- Content Marketing
- Engineering Management
- Lead Generation
- Org Design
- Technical Debt
- Technical Developer Content
Early Stage Content Strategy: Earn Trust, Manage Risk and Build Momentum
In this Speaker Series presentation, Erik Dietrich shares how to establish (or refine) content marketing KPIs specific to your business, hypotheses and experiments to reduce uncertainty for those just starting out, and future-proof approaches for teams of all stages and sizes.
- Early Stage Content Strategy
- The Business of Content
- Why Should You Listen to Me?
- Marketing Funnel: How Hard Can it Be?
- How to Improve on the Funnel Metaphor
- The 2 Critical Marketing Moments
- How to Think in Conversations
- Generating ROI with Content
- Common Early Approaches & Successes
- How to Turn Assumptions into Variables
- Building Your Funnel According to Conversations
- The Basics of Funnel Metrics
- Tools & Tactics for Improving Measurements
- Questions to Ask Yourself: Defining a Category?
- Questions to Ask Yourself: What's the Price Point?
- Questions to Ask Yourself: Who is Involved in Purchase?
- Questions to Ask Yourself: How Easy is it to Be a User?
- Questions to Ask Yourself: How Painful is the Problem You Solve?
- Questions to Ask Yourself: What is the Time to Value?
- Future Proofing: Understanding the Cone of Uncertainty
- Future Proofing Tactics
- Practical Takeaways: Avoid Stylistic or Tactical Emulation
- Practical Takeaways: Create an ROI Hypothesis
- Practical Takeaways: Map Your Conversations
- Practical Takeaways: Deploy Tactics to Tune & Fix
- Practical Takeaways: Self-Assess for Blind Spots & Bias
- Practical Takeaways: Look for Efficiency and Savings
- How Much Time and Energy to Commit to Writing Blog Posts?
- How Should You Shift Gears when your Traffic Numbers are Low?
- Is Backlinking a Viable Content Strategy?
- What is the Cost Breakdown of Working with an Agency?
- Thought Leadership vs. Company Content Marketing?
- Blog Posts and CTAs: How Long Should they Be?
- CTAs in Content vs. Elsewhere on the Website?
- Preferred Channels for Dev Content?
- Are Listicles and Templatized Content Worth It?
The Business of Content
Erik Dietrich: Today I'm going to talk through the business of content, touching on subjects that I particularly enjoy; game theory, casinos, to some extent, and return on investment. More specifically, channeling my own experience, starting maybe five years ago, learning about the digital marketing funnel, a simple, more actionable way to think of the funnel and one that is aimed at reducing decision fatigue so you don't feel like you're boiling the ocean in planning out a content strategy, and achieving that with metrics, modeling, and ROI hypotheses.
And then I'll round out with some actionable takeaways. Hopefully things you can just go and do pretty quickly.
Why Should You Listen to Me?
Before I do that, I feel like I'm an introvert. So I'm not really big on talking about myself, a little awkward, but in your position, I would want to know why am I listening to this person. So very quickly about me. I had a traditional CS background. Bachelor's and master's in computer science.
And the first 10 years of my career was the software engineer career path. I eventually became a CIO and then left and became a management consultant specializing in static code analysis, which was a lot of fun as a practice actually. So that's a pretty non-standard but understandable linear career arc that in no way involves marketing.
So it explains the ROI, the management of software, but where did the marketing come in? Well, in a term I would learn later, you could call me an engineer influencer for a long time. So I started a blog, built up a pretty good sized following, created content for pluralsight.com video courses.
So I did a lot of content and eventually had a couple of published books. And I have this slide in here, which is, it came up in my memories or something. But anyway, this was fun because it was the apex of my influencer stuff, where some rant I wrote earned exactly 1337 upvotes on this platform.
So during these years, about seven years ago, I was creating an awful lot of content. And what wound up happening is when I wasn't writing rants, static analysis companies were asking me to write content for their blog. I had nothing better going on during my 100% travel gigs in the Courtyard Marriott.
So I would write these blog posts and the more of that I did, the more of them paid me to do it. So eventually I had a brainstorm like, "Hey, if enough people pay you to do something and then there's probably a business."
So I teamed up with my wife and we started this content business. And over the last five years have grown it pretty significantly. So we've had a lot of relationships, generally, some pretty longstanding, long term ones. A lot of our companies, our clients are venture-backed startups. Not all, but we have had relationships throughout the life cycle of funding and even acquisition.
So that's, I guess, what I've come to know. So from the engineer's perspective, starting a business and then spending the last five years learning a lot about digital marketing campaigns, in conjunction with some of these businesses and reflecting on that learning.
Marketing Funnel: How Hard Can it Be?
So to go back to the beginning here, if I think about my own journey, channeling my consulting background, one of the first things I wanted to start doing as we worked with companies in the early going was understand what we were doing. What is the marketing funnel? And man, was that overwhelming.
