Gradient Handbook / Trendlines / Writing Newsletter Stories Edit this page

Writing Newsletter Stories

Introduction

Writing a newsletter story requires you to be creative, precise, analytical, and at least a little bit funny. It is not easy, and you should plan on a story taking a couple of hours to write.

We write two standard stories and one TrendSegments story per edition, and run two editions per month. The goal is to present interesting analysis with sharp visualizations in an amusing way. We want to simultaneously keep readers’ attention and show off our analytical skills on topics serious, silly, and everything in between.

Timeline

We run two newsletters per month. The first usually launches during the first week of the month, and the second launches two weeks later. During an ideal story writing and launch week, the timeline is as follows (we will be covering the bolded portions in this section):

  • Monday: Pick stories to run and begin plot specs/writing stories
  • Tuesday: Write stories
  • Wednesday: Whiteboarding, email the newsletter, and upload to the website

Picking Stories to Run

This may be an obvious step in the process, but we need to be mindful of which stories to run during which edition of Trendlines. For each story, we should consider whether the data will be stale if we don’t run it in the first edition, or if it makes sense to wait until the second edition to run a story (e.g. waiting until closer to Halloween to run a ghost story). We should also keep in mind how the stories will mesh with each other in each edition. If we have two non-time-sensitive sports stories, for example, it may be best to publish them in separate editions.

Plot Specs

When beginning with writing a newsletter story, the first thing to do is consider which data visualizations you want to include in the story. Sometimes this is obvious (e.g. if your story includes a MaxDiff, chances are you will want to include our standard visualization of MaxDiff results). Other times, you may want to look for particularly interesting crosstabs on Crunch – usually, beginning by testing yours (or the module conceiver’s) pre-hoc hypotheses.

The goal is to find an interesting hook for the story. If everyone has done a good job in the module analysis spec phase this part can be very easy. Even still, sometimes results can be unexpected and this process may take a bit of time.

Once you’ve found the hook(s) for the story, it’s time to specify what you would like the plots to look like. The reason we do this first, in addition to giving a clear direction for the story, is so that the person assigned to plots that week can get started as early as possible.

A good plot spec will include:

  • Type of plot (e.g. horizontal bar graph)
  • Title of the plot
  • Which specific variables should be on the X/Y axes
  • X-axis title (if any)
  • Y-axis title (if any)
  • Grouping variables and (if necessary) associated recoding instructions
  • Anything else you want to convey on the plot, including colors, filtering by certain variables, etc.

Plot specs are usually written in a comment in the newsletter Google Doc. Be sure to tag the person responsible for plots, and it never hurts to also update the responsible party on the Plot Creation JIRA issue.

Writing a Story

Just like in the ideas phase, writing the story is a creative process that doesn’t have one prescribed right or wrong answer. If you’ve never written a newsletter story before, it can be useful to check out the archive of past Trendlines articles to get a feel for the Trendlines “voice.”

In short, the voice is snarky, witty, quippy, and fun, yet focused on presenting interesting insights. The ratio between discussing the results and making jokes about said results is about 1:1 most of the time. To be sure though, some topics require a much more serious tone.

Some general notes to keep in mind while writing:

  • Use present tense
  • Try to avoid referring to “we” and “us” as much as possible
  • Highlight particularly interesting findings in bold and classic Gradient Green (#50bd90)
  • The stories should be somewhere between 250-350 words
  • There shouldn’t be more than two plots in a single story (memes/gifs don’t count as plots and can sometimes be fun to include)
  • On a first draft, keep in mind that this will go through a couple of rounds of revisions so it doesn’t need to be perfect
  • Have fun with it!

It is important to note that the writing should take place in the Google Doc associated with that run of Trendlines, which should be in the same Drive folder as that month’s newsletter instrument (Brand > Trendlines > Surveys > [YEAR] Surveys and Syntheses > [yyyy-mm FOLDER]). The title of the document should use the typical Gradient naming convention: yyyy-mm-dd | [1-3 word summary of each story].

Writing a Product Story

The elements of a product story are no different than a Trendlines story, other than elevating what the implications—and value—of the findings are to a marketer.

