Automating Our Content Marketing: What We’ve Learned so Far

This is a guest post for the Distilled blog, kindly written by David Mytton, Founder and CEO of Server Density.

“We have serious persuasive power immediately after someone thanks us.”

That was Robert Cialdini, author of one the most authoritative books on influence. If you need a dead simple reason as to why content marketing works, look no further than that.

It’s all about reciprocity

Content marketing is a fancy name for something that has worked since the beginning of time. Need something from other people? Then pay it forward. Help people do what matters to them. Karma will take care of the rest.

If marketing is a path to your goals (profit, growth, eyeballs) then content is a path to your customers’ goals. We were lucky enough to discover the marketing value of content years before “content marketing” became a thing. It was during my time at, almost ten years ago, when I observed how powerful written content can be. Our entire product offering (business listings) was, in fact, based around content.

So it was only natural to carry that discipline forward to my next venture, Server Density.

As early adopters of MongoDB, we had plenty of things to share. Documentation was scarce. There hardly were any online resources. We learned everything by trial and error. Our blog audience were sysadmins and developers interested in learning about MongoDB. That was the exact same audience our product, Server Density, was aimed at: highly technical engineers, interested in automation and productivity.

As MongoDB adoption rose, so did our audience. And when we wrote an article on “Does everyone hate MongoDB?” it shot straight to the top of Hacker News and led to hundreds of sign-ups.

Bad content is bad marketing

Looking back at our marketing strategy, Server Density has benefited from a two-fold first mover advantage. We embraced MongoDB and content marketing early on. Both bets paid off.

Over the years, many companies have woken up to the allure of content marketing. A majority of those efforts, however, tends to focus on producing short, low-quality posts, as quickly as possible. This approach is not sustainable.

Low-quality content leaves a bad taste in the mouth. No one shares bad content. Nobody feels inspired to reach for their wallet after reading bad content. Bad content is a short term hack, driven by some blind hunt for eyeballs. It’s based on the tenuous belief that web traffic automagically leads to revenue.

What’s more, blogs are often outsourced to content sausage factories who inject them with keywords and SEO hacks, hoping to outsmart Google. That approach doesn’t work either. That’s because . . .

Google doesn’t like bad content

Search engines cannot attract advertisers if their search results are bad. It is in Google’s best interests to fine tune their algorithm and penalise bad content. That’s bad news for content hacks.

Over the years, we’ve never been affected by any change Google made to their ranking rules. That’s not to say we don’t optimise our content. We just don’t treat SEO as a magic trick. There is no hocus-pocus about it. The title (and keyword) needs to reflect how your article helps the reader, and that’s about it. On average, we spend no more than 10% of our time on SEO. The rest goes on creating regular and useful content.

Good content takes time

Making time for good content is not easy, or cheap. We’re a young company with finite resources. As our client-base grew, one of the first lessons we learned was that it pays to automate. Don’t do anything more than once. Figure out how to do things and then automate them. Safeguard your time.

Over the years, we have automated many of our engineering processes. Tools like Puppet help us reduce the amount of things we do by hand. So we wanted to extend that ethos to other parts of the company, including marketing. What follows is an example of us doing just that.

Automating our newsletters

A few years ago we started sending out a weekly “DevOps Newsletter”. It was a summary of the 8 most popular things we posted on Twitter that week, followed by a funny gif.

250 installments later this newsletter has a huge number of subscribers, a 30% open rate, and 10% click rate. Our audience loved it, so we kept doing it.

The only problem was that it took us an hour each week to prepare it. Someone had to go to Twitter, get all the stats out, go to DevOps Reactions, find a funny gif, before heading over to MailChimp to create a campaign.

So we decided to automate it.

Developing the necessary scripts did not take that long. All the apps associated with the newsletter have documented APIs. The following class, for example, contains the logic for sending out a generic email campaign. Here is an excerpt:

The second class is specific to the content we are sending. In this case, it includes the API endpoints for Tumblr (DevOps reactions is hosted on Tumblr). It also contains the logic for extracting the most popular Tweets from Twitter. (Our script aggregates retweets, mentions, and likes into one variable. In the future we plan to place more weight on retweets and mentions, versus simple likes.)

We estimate to have saved something like 52 hours of manual effort so far. That extra hour per week allows us to allocate more time on other projects. For example, we eked out the time to build Fluffify, a Chrome extension that eradicates corporate phrases and bromides from every page you visit on the Internet.

We generally try to be “on the pulse” with stuff and be able to respond to trends and things as they happen. We wouldn’t be able to do that if we wasted hours on manual tasks.


Since launching in 2009, we’ve grown organically, through word of mouth, and without heavy marketing budgets. By sharing content that helps customers solve real problems, we forged strong relationships with them. As for signups, growth and revenue, those were a natural outcome. Not a tunnel-vision obsession.

Producing high quality content that helps customers, is not immediate or cheap. As a young company, time is our most precious resource. In order to afford good content, we have to eliminate low-value activity. The best way we can think of doing that is by automation.

Automating our engineering and marketing activities means we can continue to build compelling products and craft great content for our customers.

David is the Founder and CEO of Server Density. He has been programming in Python and PHP for over 10 years, was one of the earliest production MongoDB users (founding the London MongoDB User Group), co-founded the Open Rights Group and can often be found cycling in London or drinking tea in Japan.

Get blog posts via email

About the author
David Mytton