So this one, if you just Google "marketing funnel," this type of stuff will come up. Here's one with six different stages and all these pieces of advice on content to create. This one's got four and it maybe disagrees with this one a little about some of this. Here's one that talks about measuring in four different stages. And this one is the ToFu, MoFu, BoFu, maybe disagrees with some of the others.
It's easy to get overwhelmed and then to think, "Okay, I should just create some content and get it out there. That'll probably work, but maybe I should look up this thing called the funnel, the marketing funnel." And then you get to be able to name the stages.
And as you work your way down, it's easy to get twisted around the axle, fall into analysis paralysis, and then to say like, "I might just create content and worry about all this later." I'm not advocating a "just create content and hope for the best" approach. I'm saying, I would understand it if all of this is what you're trying to figure out.
How to Improve on the Funnel Metaphor
So the problem, I think with relying on the marketing funnel too much as a metaphor, isn't that there's anything wrong with the metaphor of a funnel. In fact, I think it's a great one in a narrow context, but it conflates three different concerns. And those are the ones I want to simplify and dive into today.
So if we're talking about the buyer journey, the conversation that you have with people, if we're talking about how you invest money and what kinds of content you create, and if we're talking about how we're measuring success, those are the three different things that we're cramming all into one here. And then by the way, a lot of this is context free.
You've got to understand that not all businesses are the same. Your customer bases are not all the same. So you're going to have different shaped funnels to begin with.
So let's deconstruct a little and talk about three different metaphors for content and content strategy. The buyer journey, and I like this metaphor, because it's just a conversation. This is like how you make friends with your prospects. The investment, I like to think of this as being in a casino. This is where the game theory comes in.
And finally I do like the funnel, but mainly for measuring success in your efforts. So let's first talk through the buyer journey. There are two critical moments in your digital marketing, in structuring campaigns.
The 2 Critical Marketing Moments
There is the moment when you first get someone's attention and then there is the moment where they trust you enough to give you money and do business with you. And I want to point out that these are rarely the same moment and I can't stress that enough.
So if you don't believe me, I'm going to pause, throw it over to today's sponsor Raycon. I mean, you've seen this before. So I'll just wait here while you go buy these. Except I'm guessing that was awkward because that's not why you're here. I appreciate you putting up with that ruse.
The reason I did that is to demonstrate that this is how users feel, if when they first get to your site, you start trying to sell them stuff. So I did that to underscore how rare it is for the moment of awareness to become the moment of purchase or at least trial.
Understand that what you really need to do is structure something that's more like how an actual human conversation works. Think about getting their attention and then think about being interesting and earning their trust over the course of some time until they decide to do business with you.
That's the overwhelming majority of the time how this is going to go. So what we're really thinking of, is this conversation that happens over the long haul, where you give them free value and you hope they come back.
How to Think in Conversations
Think of it like a conversation like making friends. You're doing stuff for free, you're spending time with them, you're making them laugh, having a good time. Not everyone is going to come back, but a lot of them will if you do a good job of this. So you want to start out by getting their attention and that's really the most critical thing.
When we're thinking of this conversation, if you're in the early going, you can figure out the rest of it later--how you're going to talk to them with email, sequences and all this other type of stuff. Get their attention first, add value first, and the rest will follow more easily. So that's the conversation metaphor. Next up is the casino. This is fun for those of you with STEM type backgrounds.
Generating ROI with Content
When we're thinking about creating content and structuring this conversation, there's all kinds of plays that you can make. You're going to maybe create a podcast or create blog content or whatever the case may be. And you don't know that these things that you're doing are going to work. You can't bring that risk to zero.
So in a sense, your investment is a series of bets. And you are not going to treat ideally each piece of content that you create as some kind of project, in and of itself. It's more like a hand in poker or a game in the casino. Your goal isn't to win every hand or agonize about every hand and it isn't to try to find schemes to create moonshot one-in-a-million payouts.
What you really want to do, if you're in this casino, is you want become the house. You want to create a strategy that stacks the deck in your favor and then start playing enough hands until you win.
Common Early Approaches & Successes
So to illustrate that, here are some things that I hear all the time if we're in the early stages of talking to folks, especially if you're just getting going like technical founders that have raised a seed round or something where there's ideas for strategy and for approach. And it might be things like, ranked in order of, I will say maturity here.
Somebody I know got traction on Hacker News, so let's do that. Or my first article went viral, it's a little better because it's based on your data or maybe we just start putting out good content and see what happens. And for what it's worth that's what I did as an influencer. I like to think it was good content, but I just started writing and it paid off over the years. And then finally, I want to bring organic traffic to the site.
So briefly, if we're talking through the casino metaphor, why I like this is because it grounds things. So if you're saying, "Hey, I know somebody got traction on Hacker News. I want to do that." That's like, given a lot of folks I'm guessing around the bay area, "My buddy went to Las Vegas and won at roulette, so I want to do that." I mean, yeah you can try that. I wouldn't. Likewise, if you wandered into the casino and won early on at roulette, that doesn't mean you're going to keep winning.