Here are some guidelines:

  • Keep the story concise and focused on one or two key findings of the analysis.
  • Consider the questions a brand manager/marketer/B2C product manager would ask to better understand their audience. Clearly explain the business case for the methodology, as well as the outcomes of the analysis. For example,
    • “Imagine you’re a brand manager; the question keeping you awake at night has to be, ‘How do we market our brand to increase its value?’ or ‘How do I know my marketing is actually working?’”
    • Note: Doesn’t have to explicitly call out the specific role, but can.
  • Explicitly call out the methodology and its benefit to a marketer (more than just a link to the product page), for example:
    • “The real power of Periscope lies in its ability to pinpoint the key brand attributes that drive investment preferences.”
    • “A conjoint allows an organization to pinpoint the exact attributes of their brand that drives a consumer down the purchase funnel.” (This is not a quote from one of our current stories… but could be.)
  • Emphasize the consumer behaviors that marketers need to be aware of for example:
    • “When it comes to purchasing beer, the most important attribute consumers look for in a brand is its sense of humor, followed by an easy-drinking beer and ‘cool’ factor.”
    • “Americans-pretending-to-be-investors view brands with high growth potential, premium products, and a loyal customer base as more likely to be considered to receive their investment.”
  • Reference the decision that brand marketer’s could make given the results of the analysis for example:
    • “Athleta, for example, would benefit most from updating their positioning via marketing to convey their products as premium and serving a niche audience.”

Previous approved product stories: Beer brands (brand metrics) Athleisure brands (Periscope)

Front Matter

After writing the stories, it’s time to entice people to read the content. Generally speaking, this is the spot to really bring the wit, humor, and wordplay to get people to open and read the stories.

Subject Line

It is probably the most difficult to incorporate wit in the subject line because it needs to be pretty short, but in general the subject line should be enticing enough where people will want to open the email. It should also make reference to each of the stories. Some examples:

  • Not-so-holy Ghosts, Infrastructure Holy Water, and Smartphone Samesies
  • Goodbye to DIY, Hello to Olympics Controversies, Tracking Santa’s Cookies
  • Weeding, High School Horrors, and Critics of Critical Race Theory
  • Falling for Fall, Afghanistan Blame Games, and Carnivores

Preheader/Preview Text

This is the text that will show up next to a subject line in an email inbox as a preview before the email is opened. So like the subject line, it also needs to be short and enticing, though there is a little bit more room for word play. Sometimes it comes down to essentially rewording the subject line. Some examples:

  • In this edition we’re covering super powers, super allergies, and super fans.
  • In this edition we dive into the future of grass and the past and present of school.
  • In this edition we’re bringing the pumpkin spice, war reflections, and dietary regrets.

Email Introduction

The space between “Welcome to Trendlines!” and “Enjoy reading.” is fun to put together. The main goal here is to re-introduce ourselves to our subscribers and briefly summarize each of the stories in the edition.

There are no strict rules about how to do this. One common approach is to include an “ice-breaker” that mentions the fact that we are back to deliver some data-driven insights and/or makes general reference to the time of year or something in the news, followed by a bulleted list featuring the story summaries. An example:

Welcome to Trendlines!

It’s November and we’ve finished gathering our data-crunched leaves into a big digital burlap sack in order to show off our neatly manicured Trendlines lawn. In this edition, we:

  • Show that Americans support providing homes for the homeless at the expense of billionaires.
  • Try turning the supply chain off and back on again.
  • Save you a big Friday night kerfuffle with your significant other by recommending the movies to skip because they’ll suck, and which to skip because the theater will be too crowded for the booster-ineligible.

Enjoy reading.

Another approach is to dive right into the summaries:

Welcome to Trendlines!

This go-round is filled with more highs and lows than a Christina Aguilera rendition of the national anthem.

The highs begin with a data-driven puff of Americans’ views on marijuana legalization. The people want to get high, but politicians and lobbyists are like totally bumming them out.

The highs continue as we asked Americans to reflect on their high school experiences. Join us at the cool kids’ table for the gossip on who was hot and who was not.

What would the highs be without getting the low-down on Critical Race Theory? Lo and behold, Americans may be more down on the name than on the concept itself.

Come for the insightful analysis, stay for the gratuitous puns. We’ve got pun-ty of both.

Again, there really are no strict rules for how this part should look. But the goal should be to entice people who have opened the email to want to read the stories.

Whiteboarding

Whiteboarding is the Gradient term for the semi-synchronous editing process that takes place directly before launching the newsletter (typically scheduled for Wednesday at 10AM ET). The person in charge of making the newsletter plots along with anyone with a whiteboarding issue assigned to them will partake.

There is no Zoom meeting necessary, but it is expected that everyone involved in whiteboarding will monitor comments in the newsletter document and the Slack channel for approximately two hours on the morning of email launch. Those two hours are set aside for a close review of each story, each plot, and the front-matter, with reviewers making suggested changes and comments in the document.

When plots are finalized, the high-res .png files for each plot in the story should be uploaded to the newsletter Slack channel, so that the person launching via Campaign Monitor can access the correct files and include them in the email launch.

Once everyone has done a close reading of the document and all of the open comments have been resolved, it is time to move on to launching the newsletter.