had to cap the number of subscribers at that number.

Whatsapp just wasn’t made for numbers like this. There were no publisher tools and you had to add and remove subscribers by hand. This was simply starting to take too long.

So we decided to build a beta app around the Whatsapp experience and scale that. And that is what we did.

In November 2015 we launched our own app, which is basically a chat platform for news (yes, we launched before Quartz announced their news chat app in February).

Here’s how our app works:

1. You don’t visit articles, you get messages just like our Whatsapp digest (there are links but we try to stay chat native):

2. There’s a main chat and some additional “rooms” or “chains” for you to check out. They’re centered on a theme like New music, Real News, Fun, All Things Trump and so on:

3. Everywhere, you can chat with us, just like on Whatsapp. Just type or send a photo and we’ll get back to you.

4. By using the app, you also get coupons to use at local businesses. This is the main way of monetization, instead of banner ads:

5. You only get a few notifications a day, for the digest, for our daily comic strip and major stories.

Does it work then?

User adoption has been slower here than on Whatsapp.

It’s more of a hassle to install a new app than to get something on existing platforms. We also haven’t nailed the first user experience yet. We’ve got many times more users now compared to Whatsapp, though.

So it does scale.

The relationship with users is also still super strong. We get a lot of messages, tons of photos and spontaneous thanks from our users. We turn those to stories on our website as well (we average around four hundred thousand weekly uniques in a country of 5,5M people).

Our main business is still our website, where we publish both original reporting (often on news topics with implications for young people or just plain funny stuff) and some curated content, for example from our parent newspaper.

So what kind of content works in chat?

Our three most popular things are the daily digest, the daily comic and the channel curating “real news”. In the curated channels people seem to have a less is more attitude: they want a digest, not a firehose. Photos, illustrations and comics seem to work well. Practical information, often local, is also popular as are everyday lifestyle topics and “lifehacks”.

So, what have we learned? Here’s a few bullet points to summarize:

  • Have a distinct editorial voice. People don’t want to chat with a dull institution — not even if it slaps emojis on its messages.
  • Have a distinct form, like a daily digest or a morning comic. Just don’t send random headlines as chat messages. That’s weird.
  • Offer a service. People love asking you questions. Answer them. If you go for scale, you might need bot help, but bots alone don’t cut it. Using only them puts your editorial voice in danger.
  • Build a community. Take what people send you and make something of it. Send it back to other users or make articles about it. People love getting noticed.
  • Don’t do just breaking news. No-one likes a chat friend that spits out wire flashes. Offer insight and context but format it as a message.
  • In general, be that awesome friend that always has great tips and helps you understand whats going on around you so that you don’t need to.

If you speak (or want to learn) Finnish, you can check out our app on the Finnish app store here and on Play store here. The app also won a gold medal at Inma global media awards this week.

Jussi Pullinen is a news editor at Nyt.fi, a site run by Helsingin Sanomat, a daily Finnish newspaper published in Helsinki.

--

--

Jussi Pullinen

Visiting professor in journalism (2017–2018) at Tampere Uni, News editor at Helsingin Sanomat. Finland. Digital media, digital news ecosystem and storytelling.