Number three is interesting, because this is like, "Let's go to the casino and let's just see what happens." And you might win that 30% of the time, 40%, you come out ahead so it can work. It's not the best.
The best is number four. "Let's start a business, a casino, if you will. And we know the rules, we win more often than we lose. And then it's just a matter of playing enough hands." So number four is really where we want to get to. And to illustrate this with maybe a more concrete example, this is how we generally model campaigns when we're working with businesses.
So let's say you think I'm wrong and you think Hacker News is really the way to go. Let's assume that you're going to commission articles around $500 a piece. And we think that there's maybe a one in 20 chance of getting viral traction. I picked that number based on anecdotal experience of my own site. In the success case, we'll get 25,000 views and in the failure case we'll maybe get a hundred, your current number of subscribers, let's say.
So what's the expected value of this equation? It's 1300-ish visitors per article. So we can reason about the cost per visitor. And if we assume one-in-a-hundred visitors becomes a lead, this is the cost of a lead. Now this is very back of the napkin math, I will full-on admit that.
So you might look at this and think, "Erik, you're way out-of-date. You haven't been on Hacker News in seven years. Your model is trash. It's going to be way better odds than one-in-20." Great.
How to Turn Assumptions into Variables
We're going to turn our assumptions into variables in the model and then have a discussion. So, okay, let's do it your way. One-in-four odds here instead of one-in-20. So that's a 0.25%, 0.25 chance and then 75,000 views in the success case. So we've created an argument, if you will, and we're going with that. It creates an upper bound and a lower bound on what we might expect.
In this model that we've revised, this is actually pretty comparable to the control, which is organic. And by the way, I'm not going to dive into this in the interest of time, but that's a much more stable one that we have a lot of data on. So we have turned these assumptions into variables for the sake of discussion and we're going to proceed.
The model is really about creating an ROI hypothesis where we think we can see a path to earning a return and then prioritizing the approach based on whether it seems feasible or not.
So what you do once you start producing the content, feed it back into the model. So if we play four hands, do any of them go viral? Do we create 10 articles before one goes viral? Well then we can feed that back in and say, "Hey, it looks like there's about a 0.1 chance. So we're going to update these figures accordingly."
And this has some nice traction in the world of this casino metaphor, because you are taking these hands and you're learning from them and feeding them back into the machine. And you just keep doing that throughout and adjusting your expectations and making sure you're earning a return.
Building Your Funnel According to Conversations
Now, speaking of return, let's talk about the funnel. I'm not going to walk through each one of these bullet points per se, because I think you get the gist. But the idea of a funnel, the reason I like the funnel as a metaphor for measurement, is it captures the idea that you're going to have more people's attention than are going to visit your site. You're going to have more site visitors than are going to become leads and so on down the line.
So you can start to look at the model that we created for instance, as a crude funnel. Meaning when we're thinking about the money that we're spending and the number of views we're getting, the success views is going to be the number of visitors.
Down here, we've made an assumption that one-in-a-hundred visitors becomes a lead. So hey, that's a funnel. The top of it has a hundred people, the bottom has one. And what we can start to do as we proceed is adjust our expectations and then use that to try to improve the different pieces of the funnel.
The Basics of Funnel Metrics
So some terms of art, I guess in this world, as we're modeling things out, impressions is generally the number of people in any medium that are exposed to your brand. Visitors is the number of people that land on your site.
Leads, there's a lot of different ways to talk about leads, but you could think of that as how many people trust you enough. They feel warmly to you and maybe you want to become a customer. Qualified leads in general is how many of those would you want as a customer? And then the conversion rate is this composite metric that just takes conversions from one stage to another.
So it could be, holistically, how many people that become aware of you ever become customers. Or you could have something very granular and specific, like I've written a blog post. At the bottom I have a call to action to join the mailing list. How many people read that blog post and join the mailing list is that granular conversion rate. Acquisition cost for something in the funnel is basically what is your spend to get people to a different stage in the conversation, stage in the funnel.
And finally your customer LTV is closing the loop, like how much revenue does turning someone into a customer feedback into the business.
Tools & Tactics for Improving Measurements
I'm not going to walk through all this in detail, but it does speak to the next couple of slides.
It's really hard to measure these things. That's why marketing, I mean, for among other reasons, it's so hard to model this and to get good data is because at every stage here, the data is really hard to get and hard to reason about.
Some things you can do to make that easier in the early going are install Google Analytics, measure visits, measure onsite behavior. It's not perfect, but it's pretty good. Google Tag Manager can help you with some automation, some measurement. It lets you pixel people, which means you can remarket to them through media like Facebook or LinkedIn.
If you're so inclined, HubSpot or Marketo have better tools for data gathering and automation. We actually have a partnership with HubSpot. So I have some insight into what you can do there. I would also recommend an abstraction over the whole thing, which is a dashboarding tool for tracking the different metrics and seeing how you're doing so that you can adjust.
And finally good old Google Sheets, as you can see. Use that one a lot for the back of the napkin math and then even more refined models as we go.
