The discipline of “audience development” — or whatever you want to call it — is finally getting lots of attention among news organizations that realize reaching new audiences, and engaging with them, doesn’t always happen organically. It takes plenty of work.
A. Expose our brand to roughly 4 million civically engaged Texans (or about 1 in 5 adults in the state) to build future loyalists.
B. Deepen engagement among 400,000 or so politics/policy insiders (and monetize them).
C. Create an agile, data-driven culture internally to make the first two goals happen.
We’re off to a good start.
Monthly users are up more than 25 percent so far in 2015 versus the same period last year. We’ve had explosive growth in social referrals (121 percent), and most referral sources (search, direct, third-party) are up. And we’re on track to hit our aggressive goal of 45 percent growth in “consumer revenue” (sub-$1,000 donations, memberships and subscriptions).
But we have a long way to go.
Building a test-and-learn mindset and developing the capability to constantly evolve our products to grow audience — without ever sacrificing our mission, of course — is key. Thanks to a generous institutional fellowship grant from the Donald W. Reynolds Journalism Institute, we’re now a lot smarter about how to do this.
I wrote about the pilot here. Essentially it’s this: Test whether the concept of growth hacking, a tech startup staple, can work when embedded in a media organization. (A growth hacker — part engineer, part marketer — focuses on low-cost, innovative alternatives to traditional marketing.)
Our first challenge was to find someone actually up to the task, which was no small feat. A lot of amazing talent in this space is working, not surprisingly, in Silicon Valley. Elsewhere it can be a challenge.
We were fortunate enough to find Lindsey Lee, a multi-talented writer/designer/marketer/developer. She had digital marketing chops and had recently graduated from MakerSquare, a coding boot camp in Austin, Texas.
The second challenge was how to construct the pilot. Given the short duration of the test, we didn’t have the luxury to have her work hands-on in our code base (it would take longer than three months just to learn our various systems and how they fit together). So her work needed to stay off-site. In order to remain connected to us, she worked virtually with a small team: Chief Innovation Officer Rodney Gibbs, Consumer Marketing Director Allison Netzer and Director of Technology Amanda Krauss.
As a group, they first brainstormed test ideas (concepts had to deliver growth in one of three areas: raising brand awareness, increasing brand engagement, or growing consumer revenue) and prioritized them based on a graph of hypothesized impact and viability. Ideas that had high-potential impact and were relatively quick to execute (aka “more bang for our buck”) were at the top of the list.
And that’s where the experiment pivoted from its original intent: Instead of just a siloed hacker, we ended up with essentially a growth hacking team. The act of working together yielded a host of other tests beyond what would ultimately fall to Lindsey’s plate. In fact, ideas in the high-impact bucket were often impossible to execute without having a hand in our code.
Here are three of the tests we ran:
The Tribune, from the beginning, has freely syndicated its content to print, radio and television news organizations throughout the state (and, as of January, nationally in partnership with The Washington Post).
But we also want to make it easier for media partners and other organizations (like trade groups, associations and state agencies) to publish our work. And not just for altruistic reasons, although that’s most important; we also know a significant number of readers say they first discovered us in or on their local newspaper, radio or TV station. So Lindsey developed a Tribune headline widget for third-party sites, leveraging an open-source tool called Feed Wind. The beauty, we hope, is in its simplicity: A single line of code for sites to include, customizable by size and by topic (top stories, most read, politics, health care, water, etc.)
The result: We’re now starting to pitch the concept across the state, thanks to Natalie Choate, our membership guru turned director of media relations and partnerships.
Another suite of test ideas centered around taking friction out of the donation process. The Tribune has had success in crowdfunding and is now experimenting with concepts like text-to-give campaigns and making it easier to donate to the Tribune as part of our event RSVP process (early signs on the latter look promising).
Another idea was to integrate donations and member sign-up into Facebook. Lindsey created a Facebook tab, which essentially loads our join page in an iframe.
The result: We’ve had a few hundred referrals from Facebook to the join page. We don’t yet have conversion analytics in place, so we can’t say what the financial result of the program has been. The bigger challenge is to make a more concerted effort to drive traffic to the tab.
Email, one of the biggest guns in our marketing arsenal, is among the best ways to encourage lasting engagement and often our primary channel for member acquisition.
Led by Netzer’s creativity and hustle, the team ran several tests, including:
- Topic-based alerts: The team developed a new suite of RSS-driven email newsletters. Users can sign up for daily or weekly headlines on subjects we cover, like education, energy and environment, race and immigration, and transportation. We also built a hub for users to add and manage their email newsletters.
- Better on-site sign-ups: Netzer and team experimented with a couple tools to make email sign-up more visible on the site. We settled on AddThis, a tool that easily embeds on the site as a header or footer:
- It’s customizable in a basic sense (by color and text, for example) and allows for easy real-time A/B testing and data on conversion rates, so you can adjust on the fly based on what’s actually working. (There are a number of other applications we’ve tested: Sending people to our membership page, encouraging users to learn more about us, announcing a crowdfunding campaign, etc.)
- Optimizing send times, sender name and subject lines: For our daily summary email (The Brief), 6 a.m. was the clear winner. Another big takeaway? Including the sender’s name (e.g. “John Reynolds, The Texas Tribune”) had the most significant effect on open rates.
- Email subscriptions grew 300 percent (to more than 160,000, including subscribers who’ve signed up to multiple emails) and we saw a 16 percent improvement in open rates.
At the Tribune, we’re exploring ways to fund additional resources to do these sorts of tests on an ongoing basis. The goal: Work as a team — led by and executed by a growth hacker — on specific, time-boxed challenges (bump up membership conversion rates for those who make it to our “support us” page, for example.) Given additional resources, I can imagine having a growth hacker lead, say, a four-week sprint focused solely on tests that encourage first-time site visitors to turn one additional page — an engagement exercise that would have potentially huge benefits. And, of course, we’ll share results of everything we test as widely as possible.
This post appeared in the Reynolds Journalism Institute blog and has been republished with permission.