How Grammarly Built a Social Media Empire of 7 Million Grammar Geeks
When you think of correct grammar, fun probably isn’t the first thing that comes to mind. Maybe you uncover repressed memories of red pen marks or condescending quips from middle school English teachers. Or you remember that frustrating acquaintance from high school who still doesn’t know when to use the correct your/you’re in status updates.
But if you know where to look on the Internet, language geeks have started to turn proper grammar into a comedic art form.
Just ask Grammarly, the bootstrapped software startup hell-bent on fixing your grammar. The company’s popular (and free) software highlights instances of incorrect writing and suggests the proper changes to make. It’s essentially Microsoft Word on steroids, claiming to be the “world’s most accurate grammar checker.” And since 2008, Grammarly has been complementing that software by building a veritable social media empire that provides tips, jokes, and comics to millions of followers who are interested in language.
— Grammarly (@Grammarly) April 12, 2015
Relative to its institutional investors (zero) and team size (100 employees, with three dedicated to social), Grammarly has built an absurdly large audience. The company’s total number of followers and fans is now nearing 7 million across its social channels, which begs the question: How has it managed to make language fun and accessible?
I talked to Grammarly’s social media manager Kimberly Joki over email to find out. Joki touched on how the company runs social and content marketing, how it’s been able to create such raving fans, and why it reposts so much old content.
Your social audience is huge, especially compared to the team size. What fueled that growth? What’s your high-level approach?
It’s been very iterative. Before I joined Grammarly, we were focused on creating content for people we intuitively thought would be most interested in grammar and writing. It was a small, specialized audience, and it just didn’t deliver. It was clear that this “intuitive” approach was failing because we just couldn’t seem to leverage the group. Only after branching out to different content types and themes, while simultaneously changing our focus to data-backed results rather than projecting what we thought was best, did we find a format that resonated with a broader base of people.
For the last four years, our social media success was measured in terms of organic growth in engagement and reach. Recently in the last several months, we have been shifting to traditional ROI metrics (CTR, new users, etc.). Today, our social team looks at everything from social metrics to the number of premium subscriptions. This is possible only because we are now able to capitalize on our engaged social audience to drive traffic.
We set benchmarks based on our best performance historically and on pages that we want to emulate. Our main metric is organic engagement ratio (engagement total divided by organic reach) though we sometimes look at total reach when relevant. We adjust our historical benchmark quarterly. We also look at other pages [aside from Facebook]. While we can’t get insight into their reach levels, we can see engagement and PTAT [people talking about this] in relation to follower count. Again the higher the ratio, the better.
How much paid acquisition or exposure to followers has helped you build your audience? What specific channels worked there?
Somewhat unconventionally, we’ve kept paid acquisition and social media separate. Our earlier startup days demanded that we did as much as possible without spending a dime. We looked at the most successful social pages across every industry and saw quite clearly that money was not a requisite for social success. So that was what we set out to do.
After Facebook’s algorithm update in late 2013, we explored some paid options, but in the end nothing has worked as well as an obsessive devotion to testing and optimizing for organic exposure.
How much do you talk about your product versus stay top-of-funnel with more playful or tangentially relevant content? Why?
Until we have a product that everyone can use, we have intentionally been focused on increasing brand awareness. This will change soon, however! Within the next few months, we will be releasing the free Grammarly extension for Safari and Firefox.
— Grammarly (@Grammarly) April 13, 2015
For new ideas, we search for relevant inspiration everywhere, but we get a lot of great ideas from our fans. We subscribe to a daily digest of relevant content from FanPage Karma. From there, the content is pre-screened by a VIP group of fans using PollDaddy. If a post performs above the benchmark established, then it is filed as approved. Dropbox makes the shifting and changing files manageable across our distributed team.
(Ed. note: Grammarly’s 13-person marketing team is split between its San Francisco and Kiev locations.)
I then pull a mix of content from this “approved” bucket as well as from previous high-performing content, align it with the needs in the editorial calendar for social and the blog (managed in Google Sheets), and send the whole plan over to design.
Our designer, Elena Godina, is also a skilled illustrator. Using Adobe Photoshop and Illustrator, she is able to do everything from reinventing old content to simply adding necessary logos or attribution. Once design is approved by our content designer or a senior designer or art director, the posts are scheduled. Ann Edwards, our community manager and Twitter master, uses TweetDeck for scheduling on the network.
You use a lot of repeat content. Why? How do you decide which pieces to use again?
We repeat content to find ideal designs and to expose new followers to our best content. Few will scroll through months of content on our pages. We decide on content based on pre-testing and/or comparing a post’s previous performance (engagement and organic reach) to the established benchmark for the quarter. If it meets the benchmark or passes it, we use it again. If not we find something else.
We split test across geography and time on our Facebook page (no ad dollars spent) and split test across time for Twitter. We also survey our audience using a sort of focus group. We have a dedicated group of followers that are asked about possible content. Based on their feedback scores, we decide whether or not to post something.
Have you received any negative feedback by sharing the same thing multiple times?
Few people are so involved with our Facebook page that they know every piece of content we have published, and on Twitter things move so quickly that we rarely hear any feedback about reposting content. We haven’t seen a significant increase in negative metrics that would be a sign of some more systemic discontent.
What’s your most successful channel? What’s your guess as to why?
Facebook is our most mature channel, and with the recent algorithm tweaks on the platform, we’ve seen some great conversion behavior from the community there. Due to sheer scale, it is our most successful both in terms of social engagement and direct ROI. (Ed. note: Grammarly reports that their average organic reach is 10 percent of their total followers.)
But we are seeing some promising trends in both growth and engagement on Twitter, though the strategy there is different.
So THAT’S what they’re called! pic.twitter.com/Kjg7wUxcOn
— Grammarly (@Grammarly) April 13, 2015
What have you learned about the different social channels?
Audiences across various platforms have more in common than a lot of marketers seem to want to admit. There are differences, but, at least for our keywords, they aren’t significant enough to warrant huge variations in content strategy. Instead, it’s valuable to spend time personalizing the implementation of your content for each platform.
For example, a question might perform best on Facebook as an image, while on Twitter the same question should be posed as a text tweet. The implementation is adjusted for audience preferences—usually confirmed by testing and detailed tracking—but the content is essentially the same. By taking advantage of these similarities, you can save yourself and your team a lot of work.
Where did you start? Did you pick one channel and decide to be great at it, or did you always syndicate across many channels from the beginning?
When I started as social media manager in 2011, the company was tiny and the marketing team was basically me and one of the founders who handled ads. We just didn’t have the bandwidth for a broad social presence. One of the first steps we took was to compare the channels that already existed and decide on the one that would be the best investment or time and resources. Due to the size of Facebook and the better overall engagement on the platform, it was easy to decide to focus there. After finding something that worked well on Facebook, we then applied those overall lessons as much as possible to other social networks.
What would you advise to others in your situation who are just launching their strategies or refining them to be more effective?
Many professionals and friends working at various stages of the startup journey have asked me this question. If you are just getting started, I highly recommend focus. Look at where your audience or prospective audience hangs out, and be there first. Hone your style there.