Tactically in the funnel, I can't stress this enough, create content on your own platform and then syndicate it to other places. I would not create content on other platforms as a primary way of getting attention.
What I've got here is a screenshot of dev.to. For anyone familiar, I ran an experiment in 2019 where I took articles I had written on my site DaedTech and all I did was syndicate them on this platform. It earned me about 5000 followers and 44,000 page views. I didn't have to create original content.
I still had DaedTech and all the tracking there, but I was able to get good use out of another platform. By the way, if you're going to do this, be a citizen on that platform. You don't want to just spam it with content.
So the point is, you can make good use of syndication, but you really want things to be originally under a site you have a lot more control. I would also suggest installing analysis and data gathering tools as early as possible. If you fail to gather data for a year, you can't go back and magic that data into existence. So start gathering it early and often.
And finally pixel visitors, as soon as you can, because even if you don't know how you want to continue the conversation just yet, at least you have the option later, you could remark to them.
Questions to Ask Yourself: Defining a Category?
So questions to ask yourself when you're thinking about your conversation with your prospects and your funnel. These are things that are going to define what your funnel, what your conversation is like compared to other businesses. So first of all, are you defining a category?
And the thing that I came up with here is, imagine the year is 2008 and functional programming isn't very big and you are offering a functional programming tool. You're going to have to educate people to what that even is before you start selling it to them. That's a long conversation that requires a lot of trust. Let's say instead, you're just defining auto-complete for Syntax or something. Very easy, might not require as much of a conversation.
Questions to Ask Yourself: What's the Price Point?
What is your price point? This is another fairly straightforward one. If it's really expensive, if you're selling, the example I have here is, let's say, I don't know, 1.5 million dollar value stream management platform. You're going to have a longer conversation. And the bottom of your funnel is probably going to be something more like scheduling a sales discovery call than somebody swiping a credit card.
Questions to Ask Yourself: Who is Involved in Purchase?
Who's involved in the purchase? This is another big one. Do you want to go with a bottom-up type approach, marketing to people that are actually going to be your users? Or a top-down approach, talking to the CTO? Or a little bit of each? Is there a buyer committee? Are all of these people involved in your purchase decision? Is it even something like, say, a headless CMS, where you have to get marketing involved? The more people that are involved in purchasing, the longer your conversations and the more multi-threaded--You're going to have to pursue different threads of campaign.
Questions to Ask Yourself: How Easy is it to Be a User?
How easy is it to become a user? So do you make linter or a mass scale ETL solution? If it's really easy to become a user, if it's a friction-free process to sign up for a demo or a trial, you might make that happen with less conversation. If it's really hard, not only do you probably need more education and upfront value in conversation, you probably also need to think about content to help with the activation once they do sign up.
Questions to Ask Yourself: How Painful is the Problem You Solve?
How painful is the problem you solve? I mean, this is an extreme example, but if you are helping a million dollar a day, e-commerce site not crash, that's a very painful problem if it goes wrong. You're probably not going to need to convince people not to lose a million dollars a day. That sells itself.
If, on the other hand, you're doing a little bit of lifestyle improvement for an engineer using an API, you're probably going to need to build a lot more trust with those engineers to get them to want to do business with you.
Questions to Ask Yourself: What is the Time to Value?
Finally, what's the time to value. Can the purchase payoff today immediately, or are you going to have sales engineering involved? There's a six month adoption and training cycle, et cetera. So again, if you are having a long lead time to value, more content in the front and then once they are activated, you need to think about content there as well.
Now let me go into future proofing a little. So you're thinking about building out this funnel, or you're looking at your funnel. Maybe you have imperfect knowledge. What can you do anyway without getting into analysis paralysis and looking at all the funnels?
Future Proofing: Understanding the Cone of Uncertainty
One of the big things I like to talk about, and I brought this along from my management consulting days, is the cone of uncertainty. You're going to know more later, you're going to make better decisions later. So one of the things you can think about is deferring decisions whenever possible, until what's called the last responsible moment.
So if you can defer the decision, go for it and you'll know more later. Otherwise, if you can't, for instance trying to go after a long term, content play like organic search, go after something that's more blue chip where there's a lower chance of being wrong.
Future Proofing Tactics
For instance, here are some tangible ones, if you have early stage employees or founders that want to write about your manifesto or your value proposition, product marketing type content, that's unlikely to change a lot, go for it. That's not going to be wrong later. Get that content up on the site. We can always solve the problem later of how to get traffic to it.
Organic traffic is a great early play. It's like compound interest. There's no better thing in the content world that embodies that idea of the best time to plant a tree was 20 years ago. So build organic traffic only if you are fairly confident that you're going to be talking to the same segment a year from now.
So if you are confident that right now, you want to go after data engineers and in year, you also want to talk to data engineers, organic's great. If however you think you might pivot away from data engineers to ops people, then maybe hold off. But all you really need for organic is to make sure you're basically talking to the same people.
