There are many ways to go about implementing content marketing for your startup. The best is when you have an experienced content writer (or two), a talented designer, lots of money, and plenty of time. In this blog post, we’ll talk about pretty much the exact opposite: how a startup that consists of two Israeli developers — who write excellent code, but aren’t necessarily content marketing experts — got leads and sales opportunities through content marketing.
The Setup
Clay Sciences provides a solution to a problem its founders frequently encountered while working as data scientists. To train the machine learning models they were building, they had to use clean, labeled data. They often ended up manually reviewing and labeling data themselves — a process that was tedious, lengthy, boring, error-prone, and probably not the best use of their time. Now, after two years of development, the Clay Sciences platform offers a crowd-sourced way to label high volumes of data quickly, accurately, and cost-effectively.
What they did
The two founders secured their first customers through introductions and outreach to their personal network. At some point, they wanted to broaden their reach.
Their target buyer (and user) is technical: data scientists, software engineers, CTOs – anyone working on building machine learning models. They decided to take a stab at writing technical articles describing industry challenges, their product, how it works, and what its benefits are. They wrote the articles themselves, and had us do basic proofreading and copyediting.
For distribution, they posted the articles on LinkedIn, and asked friends & family to do some initial liking & sharing. For the second article, they also posted it on Hacker News.
You can read the articles here:
“Machine Learning is 99% Manual Labor” — and what I’ve been up to
How to get crowd annotations with expert’s quality
Now, let’s be honest. The content is technical and quite dry (despite a few jokes here and there). There are charts, industry jargon, and some awkward phrases.
Whoa! A chart! From Clay Sciences’ article.
The results
Here are anonymized screenshots of a couple of the responses to the posts:
A data scientist working for the Israeli self-driving development team for one of the world’s largest car manufacturers reached out after a colleague forwarded him the post. This eventually led to a pilot with the car manufacturer.
Similarly, this led to a conversation with another global car manufacturer working on its self-driving car technology in Israel.
In terms of post views, the first post was viewed 580 times, the second 896 times (as of June 2018). Not huge numbers, but when you’re just starting out, even hundreds of the right views, coupled with two qualified leads, can be significant to your business.
July 2018 update: Clay Sciences’ third post, Why annotating data directly on videos is way superior to annotating images, did even better. It received over 1000 views on LinkedIn, 128 Likes, and sparked a mini-debate in the comments. A month after being published, it continues to drive qualified leads.
Why did it work?
- A technical buyer. First off, this won’t work for every type of buyer. If you’re selling to CMOs of large US fashion brands, your content game needs to be at a whole other level: polished, eloquent, differentiated. Technical buyers judge content primarily on technical sophistication and accuracy, less so on grammar and writing chops.
- Work on your polish & style, but you don’t have to be perfct. That said, we’re not advocating for forgetting the spell-checker or leaving blatant grammar mistakes. Content that’s too rough will ultimately reflect poorly on you, your company, and your product. The key is to strike the right balance. You’re a developer yourself, so ask yourself – is this something I would enjoy reading? Would I respect the author? (pro tip: these are good questions to ask about any external marketing campaign or material you’re putting out there)
- Distribution is important, existing platforms are a good start. As with any content marketing strategy, content distribution is as important as content creation (if not more). In this case, LinkedIn proved to be a suitable platform for reaching a relatively large and relevant audience quickly, and essentially for free. Hacker News helped distribute the content to additional relevant audience.
- The right message to the right audience at the right time. Last but not least, this beloved marketing cliché rings true again. Clay Sciences is addressing a huge need in the market with a solid product. Their content was authentic and high-quality, and their audience responded enthusiastically.
The bottom line
Are you a developer who loves Scala but hates spelling? Do not despair. If your target user or buyer is technical, chances are they’ll be interested in what you have to say (or write) about your product, even if it’s not Pulitzer Prize-worthy yet. Give it a shot — write a blog post, share it on platforms like LinkedIn, Reddit, and HackerNews, and you might be surprised by the leads who’ll contact you.
