In his book "The Power of Habit," Charles Duhigg writes about a three-part process that every habit undergoes: First, there’s a cue that tells your brain to go into an automatic mode and let the behavior take place; then, there’s the behavior itself (eating a cookie, checking your email, heading to Facebook), and finally, there’s the reward or the dopamine rush from the behavior itself. That’s the part of the process, Duhigg writes, that helps your brain remember the “habit loop” the next time you’re in the same situation.
I was reminded of Duhigg’s book when I subscribed to the New York Times’ Space Calendar, which is available through both Google Calendar and the iOS calendar and has 80,000 subscribers. Like many working people, I check my calendar throughout the day — it’s a habit — and now every so often, I see an upcoming space-related event that looks like this:
The Geostationary Operational Environmental Satellite-S is part of a series of satellites to improve the National Oceanic and Atmospheric Administration’s monitoring of weather and other environmental phenomena on Earth. See recent images of Earth taken by other GOES satellites here: http://nyti.ms/2D5a6w9
This is an incredibly smart strategy. First, it’s meeting and connecting with people in a place where they already have an existing habit — their calendar. Second, it’s priming them to associate an interest — space — with the coverage from The Times. Third, it’s sporadic enough to feel special, doesn’t feel as overwhelming as a push notification, and is a way for a news organization to really take ownership of a beat (and a platform.) I love it.
So I reached out to the team at The Times behind the Space Calendar to learn more. Below is our conversation.
How did you get the idea to create a calendar for space events and why did you start with the space vertical?
Michael Roston: The origin story for the Space Calendar starts with “The Simpsons,” as do so many good things.
The cable TV network FXX acquired the entire back catalog of the show, and announced that they would do a marathon of “Every Simpsons Ever” in 2014. I was a social media editor for The Times then, and assigned to work with the culture desk. We asked readers to write a blurb about their favorite episode, and we published a collection of our favorite replies. As I assembled this round-up, it occurred to me that it might help readers if they knew when FXX would air that episode, and get a reminder so they wouldn’t miss it. So I created a public Google calendar and linked to and embedded it within the Artsbeat blog post.
This was kind of a rogue effort, just something I added to a post I wrote without help from anyone else in the newsroom. No metrics or anything, I really don’t know whether anyone used it. But the idea stuck in my head as something I might want to try again. So many people spend so much time in their personal digital calendars every day, and it seemed like a zone where our journalism might fit, too.
The next year I moved to Science Times as an editor. As I started working on more space and astronomy stories, I came back to the calendar as an organizing concept for our department’s coverage. Ben [Koski, the director of the Interactive News team] and I first discussed the idea with some other editors on interactive news in 2016, but the head of steam wasn’t there. As 2017 got underway, and it became clear that the “Great American Eclipse” was going to be one of the biggest news events of the year, we returned to the idea and it seemed like the right moment. Clearly the eclipse was going to remind a lot of our readers how much they loved space and astronomy. So we got to work, and were fortunate to have Britt [Binler] and another developer, Sherman Hewitt, on board to find solutions for a lot of the technical questions.
There are many fixed, or at least semi-fixed dates for space. Eclipses and meteor showers happen on predictable dates. Rocket launches have windows that begin on announced dates. Spacecraft that are traveling around the solar system have flybys and landings that space agencies schedule down to the minute. Staying on top of all this stuff can be a challenge, so I thought it might help our readers who are interested in this subject area — not an insignificant number of people — to be informed about some of these events.
Online calendars are common across the space and astronomy web, and there’s a lot of great information out there produced by space agencies and space news websites. I hoped we could create something that was more interactive and very user-friendly. Space and astronomy aficionados might keep an eye on another calendar already, or more than one. But we’ve curated key events that I think more casual space fans will take an interest in, and where they’ll find value in getting an advanced notification that an event is coming. Maybe it’ll encourage them to look at the sky one night during a meteor shower, or tune into a livestream of a rocket launch or just learn about something wondrous that they might have forgotten about in in their busy lives.
I read that you’ve amassed 80,000 subscribers and are now expanding into a similar calendar for the Book Review. I’m curious how you’re thinking about what verticals to expand into next, based on what you’ve learned so far.
Britt Binler: The production of each calendar is driven by providing utility and creating a meaningful experience for subscribers. We avoid events connected to specific geographies in the Books Calendar. We don't set event notifications. We're experimenting with creating rich event descriptions that update over time with the recently published Election Calendar. Generally, we’re still discovering the limits and value of the format.
What has the reaction been from readers? Are they coming back? Is this helping with subscriptions?
Michael Roston: The reaction has been overwhelmingly positive, and we’ve also received some very useful feedback about events we should add to the calendar. And we definitely see many readers visiting the stories we link to in the calendar FROM the calendar, so I think that’s a sign that they really are looking at these entries in their digital calendars and engaging with them.
Because this began as a product dreamed up in the newsroom by an editor, I can’t say I started off asking whether this would help attract new subscribers. It’s more like I thought this would be a useful journalistic tool that would help us tell our stories and engage the audience that I knew was there. But I’m hopeful it contributes to the value of a Times subscription, the same way I find it valuable when my Netflix app sends me a push alert informing me that they’ve added a new show or movie that they think I’m interested in.
I’m curious how you created this from a technical standpoint. Was it manually entering entries into a calendar or are you using some kind of CSV?