Another thing I really like is running paid ads or promotions because you can turn off the money spigot just about any time you want. It is a very flexible play. And it also feeds into learning in the early going. If you've put up a landing page and you're wondering if people like it, but you have no traffic to your site, run some ads. And for not that much money, you can run a little A/B testing and answer some of your questions.
And then finally in general, raise awareness of your brand as early as possible. Get that mind share with people. You can always figure out the rest of the conversation later, so you don't need to wait until you have an email sequence and a course and a lead magnet and all that stuff to get in front of people. You'll be able to continue that conversation later. All right, so I'll round out with some practical takeaways, based on my experience over the last five years of talking to folks and seeing how it goes.
Practical Takeaways: Avoid Stylistic or Tactical Emulation
The first one is, as you're thinking about your conversation, as you're thinking about your content plays, avoid what I'll call stylistic or tactical emulation. So I've got the little owl here saying, "I like GitHub's blog. I want ours to be like that." I get that from an editorial perspective, but a business that you're emulating has made decisions getting to that point for specific reasons.
So usually the editorial decisions, the tone, things of that nature are undergirded by business drivers and your business isn't their business. So I would hesitate at the idea of doing that style of emulation, unless you can articulate why it helped who you are emulating.
Practical Takeaways: Create an ROI Hypothesis
Create an ROI hypothesis. I can't stress this one enough. This at the bottom here, I say, you don't need to know, you just need to have a hypothesis. It doesn't have to be perfect, but you should have some idea about how you're going to make this make sense.
And that idea can come in the most simple way with ROI, which is, "We're going to create content for three years at a spend of half a million dollars. And we think it's going to bring us a million dollars in revenue, let's say." That's an ROI hypothesis that's very straightforward.
But it could also be something like, "We're pretty sure that if we get a hundred thousand visitors to the site, we can convert that into 10,000 daily active users. And that's going to help us raise our next round of funding." It's still a business driver. It's still important. It's a little different than the first one, but perfectly valid to go after.
And then another common thing, maybe the second most common thing that we'll see is opportunity value. So you're spending $10k/month on PPC. You think you can bring that spend down to $3,000 or $4,000 if you were to use organic over the long haul. So you're creating a cost arbitrage. That's also a great ROI hypothesis. So again, you don't need to know, you just need to start with that back and then napkin math. I think this is how it'll play out and then tune as you go along.
Practical Takeaways: Map Your Conversations
When it comes to the conversations and then subsequently the funnel you're using to measure success, lay out a roadmap. So how are you going to get people's attention? And then what are you going to do to keep it? You're going to get them to the site.
And then are you going to remarket to them? Do you want them to sign up for a mailing list? Start to model out how you're going to have that ongoing conversation, which lets you establish checkpoints along the way that you can measure. And then figure out where maybe you're having trouble.
Practical Takeaways: Deploy Tactics to Tune & Fix
And then once you see where you're weak and where you're having trouble, then sure, absolutely deploy tactics to address that. But I would definitely suggest when you deploy a tactic, start with not only a success criteria, but fail and pivot, and then a time box or bailout, meaning if you're trying something, hope isn't a strategy.
Don't run something, I think this is going to improve our conversion rate and then just three months later, hope it eventually starts. There's a time box when you should stop doing that. Or there's maybe something you're seeing that indicates failure. So you can stop, back away, and deploy a different tactic.
Practical Takeaways: Self-Assess for Blind Spots & Bias
One big one, especially in the early going, I would say self-assess for blind spots and cognitive bias. I'm not accusing anyone watching of this. I can speak to this from my own experience of building my site and my following.
One big one is that your users slash readers might be a lot less interested in your problem domain than you are. So an example just off the cuff of that might be like, imagine, sometimes I could get into this with the static analysis, that I was talking about gets a little wonky for most readers.
But maybe an easy one to understand is imagine that you make a regulatory compliance solution. You want to get engineers to the site, your message to them probably ought to be something like, "Hey, isn't regulatory compliance a bummer. Wouldn't you like us to solve that for you?" Rather than assuming they're going to be interested in regulatory compliance because they won't. So don't think that your interest in your problem domain translates to everybody.
Avoid the spotlight effect in the early in the early going. This one I totally get.
You're putting that first post up on your blog and you're worried about whether you should use an oxford comma or not. I hate to say this, but that first blog post, almost no one's reading it. It won't matter that much.
The spotlight effect, that biases where you assume people are paying way more attention than they are. So just be mindful of that. Are you struggling to produce content initially because you're too worried unjustifiably, so about what people will think?
And then finally, guilty here, are you teeing up a "dear diary" blog? I shrieked into the void for three years before people started following my blog. And part of that was because it was a hobby. I just like writing.
But be careful that you are not sharing your thoughts and thinking that it's thought leadership when it's really just live journaling, if you will. So the way to avoid a lot of this, is to seek out sanity checks and feedback from neutral parties, who you can trust not to gas you up and to give real input.
