What I learned about content marketing at CXL

Last Updated on August 26, 2020 by Alex Birkett

I always wish that more people would document their learnings as they go through a process. We often get post-hoc analyses that are cleaner in retrospect and spoken about from a 10,000 foot view.

A big part of my job at CXL was to do content marketing – create great shit, promote it, grow traffic, leads, etc.

Luckily, I did document a lot of what I learned while going through it.

This post covers a lot of what I’ve learned about content marketing and doing it well (in the context of CXL’s business, at least. May or may not apply to yours).

On creating good content

People tend to parrot easy-to-digest advice regardless of the veracity. This one comes to mind: “spend 80% of your time promoting content, 20% creating it.”


For every 4 hours you spend writing an article, you should spend 16 hours promoting it? How long would Tim Urban have to spend promoting his content to adhere to this advice?

These tweetable pieces of advice catch on because they are simple, easy to remember, and directionally true.

“Directionally True” is an infuriatingly amoral phrase that I believe Scott Adams coined when trying to apologize for Donald Trump’s mistruths (“You don’t want to assume his facts are literally true, although they are usually emotionally or directionally true,” he says).

Directional Truth does help here, though, in that most people should spend more time promoting their content. Most people hit publish with no pre-promotion plan and no post-promotion efforts and wonder why they don’t get traffic. Most people should do more.

But 80/20 is a proportion that only someone with an agenda would advise.

Content creation is important.

Don’t skimp on it.

First, you shouldn’t add to the noise of bullshit content online. It’s ruining the internet.

Even if you don’t care about your poisonous externalities, it’s more effective to put together 10X Pillar Content. This type of stuff tends to weather Google algorithms and other uncontrollable things. Focus on quality.

Hacks, tactics, listicles

10X content doesn’t simply mean long content. There may be a certain je ne sais quoi about 10X content, but I know for certain the factor here is not content length. There are lots of 5000 word fluff pieces out there.

The truth is, it’s hard to say what an excellent piece of content looks like. Here are a few publications I think are doing really well:

(Mandatory note: businesses need to care about ROI. You’re not a publishing company (probably), so first and foremost content is a business asset. The argument I’m trying to make is that, most often, you’re better off spending time crafting valuable and high quality content. The business case and choice is yours, though, as I’m sure some industries are better off pumping out mediocre stuff at scale. There aren’t many black and white answers, sadly).

On content promotion

Content promotion is about putting great content in the right places. Strategically, there’s not a lot of wiggle room here. If you put great content in the wrong places, it will be irrelevant. If you put bad content in the right places, it will be worse than if you had done no promotion at all: you simply annoy and spam potential visitors.

Tactically, what I’ve found is most effective is the least “scalable”: you have to build genuine relationships.

People are now obsessed with phrases like “influencer marketing,” but I think that’s the wrong way to look at it (morally and strategically). In a Reddit AMA, Ryan Holiday said about the same:

I’ve gotten enough cold emails to know that most people look for the easy way out. They read a Backlinko blog post and copy the process down (to the very email template recommended). It’s easy to see right through this (especially when you’ve seen the same template a hundred times).

Image Source

You’ll find that authentic communication prevails. When you join a community, you want to be the one adding value, not just spamming your own content. When you email someone, you can’t just ask for favors all the time.

Voting rings and content sharing/upvoting quid pro quos can work to bring in traffic, especially if your content is actually worth being on the front page of a community.

But otherwise, it’s a stale tactic and you’re often marketing to the echo chamber. If you frequent Inbound.org or Growth Hackers, you’ve seen the mediocre posts on page one with relatively few or no comments (and if there are comments, they’re bland statements like “nice work!”).

How frustrating it must be to be a moderator in these situations, because the power users of the site are the ones causing the most abuse.

However, being a meaningful contributor to those communities was definitely valuable (still is). You forge connections, build credibility, and lo and behold, people share content for you. You don’t need to spam a bunch of strangers asking for upvotes (which simply feels bad for the soul).

One point that may not be helpful but is a personal vendetta of mine: when you are asking for a link or a social share, just come right out and say it. When you try to obfuscate your request through some weird request to “get you thoughts,” or “get your feedback on this article,” it comes off as super disingenuous.

You don’t want my feedback (trust me); you want my backlink. Stop sugarcoating things.

And whatever you do, if you’re sending outreach emails, at least don’t send a comically irrelevant pitch:

Apart from my rants, content promotion is mostly about finding a distribution channel where you have an unfair advantage, where the spoils outweigh the costs. If you can find a way to make scalable cold outreach or community spamming work, who am I to tell you not to? It’s just in my experience I’ve seen that may not be the most effective use of time.

On content optimization

It wasn’t until very late into my role at CXL that we looked into content optimization seriously.

Well, we started early on fixing broken links and testing different opt-in offers, but that’s table stakes, low hanging fruit. When you look at content optimization as a lever which you should be focusing on as much as creation and promotion, things change tactically and strategically.

