My startup, GiftRocket gift cards, first launched our blog in June 2011. Our objectives were to increase awareness of our product and to build link-worthy content. All pretty standard stuff for an early stage startup who wanted to do better in the rankings.

The problem was that while everyone told us to write “link-worthy content”, we really had no idea what that was. It took us a few months, but we did figure it out and started to get pretty damn good at it. This is the story of what we learned. Below is a map of our blog posts over time and how many social media shares we’ve been able to get per post.

We went from 0 from the first few articles to hundreds for all subsequent ones. This isn’t amazing-- Rand’s recent post got about 500, so we're still behind him-- but it isn’t bad for a young startup. Here’s the chart again, but with some important inflection points labeled. I’ll talk you through our thinking at each stage and how we kept advancing.

1. “What the hell are we going to write about?”

Again, our goal in launching the blog was to build links and get attention to our site. GiftRocket is an online gift card service. So last June right before Father’s day, we decided to open our blog and post about the Top 5 Father’s Day Gift Ideas. “Brilliant! Tons of people are going to be looking for gift ideas, and we’re going to dominate.”

This was an utter failure. We even blasted 1000 or so people on our mailing list and got no retweets or attention.

The biggest issue was that we weren’t writing interesting content. The stuff that gets handed around the internet is thoughtful, emotional, hilarious, informative, instructive, or just plain cool. Look at content that the Oatmeal puts out, or the infographics that Mashable publishes. Our post on gift ideas that spent most of its time plugging our own product was... boring.

2. “Let’s write outlandish taglines about timely content”

So we decided to get more ridiculous. When the US government announced that it was going to be shutting down NASA, we wrote an article titled Restaurant Gift Card Market Bigger Than NASA Space Exploration Budget. We tweeted it out to a few people and tagged it with #NASA, and it got a couple tweets and likes. We were jubilant.

We also tried to go after getting retweets from a few high-value Twitter folks. Nathan Fillion had just posted something about Comicon so we wrote up an article starring him. We did something similar with Matthew Berry of ESPN.

This stuff wasn’t a total failure, but it wasn’t anything great. We’d get a few tweets, a few chuckles, make it into a roundup or so, and it’d be done. The reason was that we hadn’t answered the following question: who the f*ck cares. We were making mediocre content for people that weren’t that interested in it, banking that a few really influential people would take the time out of their day to read it and retweet.

3. “Let’s write content for content-hungry communities”

There’s an intensely devoted community out there focused on entrepreneurship, startups, and computers. I’d say the center of it is Hacker News, a Reddit-like community where stories about startups, programming, and technology are upvoted and downvoted.

I’ve also been a startup entrepreneur for a while, and one day scribbled out the first post we submitted. It was a bit of a rant, actually, about people being transactional with each other in Silicon Valley. We titled it Don’t Burn Bridges, posted on our blog, and submitted to Hacker News.

The content was upvoted and at one point made it to #1. Some 15k visitors poured onto our blog and read the post; it was tweeted almost 150 times. I repeated this process with a post focused on the design community, another set of folks focused on devouring content. It was an in-depth analysis of our site’s redesign; it took me something like 15-16 hours to scope out and complete. It earned 300 tweets, tons of links, and still gets traffic today.

The thing that changed is that instead of writing general content and marketing it to no one, we started writing good content for communities that were interested in it. I knew it was good because it was the type of stuff I liked reading every day. And because there was a community intensely devoted to it, distribution stopped being a problem.

The sheer amount of time we spent on writing that content had an incredible correlation to how popular it ended up being (and how many links we got as a result). People started commenting at the bottom of our posts thanking us for what we had put together.

4. “How do we keep this going?”

We wrote further content about design and startups and they’ve all done consistently well.

The newest problem we’re focusing on is relevancy. People like hearing about gift ideas, but it isn’t central to their livelihood. No one is a gift “hobbyist” that spends their Saturday afternoons thinking about gift cards.

Hopefully we’re on the cusp of solving that as well. Because while people aren’t always interested in gift cards, they are interested in a lot of things related to gift cards-- for example, data, pop culture, visualizations, celebrity, whatever. If Chuck Klosterman can make breakfast cereal into a riveting topic, we can do the same with gift cards. And when we figure it out, we’ll blog about it.

If you’ve got thoughts about things we should do, we’d love to hear about them (email me at [email protected]).

Author bio: Kapil Kale is a co-founder of GiftRocket, a YCombinator-backed startup. GiftRocket sells online gift cards to any business in the US. Gift cards have been largely unchanged for the last 20 years and we plan on changing that by making them more flexible and easy to use. Visit our blog for gift ideas and more.