Practical Takeaways: Look for Efficiency and Savings
Finally, to close out, a few tactics around efficiency and cost savings.
Once you close the spend-to-ROI loop, so you find something that effectively moves people down the funnel and is working, go hard and go fast on scaling that up because digital marketing moves fast, your competitors move fast, tech moves fast. Your window for that isn't going to be long. So once you've got that signal of success, amplify it.
Understand the vendor to contractor to FTE cost reduction life cycle, if you will. Meaning, start out using a vendor. As you realize success and scale up, contractors are going to be cheaper than vendor partners, FTE is cheaper still than contractors. So have a plan for how you're going to eventually assume and bring these operations in-house.
Deconstruct, analyze, and cost assess. So let's say you're building a podcast. There are different parts that go into that. And you especially, if you are outsourcing the whole podcast to begin with. You might not realize this, but there's going to be guest outreach. There's going to be the production or recording of the episodes, production, editorial, quality assurance, et cetera.
Once you start to get established, break those things down into their components and then ask if you're right sized in terms of your spending on those things. This is something to do as you go along. You'll be able to improve your ROI situation once you're mature and sophisticated enough to understand where your costs are in that cycle.
Batch and eliminate content decisions. This is an important one, in my experience. You don't want to have a weekly meeting with four people to decide what to blog about this week. Do something, go take a look at something like DigitalOcean.
We're going to create all kinds of tutorials. We're going to scale this up, have people write for us. Make the decisions in bulk about what kind of content you're going to produce, so that you don't wind up with just endless decision fatigue that will torpedo your content production efforts.
And finally, the last one, this is fun, we look at sometimes in our, at "subscribe labs," we're calling it--Uncommon plays. So everybody out there is going to write organic content. Everybody out there has a blog, et cetera. Could you do interesting things, like go find Erik from seven years ago and say, "Hey, I want to buy out your site and your audience. I'm going to put you on an earn-out and we're going to get in front of these folks."
Or maybe you go out and find influencers in the space that don't know how to monetize their audience. They would probably appreciate the win-win of you helping them monetize their audience while you're getting in front of their audience. So that's just a handful of things but there are a lot of unique plays out there. Don't hesitate to think and go down paths that aren't well-trod. So that is it.
I will stop there and throw it open to any questions folks have.
How Much Time and Energy to Commit to Writing Blog Posts?
So I would say that, we commission a lot of content through different authors and it depends on how prolific you are with drafting. So one of the reasons I wound up in the business was because I could bang out a 1200 word blog post in 45 minutes. That's probably not common. I would guess based on my memory of historical data with some of our authors, that you're maybe looking at two or three hours to draft a blog post for the average person on a topic that you're familiar with.
How long an editor or a proofreader spends on it is going to vary a little bit by that person. But I wouldn't imagine more than an hour to give it a copy pass. So if you have a founder that's writing a piece of content, maybe two to three hours to create the piece of content, somebody's spending an hour doing a copy pass back and forth and probably in the early going that's enough. Publish it and then have some promotion plan for it. That's another hour there.
I think that if you are spending 10 hours writing a blog post or something, there could be one or two things going on. Either maybe that's not the best medium for you to create content, because you're just not a prolific writer, which is fine. Some people do better at video, podcasts, et cetera. Or it could be that you're agonizing too much.
So if you are creating content, and again, like for a founder, an early employee, you're presumably writing about topics you know very well. So if you are spending hours agonizing over that, you know the topic, well, I wouldn't worry so much about it. You're an authority or an expert. So hopefully that answers the question.
How Should You Shift Gears when your Traffic Numbers are Low?
I want to be upfront about where I get to the edge of my knowledge, which is in some of the particulars of the tools, like if you are going to enlist our help in building out something like this, we would subcontract out a lot of that work. I'm not an analytics expert.
So the way in terms of steering, that's really where I have the most opinions, which is, if you take a tool like Google Analytics and you assume that maybe a third of your users are blocking scripts altogether, that can still give you actionable insights if you have blog post A, that's getting a thousand visitors a month and blog post B that's getting 200. Even if a third of your visitors are being blocked, those relative numbers still add up.
So you can feed the success of blog post A back into the machine. So the relativity is a big thing that we use to steer decisions with imperfect data. And then we'll also normalize our account for that. Are we talking about two different blog-- I mean, I'm making this up off the cuff, but imagine one blog post is speaking to organizational leadership or project managers and another to engineers.
Well then we might want to question whether we're getting a representative sample size, so then maybe the relativity breaks down. So for me-- And especially because often when we're helping clients come up with plans, we don't have the ability to go installing tools or to dig into their server side data. So we're very habituated to making decisions with imperfect data.
The ideal case would be like when you're mining server log somehow are getting really accurate server data. I think that, that can be super helpful. The caution I'd give you is number one, that comes with its own form of maintenance. Number two, it can still be imperfect. So I personally like the approach of assuming that no matter what we do, the data will be somewhat imperfect. And how do we minimize the problem associated with that and correct for it.
Is Backlinking a Viable Content Strategy?
