The SEO playbook for seed-stage startups is different than for established companies. Young startups often lack domain authority, volume of content, and a strong social media following, all of which makes it hard to compete with large players.
But they have other traits that they can leverage to get off the ground, so it’s important for startups to focus on their unfair advantages and embrace Paul Graham’s “Do Things That Don’t Scale” philosophy.
In my last corporate job in 2013 I was SVP of Growth at IAC, where I was in charge of creating a Center of Excellence in SEO for properties including Ask and About. That experience taught me how to approach enterprise-level SEO with six figure vendor contracts, large staff, and access to proprietary resources (eg: our own internal search engine in the form of Ask).
The other side of my career includes co-founding three SEO-driven startups (including the one I am currently running).
Comparing the playbooks of my startups and those we used at IAC, I realized that the usual SEO best practices (e.g. website architecture, title tag optimization) are far more aligned with enterprise-level companies than with seed-stage startups.
That has a very simple reason – most of that content is generated by SEO consultants, whose clients are usually companies with decent traction and the ability to pay for such services.
In contrast, most seed-stage startups can’t afford expensive consultants, and even if they could, those consultants wouldn’t have much existing content or domain authority to work with, and it would take much longer for them to return their investment.
As a result, there is very little SEO content geared towards seed-stage startups. Their founders either recognize the unsuitability of those best practices to the current stage of their company, or they execute those playbooks only to get burned and then avoid SEO altogether.
The goal of this article is to shed light on this issue and to share some insights on what has worked for my own and other seed-stage startups that I am familiar with.
Do Things That Don’t Scale
As alluded to above, the playbooks are different based on the amount of existing content and domain authority (which is a measure of how much search engines trust your domain). A common measure of domain strength is Moz’s Domain Authority. Other tools, such as Ahrefs, have their own proprietary metrics.
“Domain strength” is another way to say “credibility,” which in SEO terms means “links.” Google trusts stronger domains and therefore ranks their content quicker. This is the advantage that large, established websites have with SEO: they’re already trusted by search engines.
Generally, you want to move towards having a lot of content and a high domain authority. In the past, it used to be possible to get there by publishing a huge amount of low-quality content and also simultaneously engaging in a very aggressive link building effort.
Those days are over since Google now penalizes domains that have a high content footprint relative to their domain authority, and also they are getting increasingly better in identifying unnatural backlink portfolios.
On the positive side, Google has also gotten better in identifying high-quality content, and these days it’s not uncommon to see amazing articles published on small blogs outranking high authority competitors with lower quality content.
In other words, content quality is a lever and can be used to make up for the lack of domain authority. It is, however, an expensive lever, and a high-quality article can easily cost well over $1,000.
As the content volume grows over time and the domain authority improves, ranking gets easier and the content quality no longer plays the same level of importance as it did in the earlier days. At the same time, in order to maintain high growth rates, established companies increasingly start looking for ways to scale their content production, which often results in lower quality standards.
This is ultimately why it’s still possible for startups to rank for high-quality keywords – Google’s insatiable thirst for content quality means that anyone can be ranked at the top of the Search Engine Results Page (SERP) if they just want it badly enough.
And there is a simple reason why it makes sense for seed-stage startups to invest more in their first few articles than an established company would decide to do: they have no choice – they simply have to start somewhere in order to get noticed and start getting their inbound links, which will eventually enable them to increase domain strength and lower their content production cost over time.
In fact, the production of expensive, amazing content is just another aspect of Paul Graham’s “Do Things That Don’t Scale” mantra, which turns out to not only work as a great playbook for early-stage customer support and hiring, but also for SEO.
Step by Step Playbook
I was co-founder and CMO at Airpair, which was in the YC W14 batch. I’ll use our playbook to illustrate how we followed the “Do Things That Don’t Scale” approach to grow our organic search traffic to a million monthly uniques in less than a year.
Airpair started out as a marketplace to match software developers who needed help with engineers willing to debug their code. To attract customers, we needed to be found by people whose searches expressed intent for troubleshooting their code.
It’s worth noting that we started ranking on the first page in Google for high volume keywords like “angularjs tutorial” (15,000 monthly searches) within the first two weeks of launching our content strategy. That should be the ultimate proof that content quality can be used deliberately to make up for the lack of domain authority, although there’s a bit of a recipe that has to be followed and which I’ll document below.
In the examples below, I’ll use Searchmetrics as my SEO toolkit of choice. Note that there are plenty of other alternatives available, including Ahrefs, SEMrush, and many more.
1. Identify the Most Successful Domains in Your Content Vertical
Searchmetrics allows you to pull up a list of domains that rank for some of the same keywords as your domain.
Obviously, this is not going to be useful if you don’t have a single keyword you’re currently ranking for, but I’ve found that very little data is necessary to identify the first relevant competitor.
As soon as you have that data point, you can then search for their competitors, and wash, rinse and repeat until you have identified every high-ranking domain in your vertical.