Britt Binler: The calendars hinge on Google Calendar and Spreadsheet APIs.
For convenience, the Space and Books Calendars were initially populated using a spreadsheet of event data, including references to associated media where applicable. Desks now use a familiar interface – Google Calendar – to make changes by directly editing, creating or deleting events.
In order to more closely align with workflows between desks, however, we have developed software that can handle a variety of requirements. The Election Calendar, which was a collaboration with Graphics, relies on a spreadsheet as a source of truth. Calendar events are then programatically created and deleted.
A cron job updates calendar assets and periodically re-publishes to reflect changes on our site or in Google Calendar.
How many events is too many? Have you found that readers respond better to sporadic events?
Michael Roston: I’ve always wanted this to be not too obtrusive, and so in the Space Calendar I’ve aimed to keep this to two to five events per month. We don’t cover every rocket launch or every asteroid that enters planet Earth’s general neighborhood, so I wouldn’t want to put events on the calendar that we don’t see having significant news value. Even with that low number, I’ve had a few readers write and say that it was too much. So I think careful curation makes this a better experience for the readers we attract.
Britt Binler: We prioritize respect for the limited space available in readers’ personal calendars and revisit this question regularly.
I love this experiment, in part, because it’s creating a habit (and connection) to the Times through a platform that people use every day but don’t associate with the news. But I imagine there’s only so many “calendars” that someone would install overall. (Though, as with podcasts, there’s a niche audience for different kinds of verticals.) I’m curious what you’re learning from creating these niches, and how that feeds into the other products you’re building.
Ben Koski: One of our core challenges is connecting the right content to the right readers at the right time. Right now, our toolbox for doing so is somewhat constrained. We can use targeted push and email, but we have to be careful about these becoming too intrusive. We can make use of social platforms, but here our content is mediated by another organization. We of course have our website and native apps, but readers have to actively reach for these.
Calendars are interesting to explore not only because they offer a new channel for reaching readers, but also because the content we place there lives in an inherent context. We can place dozens of events well off into the future, and the calendar will do the hard work of introducing this to you in the right moment, in the context of time and everything else going on in your day.
One thing we’ve learned from the success of the Space Calendar is that we should keep seeking high-context opportunities. In some cases these might be experiments within new platforms, but these might also be experiments in better defining the frame for elemental push and email technologies.
We also want to do more to encourage a bi-directional relationship with these niches. Time and time again, our reader submission efforts have proven that relationships with these niches can generate valuable reporting and sourcing leads and valuable feedback about our coverage. There’s clearly an opportunity for us to better align niche content experiments like these calendars with community and newsroom relationship-building experiments.
What other platforms are you experimenting with for the Olympics?
Ben Koski: We have an exciting Olympics messaging pilot, building on the success of our experiments with SMS. In both the Rio 2016 and Trump Trip experiments, readers really enjoyed the opportunity to interact with an NYT staffer around the edges of mainline coverage.
SMS proved to be a fairly constrained platform, though. We didn’t have control over the sequence and structure of the stream. Media interactions varied widely from device to device. The cost of running SMS at scale meant that we had to limit the number and length of messages we sent and cap the size of the audience.
There was also a more fundamental issue with interactivity. With SMS, the most a reader could do is text us back, but it was hard to keep up with the volume of replies.
We wanted to create something that was more scalable, more interactive, and visually richer, and so decided to build our own interactive messaging space within the native NYT Android and iOS apps. Push messaging stands in for SMS — once you’ve opted in, you’ll receive broadcast messages (and in some cases, personal replies) from Sam Manchester via push.
These push messages will lead you into an interactive messaging stream designed by Troy Griggs and built by Isaac White and Paul Murray. We’re able to design sequences of messages that incorporate media and multiple-choice questions; this allows us to convey a sense of conversation, without leaning exclusively on Sam’s ability to make personal replies or some AI Bot. It allows a human interaction at scale.
If this model proves successful, it could be another way of serving these niche audiences. The “messenger” format lets us reframe push technology away from something generic and omniscient towards something conversational and human.
When someone has a really good idea for this — a Space Calendar — how does that get approved within The Times? I hear from a lot of product/tech people in newsrooms who have great ideas, and I’m curious what you can share about how this was championed internally and then actually built into production.
Michael Roston: Because the social media team used to be part of Interactive News, I have often worked with its editors over the last six years. That helped me find the on-ramp for this project, and Ben and company found the resources that were needed to work out the technology.
My boss, Celia Dugger, saw the value of this project, too, so that helped make it a science department priority. The three reporters who work most on space — Kenneth Chang, Dennis Overbye and Nicholas St. Fleur — all shared their expertise to make sure that it was a journalistically sound project, as well.
Britt Binler: We have both taken pitches as well as approached newsroom partners for collaboration.
After the Space Calendar launched, Pamela Paul shared her vision for the Books Calendar with us and has since championed the effort. Lovia Gyarkye, Greg Cowles and Rose Courteau contributed content.
We pitched the Election Calendar to Wilson Andrews in Graphics. There was simultaneous development on a similar type of project, and a subscription feature complemented the work.
Ben Koski: Within Interactive News, we actively edit away from one-offs and towards projects that work at scale. This project seemed like a great opportunity to experiment, since it scaled internally (easy to keep growing the audience over time) and horizontally (to other topic areas) without requiring tremendous resource investments from either Science or Interactive News.