Instead of simply looking forward at your editorial calendar, you ask questions like:

  • Which blog posts are getting high traffic but are driving very few conversions?
  • Which blog posts drive few, but very high value, conversions?
  • Which pages rank at 7-8 in SERPs that would be worth trying to push up to 1 or 2?

If, for instance, you find that your highest performing pieces of content in terms of traffic are bringing in little to no quality conversions, you can look to 1) improve the offer on the page 2) improve the content itself 3) focus on producing different content that actually aligns with conversion intent.

Once you’ve produced a pile of content, it’s important to learn from the data and shift the strategy. It’s too easy to look forward and keep pumping out the content you know how to without seeing if it’s actually effective. Plus, the lowest hanging fruit, in terms of traffic and conversions, is usually in this optimization lever.

It’s quite a similar approach, here, as you’d take with conversion optimization in general: look at the data, run experiments, and move the needle.

The bottom line is: this will increasingly be a powerful and competitive lever for content marketing. Don’t skip it.

On content marketing program management

Content marketing, like anything else, should be viewed as a system.

There are multiple levers, or inflection points, that can affects your outputs. Most people, as I mentioned earlier, have a keen focus on both content creation and content promotion (more so with creation), but they forget about content optimization.

Image Source

All three of these levers matter when it comes to quantitative outputs. A parallel is how most people view growth marketing – there, we tend to agree that we can experiment across the different nodes (Acquisition, Activation, Retention, Revenue, Referral).

You may choose to focus on one of these levers for a quarter or another mark of time, but all of them matter in terms of your end results. Usually, one lever is substantially weaker at a given point in time, so focusing there will bring more results.

An example with content is that if you’ve optimized your content creation lever and you’re churning out 4 stellar pieces of content per week, you’ve got that down. Let’s also say you’ve got an SEO/growth person that continues to optimize content, but you haven’t increased social or referral traffic in 6 months. You may, then, want to look at content promotion and how you can improve that area.

In addition, bringing efficiency to processes within each lever is important as well.

As an example, how you manage guest or freelance writers is super important to your content creation productivity and results.

That was one of my earliest challenges at CXL. I was a solo marketer, and I had other things to do than writing content. Getting guest writers would help free up my time for other strategic efforts, I thought.

This is true in theory, but rarely the case in practice. As someone managing other writers, you’ll often resonate with the following cliche:

“If you want something done right, you better do it yourself.”

Lots of things go wrong, mostly in the chain of communication and expectations, so you end up spending more time and energy to get a guest post published than if you had simply written it yourself. This isn’t always true, as you’ll surely work with some straight professionals (and this frustrating process allows you to find the gems that you work with longer term).

But be careful as wasted time begins to creep up on your productivity as well as your stress levels.

There’s been a lot written about the relationship between marketing manager and guest/freelance writer, but here’s what helped me:

  • Find writers whose work you already know and admire.
    • I found those whose work I was already reading on Growth Hackers, Hacker News, etc. I simply reached out to work with them.
  • Create editorial guidelines and add a “brown M&M’s” clause.
    • This should be a detail that isn’t necessarily hidden but will prove to you that the potential guest writer has read the guidelines.
    • Mine was a specific subject line that the guest writer needed to lead with.
      • If the subject line was different, I threw out the email.
  • Create a shared Trello board so guest writers (once accepted) can see where you’re at in the process and what other writers are working on.
  • Try to work with writers who actually do the thing they’re writing about. I know there’s a big debate on subject matter experts vs. writers, but at least in the domain of CRO, it was always way better working with subject matter experts than those that could write well. I can make the writing pretty, but I can’t put knowledge in someone’s head.
  • Give up after three drafts back and forth.
  • Give clear, straightforward, and constructive feedback. Don’t focus on minutia (commas), focus on structure and themes. Critique like you’re your own audience (what questions would they have?). I like blunt feedback, but some writers aren’t going to be prepared for that.
  • If a writer works out well, try to get them to write regularly, or if possible, hire them. Good writers are really hard to find; don’t think your ad hoc guest writer pipeline will expand forever.

Similarly, I created a baseline process/checklist for content promotion early on.

A lot of research went into who the movers and shakers in the CRO space were, which communities were vibrant, and what channels we could have an unfair advantage in.

We used this for a long time in the beginning, but later on as we learned more and got feedback to the process, we slowly shifted to higher performing distribution channels and dropped out the low performing ones.

Bottom line: If it’s tough to create an individual piece of content, it’s much harder to continue to produce great content, and promote and optimize it, with consistency.

Build the system, don’t obsess too much on the one-off tactics.


There’s a lot of people who will simplify things in content, saying things like “you should spend 80% of your time on promotion and 20% on creation.” Be careful when listening to black and white maxims like this.

In fact, be careful listening to the advice in this article. It’s simply things that I’ve learned in my specific context that worked for me.

In truth, anything worth doing is difficult and, while you can follow a content marketing playbook of sorts to get a baseline, to truly excel at the sport you need to operate using first principles, systems thinking, and a whole lot of brilliant tactical execution. How that looks on an individual level differs, but it certainly never looks like the mediocre, templatized version of content marketing you often see written about.