Stories about Software


The Phases of an Organic Campaign

Don’t look now, but it seems like I’m in real danger of finishing this series of posts.  One day, I might even become a blogger again.

In the last few posts, I went into a good bit of detail about how to do keyword research.  This culminated in a post about assembling your research into campaigns and even fully-formed marketing funnels.

But today I want to switch gears.  Instead of talking taxonomies and hierarchies, let’s look at timelines.

What are the phases of organic traffic campaigns?  Or, perhaps more precisely, what are the phases of your site’s relationship with the search engine?

Caveats and Philosophy

Because I can’t resist a good “before we get started” caveat, I’m going to do that.  But I promise to keep it brief.

The first thing I want to mention is that the rest of this post is biased towards maximizing long term value and ROI for your spend on the campaign.  If you’re burning through your runway furiously and need a lot of leads, fast, to survive to raise your next round, I sympathize.  But you’ll need a different philosophy (and probably a different acquisition channel) than the one I’m laying out here.

The other thing to mention is that I encourage you to kick any agonizing, fretting, and general decision-making about anything as far down the road as possible.  You can always refresh a piece of content later, or throw a different CTA on it.  But you can’t rip a hole in space-time, go back and publish a post earlier.

So I likewise I assume a bias toward pushing content early and often.

With that said, let’s talk about the 3 phases of organic, plus phase 0.  (You can take the boy out of programming but you can’t take the programmer out of the boy — of course there’s a phase 0).

Phase 0: Pointing the Shotgun Away from Your Foot

You should spend as little time in phase 0 as humanly possible.  Every second you spend in this phase is a second you haven’t indexed your content and given it a chance to rank.

But you should execute phase 0 in earnest, nevertheless.  It’s essentially a mix of gut-check and checklist, so you’ll go fast if you’re in the right frame of mind.

What’s Your Topic Selection Algorithm?

Speaking of gut checks, let’s take a gulp of air and get the biggest one out of the way first.  If want to decide what to write about every week during some kind of “ideation” session before drafting, give up right now and go home.

That is far, FAR too slow and too inefficient.

In fact, even ideating around topics in groups of 5 or 10 is probably too slow and inefficient.  Thinking about creating organic content as if you were stacking a backlog for an editorial content is a failure pattern.

You’re not picking topics when you start out.  You’re picking an algorithm for topic approval.

Here’s a simple example.

Waldo has been a Hit Subscribe client for around a year and a half.  They make a tool for testing mobile apps.

Do you know who does a lot of testing of mobile apps?  Engineers that build mobile apps.

So what did we do to generate a lot of content and traffic?  Published a lot of tutorials about mobile app development on the site.

Thus we’re not making a decision with each article except “does this or does this not fit into the general heading of ‘mobile development tutorial’?”  Topic ideation becomes dead simple and can focus on the important variables of segmentation, volume, and difficulty and not “gosh, what do I feel like talking about?”

Are You Sure about Your Persona?

Once you have your theme — your algorithm for topic vetting — you’re mostly ready to crank.  Now you have a couple of sanity checks to do.

First, is there any chance that you won’t actually want to reach the segment you’re targeting?  In other words, are you teeing up a bunch of tutorials about unit testing when there’s a chance that you’re later going to want to talk to gardeners in desert climates?

You can figure out a lot of things later with organic, and you can correct a lot of errors.  But segmentation (who is asking the question you’re answering) is the one thing you can’t really do anything about.

So you can start targeting organic as a channel before you even know key things like pricing, minimum feature set, etc.  But you can’t start before you know for sure who you’ll want to talk to when the traffic starts coming in.

The Bare Minimum SEO Checklist

If you’re reasoning algorithmically about topics and know who you want to talk to, the last remaining sanity check is a “don’t shotgun-foot” checklist.  I’ll just throw out a list of the most common mistakes I see that will substantially impact your efforts.

And by this I don’t mean “stuff that could be better,” like a nicer font or something.  I mean stuff that will unambiguously hurt your performance.