So I guess to give everybody some context, if you're building organic traffic, there's this idea of domain authority, which is a proprietary Moz metric and domain authority, or if you're using other tools, it's a measurement of how much does the search engine trust you. That is really heavily undergirded by how many links you have to your site.
So if you're doing an organic traffic campaign, one of the big success factors is going to be how many people are linking to you. Because the search engine views that as-- If on my site DaedTech, I link out to some Wikipedia article, I'm saying, "That's authoritative. This is a good source of information."
So with that context in mind, there is an idea in the industry around SEO that you should go out and build links to help the search engines trust you more and thus rank you more. 100%, that is an important thing to be doing. A lot of the campaigns that we work with in terms of clients, we're relying on the content itself to attract links. So we do a lot of modeling that assumes no backlinks are going to be built in any way. And then consider it a bonus if people get links.
That said, I would absolutely be looking for links. So great ways to get links for the niche that we're in here, you can syndicate to a site like a DEV or a DZone. So syndicate everywhere that you can. You'll get follow backlinks from those sites. And you can publish something that's not an organic article.
Something that's like thought leadership from a founder and that'll be welcomed on the platforms. You can be a good citizen and get a follow link. In the early going, you want to try to get yourself under review sites. That is a good source of valuable links. Your PR folks are going to be a big help there. So every time they place you in SD Times or CIO.com or something, that's more backlinks.
What is the Cost Breakdown of Working with an Agency?
So in general, there's a playbook that you can execute to help build links. There are also more advanced tactics that you could think about like creating widgets in blog posts. Infographics tend to do this well. Or if you build some calculator widget, like when I was doing static analysis tools like, "Hey use this tool to calculate tech debt." Things like that. Tools, canonical sources of information, just to attract a lot of links. Those are all tactics I would recommend.
On the caution side is hiring firms who do a lot of link building outreach. It's not that this is a bad play. We certainly-- We don't do that or broker that. But I know businesses that do. The danger in tech is that if you do too much of that, it's your target audience that is getting these outreach emails.
So you could find yourself getting roasted on social media if people are spamming engineers on your behalf. So if you're going to enlist a link building firm, I would be careful. Dive into their tactics. What do they do? How do they do this outreach? So hopefully that's helpful.
I've seen and actually talked to some link building agencies about how they do price and usually what they're doing is charging on a per link basis that's a multiple of the domain authority. So it might be something along the lines of maybe for a 30-domain authority link, they're charging you $150, like 5X the domain authority, something like that.
And what they'll typically do is want a retainer where they're saying, "All right, I'm going to build you this amount of roughly ballpark domain authority links per month. So we'll do a thousand dollars a month campaign and then I'll get you five to six backed links at like 50-domain authority." That's pretty common. I think a lot of them will work within your budget, depending on how robust their network of sites is and how good their outreach is.
Thought Leadership vs. Company Content Marketing?
So for a founder, somebody with equity in the business and a stake in it-- I think, I mean, I can relate to this a little bit because I think, more than the business I have now, my own personal brand has more of a following.
I think that typically, if you're building a following like that, you are building relationships, parasocial relationships, with people that follow you. So they're talking in those personal channels about your business and about your travels and your experiences, is going to quite naturally get a lot of people onto the site. So it's a very natural brand awareness plan. It's bringing warm visitors.
So one of the things I do, if a founder with the business brand and then the personal brand, talk a lot about the business and just invite people to check out what you're doing, it's going to get people onto the site. I think it can be as simple as that.
If you're talking about hired employees and they're not something like a DevRel that you're paying to do this, I would think about some form of compensation or consideration for them lending their personal brand to the business. Because the thing I'd always want to be careful about there is, I'm just imagining say I hired an engineer for a business to do some automation and said, "Hey, I can't help but notice you've got 20,000 Twitter followers. Can you tell them about our business?"
If I'm in that person's position, I realize pretty quickly, "Oh, you're asking me for free value." So one thing you might do is be aware of what you would pay an influencer with that size of audience for a mention or a spot or something.
And I'm not saying treat them as if they had no relationship with your company. But be aware if you're going out and asking for something that an influencer would charge a thousand dollars for. That's a pretty big ask. So I'd say if you are doing non-equity type employees, be more careful with that, but I think it's perfectly valid to ask.
The other thing I'll say is a lot of them will just do that naturally. They'll come to work for you, be excited to work for you and reach out to their audience and talk about it. I would just consider that found traffic.
Blog Posts and CTAs: How Long Should they Be?
This feeds into what, I probably rush through a little bit, but where there's no one-size-fits-all conversation or journey through the buying sequence. So I guess to answer the, "Should every blog post have a CTA?"
My answer to that would be no, because there might be-- For instance, let's say in the early going, you want to build up domain authority, you want to rank. CTAs are not directly going to impact your chance to rank per se, but it is a marketing thing on your site, which makes viewers, especially engineers, ever so slightly less receptive to the message.