2. Identify the Top Keywords
The next step is to identify which keywords drive the majority of traffic to those competing domains. This is also possible with Searchmetrics, as shown below.
3. Group Top Keywords and Most Successful Content Angles
Export the top 50 high volume and high intent keywords from each domain, and then group them by topics. In Airpair’s case, most of our competitors’ organic search traffic comes from keywords related to programming languages, frameworks, databases, and common error messages.
The next step was to identify the most successful content angles for the above topics, and we identified that the best-performing articles often followed one of the formats below (where the [l/f] placeholder stands for “language/framework”):
- [l/f] tutorial
- [l/f] vs [l/f]
- [l/f] interview questions
- [l/f] common mistakes
- Introduction to [l/f]
- Fatal error: [error code / name]
All of the above topics would allow us to drive a lot of traffic to our site, but some have a higher purchase intent than others. At this point, I should point out that Airpair is a marketplace for on-demand programming help via Hangouts. With that in mind, we assumed that of the themes above, the “Fatal error” one had the highest purchase intent, followed by “[l/f] tutorial.”
4. Evaluate Content Distribution Options
It’s important to view each article as a distinct product that requires a budget, has a clear purpose and performance goals, and also needs a distribution strategy. The last item is particularly important for seed-stage startups since they don’t usually have a large social media following or email lists, and therefore need to be able to attract 3rd party communities to share their content. In other words, the choice of topics can make or break your distribution strategy, and you should choose your editorial calendar very carefully.
After much research, we narrowed our initial content topics to two themes: 1) tutorial content and 2) articles explaining error messages.
The tutorial content was going to resonate with a number of communities (Hacker News, Reddit, etc), and it had a much larger search volume (15,000 monthly searches on Google for “angularjs tutorial,” compared to only 20 monthly searches for the phrase “allowed memory size,” which is commonly included in out-of-memory errors). At the same time, the number of errors you can write about dwarfs the number of tutorials. Given that we’re more likely to be able to promote the tutorial content than the error content on 3rd party sites, PG’s “Do Things That Don’t Scale” framework made it easy for us to define our content strategy as follows:
1. We’ll start off with the tutorials, since it is better suited for expensive, amazing content that will allow us to impress and attract 3rd party communities.
2. Once we establish ourselves as a destination for great content and achieve some domain authority, we’ll start publishing the content on errors, since that content will then be able to rank using our internal resources alone (domain authority, social following and newsletter).
5. Write, publish, and promote
As outlined above, we started off with a 10,000 word post on AngularJS in early 2014, right around the time when Angular 1.0 was starting to gain steam and there was still very little 3rd party documentation available (generally, the official documentation often feels a bit dry, leaving the door open for individual developers to publish easier to follow guides). For this task, we hired the best AngularJS developer in our marketplace, Todd Motto.
It’s worth noting that what we paid Todd was considerable, but still much less than what he can make in one day. So, the only reason he agreed to do it was to leverage the combined promotional effort we were going to put into this project. This included, among others, reaching out to Brad Green, the Angular manager at Google, who received an early draft and then tweeted about the guide to his 30k followers.
We also reached out to every individual AngularJS developer in Airpair’s marketplace, and dozens of other high profile developers across the world. As a result, our guide was featured in almost all AngularJS-centric newsletters, and we received over 500 inbound links and over 9,000 social shares.
Most importantly, the post started ranking for “angularjs tutorial” on the first page in Google, and it continues to rank for that term even 4 years after being published. Since then, over a quarter million people have visited our domain just because of that one post.
6. Wash, Rinse, Repeat
By the time we started publishing the templated content, we had achieved enough domain authority and social following that we managed to promote that content to just our internal community and still rank with it on the first page in Google. Content like Fixing PHP Fatal Error: Allowed Memory Size Exhausted earned us over 100 inbound links and over 1,000 social shares without almost any promotional effort.
Evolving From Expensive To Scalable
Make no mistake – Airpair did not have the funds to continue paying thousands of dollars for each article (and we never did again after that first piece). As our publishing platform proved to be a great way for developers to build their own brand, we were able to lower the financial incentives and ultimately remove them altogether – the vast majority of the articles published on Airpair came to us unsolicited and were contributed for free.
To support that user-generated content effort, we published detailed guidelines that helped standardize everyone’s format and integrated our CMS with Github to keep the content quality as high as possible.
This illustrates the evolution of Airpair’s content strategy from expensive, amazing content (tutorials) to templated content (errors) and user-generated content (unsolicited articles). Once there was enough content on our site, the next logical step was to start applying the SEO best practices, which gave us an additional boost in rankings.
SEO can make or break a company, and contrary to popular belief, it is a suitable growth channel for companies at all stages. It does require, however, a different playbook for each stage, and I hope that this article helped illustrate how we navigated that evolutionary path with Airpair.
I hope that I’ll soon get around to detailing similar paths we did at Calorie Count and Kindly Care, although the high-level strategy stayed remarkably similar to what was described here – Do Things That Don’t Scale… until you are able to Do Things That Do Scale.