  1. Do you have a sitemap?  This is essentially how you tell the search engines about content you publish.
  2. Are you indexing user-generated or other noisy content (like help forums)?  Don’t do that.
  3. Does your URL slug scheme bake dates into the URL or, worse, use URL parameters to serve content?  Do not do either of these.
  4. Is your site performance bad?  Give it a check on Page Speed Insights.  Doesn’t need to be perfect, but I’d shoot for about a 40 on mobile and an 70 on desktop or you run the risk of site performance being a drag on your search performance. (And ideally try to get it passing core web vitals)
  5. Are you using SSL (https) on your site?  Google will actively penalize you if not.
  6. Do you have 404 links in your content or have you deleted previously indexed pages without redirecting them?  Don’t leave the search engine hanging — always be serving and linking accurately.
  7. Have Google Analytics and Google Search Console.  Missing this won’t actively hurt your rankings the way this other stuff will, but it won’t hurt you in the same way that driving without a speedometer or gas gauge won’t hurt you.  You’ll be able to drive, but you’ll make bad decisions because of a lack of data.

Things I Wouldn’t Sweat

I think it’s also worth calling out stuff that I see people get worked up about, but that my experience tells me isn’t overly important.  Or, if it is even somewhat significant long-term, it’s addressable later.

  1. The CMS you use.  I’d personally use WordPress or Webflow because sooner or later, you’re going to hire marketers and they won’t consider it a badge of honor to inflict drafting in markdown on themselves.  But, do you, and whatever you do, don’t waste time on this ultimately migratable decision.
  2. Your TLD or even your brand name.  You can always change this later and, as long as you don’t mess up the migration, change without adverse effects.
  3. Graphic design and aesthetic branding.  You don’t need a beautiful site to answer searchers’ questions, so get started now and make it pretty later.
  4. The launch of a beta or a new feature set.  You don’t need to have positioning and product details nailed down to answer searcher question.
  5. Having an email list or nurture sequence.  You don’t need a list or a CTA to answer searcher questions.  (Though you may want to set up a remarketing pixel sooner).
  6. Social media accounts, distribution strategies, RSS feeds, etc.  Sure those things are good to have but again, say it with me now, you can answer searcher questions without them.

A lot of this type of stuff seems important, but can really just serve as a productive-feeling form of procrastination.  Get the basics in place and start producing content.

Phase 1: Building Search Engine Trust

Once you’re reasonably confident future efforts won’t prove vain, it’s on to phase 1: the first phase in which we actually do stuff.  At this point, you may want to briefly revisit the post where I talk about measuring keywords, because that will feature prominently here.

In this phase, we want to:

  • Optimize for low difficulty
  • Mostly ignore keyword volume
  • Favor tighter segmentation when all else is equal

Phase 1 is the long handle of the hockey stick, and you know you’re in it when it’s taking your content 4-6 months to rank.  Any brand new site with low or minimal domain authority will start here.  But you might also have the dubious distinction of starting here when you have plenty of domain authority, but a long history of publishing and indexing columns, indexing your user forums, or anything else that makes the search engine regard you as a searcher-question dumpster fire.

In either case, we start by going after low difficulty keywords as long as they have ANY indication of searcher interest whatsoever.  And yes, this means going after keywords the tools tell you don’t have any search volume.  You just want to check for a people also ask or a search-engine auto-complete.

Your ideation tactics will be fairly straightforward.  I’d suggest starting with core terms that you want to talk about and filter in ahrefs for difficulty less than 5 or so.  You can also mine people-also-ask questions and answer the public.

What we want to do in this phase is produce a LOT of content, all targeting “easy” keywords that nobody really seems to address well.  The effect is to “train” the search engine.

At the risk of anthropomorphizing it, the search engine starts to think about your site “wow, they sure don’t have much history, but whenever I rank their content searchers LOVE it.”

AP Tip: Response Posts

As a brief aside and addendum, I’d like to bang the drum for response posts here in phase 1.  This is where you find searches that are fully formed questions and create posts with titles that mirror those questions back to the reader.

So, for instance, take this search for Ruby on Rails.

If you look at the people-also-ask, you see 4 questions (or more if you keep clicking on the questions).  The search engine is basically telling you with this widget, “hey, here are things people wonder about.”

You can document these questions and then filter for the ones that seem like they’d make for decent blog posts.  IME off the cuff, the first two would be worth a check for difficulty and then worth answering if these questions were things your audience segment might ask.

But if you were to create the content, you’d want to mirror the question in the title:

  1. What is Ruby on Rails Used For? The Top [N] Use Cases
  2. Is Ruby on Rails Still Relevant in 2022? (You’ll have to update this one’s title and content every year)

The key here is including the verbatim question/keyword in the title.  If you’d like to see a case study on that, you can check out this video I recorded a few years ago about building makemeaprogrammer.com, back when I still thought it was fun to pretend I was a YouTuber.

