Web Notifications, Part 4: Sequential

Conveying a set of facts with a thread of notifications.

Madeline Welsh
The Guardian Mobile Innovation Lab
4 min readJul 11, 2016

--

Photo by Alamy Stock Photo

We’ve defined sequential notifications as a series of notifications that creates a thread, where the user must follow a predetermined sequence, or click off the thread to get more information (to open an article, for example). In a sequential notification, all users follow one path, making it different from an interactive notification, where each user determines their own path through a series of options.

Like a top 10 list, the sequential notification is simple, direct, and makes the most of the user’s time. It directly answers the question, “what are the most important things that happened?”

A recent example: Following the June 7 presidential primaries we sent a series called “10 Primary Facts.” The entire series was strung together in a queue and the action buttons were used as a navigational tool, similar to an interactive notification. In “10 Primary Facts,” though, the action buttons directed the user to move through the series by pressing “Next”, or to leave the series to get more information from the live blog or in a relevant article, rather than to choose a path.

The initial alert introduced the series and gave an expected number of alerts in the title, and gave what we determined to be the most important fact. Action buttons directed the user to the next notification or to more information on the Guardian site.

The second alert continued the interaction.

And so on…

After the 10th fact, we sent a final notification to let the user they were “All caught up” and to direct them to the live blog.

We see sequential notifications as a great way to share round-ups with readers, or to format a summary of a topic.

In the future as we recruit more users, we want to test the idea that users will stick with the notifications if they are delivered on a subject the user has a particular interest in. This will help us test the theory that when notifications aren’t data driven and predictable, that users may find it engaging to receive a sequence of notifications about a topic that they have expressed interest in, rather than one that a news organization has sent out to their entire audience.

For those looking to experiment with sequential notifications, here is what we’ve learned so far.

  • Context means characters. It’s hard to write short. We’ve learned thus far that notification titles hold generally no more than 30 characters and body copy about 270 characters. Trying to summarize a complicated point in a single notification can be difficult.
  • Create a series idea, and label it clearly. Our goal has been to introduce some form and predictability into our experiments by creating clearly labeled series, so that users understand why they are receiving notifications. We do this in particular by having consistent series-related icons, and where possible putting the series name in the title.
  • Pull from existing content wherever possible. Drawing from and adapting existing Guardian content into the notifications format makes the editorial and production process infinitely faster than it would be otherwise. A newsroom looking to start using web notifications should first look to what they already produce that could be adapted from tweets, to key events, to summaries rather than starting from scratch.

More in this series about web notifications:

--

--

Madeline Welsh
The Guardian Mobile Innovation Lab

Editor @Google Earth | @GdnMobileLab @NiemanLab, @Studio20NYU, @CairoReview alum