Sending Newsletters for Fun and Profit
Who are you and what are you doing here?
My name is Josh Gee, and I recently served as the Director of Audience Development at Boston Globe Media Partners. I drove strategy and execution for search optimization, email (newsletters and marketing), paid traffic, and business development/partnerships for boston.com, bostonglobe.com, and their constellation of awesome smaller sites.
I helped build this, and this was my baby. In short, my grandmother was very happy that I had a job in marketing of some sort at a big company.
Before working at the Globe, I worked in the fast-paced, not-Nana-approved world of political campaigns and consulting. Over the years, I’ve sent millions of emails to raise money, create action, or drive clicks. I’ve used email to help some pretty great candidates get elected. I also used it to support some pretty big digital campaigns while working here. I came to the Globe with an eye towards using some of what I had learned to help overhaul their email newsletter program.
My friend and former colleague Matt Karolian just blew the doors off the place with a great post about social, so I have been guilted into sharing what I’ve learned about email newsletters.
It works, slowly
It’s no secret to anyone that email is really hot right now in media. Everyone and their mother is launching an email newsletter. That’s because it works! As soon as we started accurately tracking Boston.com’s email newsletter traffic, we realized our relatively small (< 100,000 subscribers) morning headlines newsletter was our #2 external source of traffic after Facebook. That’s pretty great.
But it’s slow. We promoted email aggressively across all BGMP properties and only saw small yet consistent gains. That’s similar to what I’ve heard from other publishers. I don’t think it’s realistic to launch a newsletter and expect it explode unless you’re willing to invest money in list growth (I welcome a challenge to that assumption). The timeline for significant organic growth from email is quarters and years, not weeks or months.
Delivering email is absurdly complicated, so keep the rest as simple as possible
Email is old, complicated, and janky (For more on this, read this amazing story). Delivering millions of emails to the right people involves highly technical work as well as deep relationships with email providers. Even Buzzfeed, who builds EVERYTHING themselves, uses an outside company to handle delivery. The choice of technology partner is an important one and most media companies seem to be shopping around right now.
The road to hell is paved with enterprise software solutions.
Unless you are are investing in a large email production department, then making it easy for non-technical folks to produce and send emails should be your top priority. It’s easy to get caught up when large providers talk about fancy automations, IP deliverability advantages, and better outcomes. What you don’t want is a technical solution so complicated you won’t be able to even get started without investing significant resources. Scaling down is just as important as scaling up.
Based on what I’ve seen, the media companies thinking hard about email (Buzzfeed, LA Times, NY Times, Quartz) are investing in email the way they are investing in social: a dedicated team coordinating very closely with editorial that understands the format and the form but with a mandate to be creative about new products. Some publishers, like Business Insider, have invested in really fancy automations and seen some great numbers. But that takes a nontrivial amount of technical work and it seems like the trend is currently in favor of greater editorial control and building relationships between the sender and recipient.
At the Globe, we had a mix of both, but overall our editorially curated emails sent by specific people performed best.
From a product side, make it easy for yourself by keeping it simple. Email clients can’t handle complicated code or designs, so your templates should be simple. Also…
Email is mobile, and it’s a feed
Across BGMP, more than 60% of our email newsletters were opened on mobile devices. Email strategy *is* mobile strategy.
Also, email is a feed. I mean look:
Newsletters are fighting for attention the same way your tweets and Facebook posts are. The tactics your social team is experimenting with could work in email. This presentation by Amy O’Leary has a great examples of what those should be.
A/B testing is something you can do with email that you can’t with social. For example, testing the subject line [Boston.com Today 4/15/2015] against [Headline of the top story] resulted in a statistically significant increase in open rate when we used the story-specific headline. Makes sense, since I’d be more likely to click on “Tom Brady gets his day in Goodell’s court” vs. “Boston.com Today 6/27/2015”. We also saw a lift changing the sender from [Boston.com] to [Writer Name, Boston.com].
You only have the sender and the subject line to capture attention — use them. Ideally, publishers should be testing different subject lines for each newsletter with smaller list segments the same way marketing departments do.
Focus on metrics that matter, and there’s no industry standard yet
So how do we judge if our newsletter is working? How do we feed the demand for data. MOAR DATA SO WE CAN MAKE MOAR DATA-DRIVEN DECISIONS. But srsly, it is logical to ask: “Well how are we performing based on industry standards?”
…the newsletter has achieved a 40 percent open rate — a figure Time Inc. boasted as twice the industry average when the company named Schweitzer its editorial director for audience strategy in October. Click-through rates after open are about twice the industry average of 16 percent, Schweitzer told me.
That’s from an article I linked to before (here it is again). I call shenanigans and think this is made up. There is no industry standard for newsletters that I was able to find. Maybe we should all agree that it’s 20% so we can look really good, but across the 21 Globe newsletters, some had open rates around 80% while the lowest hovered around 16%.
When I did more direct-response emailing for campaigns, I was thrilled to get over 10%. Looking at editorial newsletter numbers and I was adrift.
Is a 35% good? It’s fucking terrific!
Is it good compared to what it could be? I have no idea!
With a lack of clear metrics, all we can do is compare each performance against past results on a list by list basis and keep trying to move the numbers in the right direction.
Also, we probably shouldn’t care about open rate as much as we do.
Take a look at these two emails. Quartz probably cares about open rate (look, there’s an ad right there in the email!) while Buzzfeed probably cares about click-through rate.
Rather than focusing on a bunch of data that might not mean anything, figure out what the key goal is for your email. Then use metrics based on that goal as your north star.
My goal with Boston Globe’s Today’s Headlines morning newsletter was purely driving traffic back to the site. So I didn’t care what the open rate was, only the click-through rate. I’ve run experiments that resulted in a higher open rate but hurt clicks. If I hadn’t had that goal in mind, I wouldn’t have realized I was hurting myself.
Also, give metrics time to play out. Mailchimp doesn’t even report trends for three months. It’s frustrating and slow, but there’s so much variability that you need to take a longer view before judging success or failure overall.
tl;dr: Make it easy for readers to sign up, make it easy for writers to send, and be patient. If you can do that you’re 80% of the way there.