Phase 2: “Hockey Sticking”

If you have a newish domain and push significant content for about 6 to 12 months, you’ll start to hit the fun part of the hockey stick.  I’m no hockey expert, but I think they call this the blade or the puck-slappy-part or something.

Your phase 1 grind has paid off, your articles are hitting the front page, and your traffic is starting to take off and realize 20% to 50% gains per month.  To make this more visually concrete, here is the shape of organic traffic, typically, for a new site (assume DA of 2), assuming it did 5, 10, or 20 pieces of search-attracting content per month.

As you can see, it’s a pretty serious act of faith in the first year.  Phase 2 kicks in for this hypothetical client near the 1 year mark, and stuff starts to get fun.

Now we reconfigure our thinking with our three core variables

  • Favor low difficulty, but don’t worry too much as long as the parent term seems attainable now or a little later.
  • Optimize for high volume
  • Sanity check segmentation and cast a wide net

This all probably makes sense to you, except perhaps why I recommend being tighter with segmentation in phase 1.  The reason for that is that, the tighter the segmentation, the easier it is to nail the question with content.  We want tight segmentation in phase 1 because it helps with ranking (not because we’re worried about narrowing down the funnel).

In the volume stage, stage 2, we just want to drive tons of qualified traffic to the site because we’re going to use the now statistically significant data to see what converts and to turn the traffic into a serious acquisition channel with a low cost of acquisition (CAC).

Phase 3: Conversion and ROI

The timing of the start of phase 3 is going to vary more than anything else here.  It’ll depend on your appetite for risk, your customer LTV, the strength of your nurture sequences, and a number of other variables.

But it also might flip over right at the moment that someone says, “hey, we’re getting like 300K people to the site per month, but they don’t really seem to convert.”

Whatever triggers it, this starts when you become serious about tuning for maximum ROI.

Now you might wonder why we don’t do this from the beginning.  The answer is actually fairly simple.  The sorts of things that maximize conversion (pop-ups, sign-up toasts, CTAs, marketing automation in general) really tend to piss off searchers.

If you kick off a “sign up for our DevOps Toolkit” pop-up message when you have 300K visitors per month, a 65 domain authority, and a long history of ranking, the search engine responds with a hearty ¯\_(ツ)_/¯ and keeps ranking you.  If, on the other hand, you do that with a new site and a domain authority of 2, it assumes you’ve just kicked off an illegal island gambling ring and refuses to rank anything you produce.

The things that convert in the late stages work against you in the early ones.

But now the traffic is established, it’s time to reprioritize again:

  • (mostly) Ignore difficulty.
  • Favor high volume
  • Optimize for converting your target segment to nurture

Unlike previous stages, we’re not talking entirely about new content anymore, either.  Sustaining traffic will mean producing new content, but we’re optimizing conversion on existing content by adding new CTAs and the like to see what works.

Sustaining and Winding Down

Using organic traffic as a lead generation channel creates extremely high upside, but it’s not unlimited.  Last I looked into it, the number of programmers in the world was around 20 million.  So at 20 months of 1M unique site visitors, you’ve burned through entire addressable audience.

While you might chuckle at the “good problem to have” to have here, it does become a real concern.  Content that you produce has a maintenance cost on a long time line, if you want it to continue to drive results.  And that cost mounts as your addressable market shrinks and existing content grows.

At some point you’ll go into a sustaining mode.

I won’t get too into the details of that here, but suffice it to say you’ll want to have a sense of when you’re getting to this point — when you’re seeing diminishing returns on new work.  The main leading indicator for this is when the projected traffic (and conversions) for any prospective new content starts to sit on par with the cost of maintaining and updating the existing.

Once this happens, think of your content more as a wiki or repo that you keep up to date to serve your audience.

Looking Ahead to ROI In Depth

If my mention of ROI in phase 3 seems a bit hand-wavy, I invite you to have a bit of faith.  Now that I’ve demonstrated that I’m capable of managing to write a post or so per month, you can rest assured that I’ll address that in significant detail in the next installment of the series.

And, in a way, that’s kind of a thematic close to a post about about organic traffic phases.  You can worry about ROI and conversions later.  Get started with phase 1 today and we’ll get to the money tomorrow.