So one thing that we'll often do in the early going is create content that has no marketing stuff going on whatsoever in it, with the idea that later we can go back and retrofit once the search engine's used to ranking that content.
So I don't think you need a CTA by any stretch on every piece of content and you can also mix and match and see what works. So you start out, you don't have anything, you put a CTA in there, you lose some rank, hey, you take it off. So that's a pretty fluid thing that you can experiment with.
Blog post length, I don't know that there's any particular preferred length, if you will. If we're talking about organic search, you'll see tools like Clearscope or Content Harmony, will make recommendations for how long the blog post should be.
I could probably talk on a length about my opinion about that, but I won't. I will say that my opinion in the broadest sense is that the content should be as long-- Every time somebody Googles something they're asking a question, and if you're ranking for that term, you're answering that question.
So this is a little fatuous sounding, but the content should be as long as it takes to answer the question.
I think generally, more detail tends to be better. If you make a blog post that's 5000 plus words, you probably want to put some navigation at the top of it because that's a lot of text to ask somebody to scroll through and they might just want to get to the point. But I really do think it depends with what you're trying to do.
The blog post that we produce, I'd say average around 1500 words, but we also do our modeling based on what SERPs could we get into for the client with a 1500-word post. Some of that'll be budget too. If you are going out and commissioning content, typically one of the main ways that businesses will price content is per word. So your budget's also going to be a big factor there.
CTAs in Content vs. Elsewhere on the Website?
If you look at a blog post, there's certain places that people like it ad or marketing blindness. If you're talking about a, "Hey sign up for my mailing list," in the side gutter or alongside the post, I think there's very little potential blowback with that. It's expected. I have one of those on DaedTech tech. I think I still have that. So yeah, that's no problem, I would think from a ranking perspective. I don't know how effective it's going to be, but if it's just that you're throwing it up there and getting a lead every now and then, hey, there's no downside. So I think that's fine.
The more interstitial you make the call to action--So one thing that tends to be effective with a ranking article is if you have either different and colored text or different background with a text based call to action, that gets people's attention. The more you get their attention, the more you run the risk of what I was talking about, where people are like, "Hey, this isn't why I'm here." So that's why I really like the AB testing concept.
If you have, let's say, and I've run experiments like this in the past too. If you have 10 blog posts that are ranking and bringing a lot of traffic and they're not generating any leads whatsoever, put in five of them an interstitial call to action, in another five, have a button at the bottom to sign up for the mailing list and then just see what the results look like. And you can take the ones that succeed the most and adopt that approach.
Preferred Channels for Dev Content?
To some extent, it depends on what kind of engineers. There are a lot of communities that you could get into there for Python developers or Rails developers. Those tend to do pretty well. I mean, you could have bigger aggregator type sites that maybe break those down for you. I can think of promotion I've done over the years.
Twitter is certainly a good channel, engineers disproportionately hang out there. LinkedIn, if you're trying to reach IC contributors, not so much, especially because that's, if you're an engineer and first hand, I know this, that's where you go if you want recruiters to spam you. So Twitter is probably the best on social media. DEV, in my experience, is great if you're creating more beginners content, that platform skews beginner.
Dzone is a good one if, and I don't mean to paint with an overly broad brush, these platforms might object me saying this. I think of Dzone as having more of an enterprise leaning bent. So if you're trying to get in front of engineers in the enterprise, Dzone is probably a good one. But I think it's really just a question of identifying the segment as narrowly as you can. So like what engineer and where do those folks hang out?
Another thing that you might do in addition to just pure syndication in channels or promoting on social media, is see if in those narrower segments, you can build relationships with people that curate mailing lists and communities and such, because if you build a good relationship and you make them confident that when you share stuff with them, it actually is valuable to their audience, they'll promote you into a pretty well qualified audience.
Are Listicles and Templatized Content Worth It?
Listicles as an editorial piece of content published on your blog, like, "Hey, we just published the seven best podcasts or whatever." I think that doesn't ring all that true with subject matter experts, so engineers view that as marketing-y.
That said, I think the listicle has a very valuable place in organic search because if people are doing what's called a commercial search, meaning, "I'm in the market for a CI/CD solution, I'm going to Google CI/CD tools and then what I want is a listicle. I want somebody to round up the different options, talk about the pros and cons."
So I would think of if you're going to position listicles, you're probably not going to get a lot of love in direct promotion. But those tend to rank really well and do well in solving searchers problems if they're looking for round ups of things.
Ted Carstensen: That's great. I love it. All right, folks, we are coming up on time. I just want to thank you all for coming and attending and Erik thank you so much for taking the time to put this together. If folks want to get in touch with you and ask other questions or they want to learn more about Hit Subscribe. How can they do that?
Erik: You could send me an email. It's Erik, E-R-I-K, @hitsubscribe.com. That's probably the easiest thing. My old site and brand--I am on Twitter @daedtech, D-A-E-D-T-E-C-H. Fair warning, I'm pretty bad at Twitter, but if you DM me there I will most likely see it too.