How to reliably predict SEO success before publishing content
Only the spammiest of spammy SEOs “guarantee” anything.
“Guaranteed” first position rankings, “guaranteed” traffic estimates, and “guaranteed” ROI only live in the deepest, darkest pits of spammy LinkedIn InMails. And only the unsuspecting, naive, plus frankly ignorant businesses fall for these flagrant falsehoods.
That’s not to say you can’t forecast or predict page rankings, traffic estimates, or search ROI.
Nuance and context make all the difference.
You can – if you know what you’re doing – predict the odds of success or failure ahead of time. Even before publishing a single piece of content or building a single link for a keyword.
In this article, you’ll learn the framework I’ve used to help scale seven, eight, and nine-figure companies into 10-figure companies.
Why ranking takes forever
Ranking for something new can take a looooooooong time.
Anywhere from weeks to months for already-big sites, to years for smallish ones. And it’s why you can’t really “guarantee” much in SEO with a great degree of accuracy.
Here are a few common reasons why.
You need to be a big site to rank for good keywords.
But you need to rank for good keywords before you can build a big site. Therein lies your first dilemma.
Inputs vs. outputs
Rankings, traffic, and sales are lagging indicators.
This means we need to focus instead on leading indicators like keyword and topic selection before we’ll ever see the “outputs” possibly years later.
Selecting “good” keywords involves a balancing act of:
- Buying intent.
Yet, there’s no such thing as the perfect balance. And you often have to either prioritize or sacrifice different elements based on your short vs. long-term objectives.
Less ‘at bats’
Zero-click SERPs, knowledge graphs, instant answers, new advertising slots, People Also Ask questions, blended SERPs, and more mean that:
- You have fewer organic slots to possibly rank in.
- Those are being pushed well below the fold.
The bar keeps rising
Your direct and indirect competitors are getting smarter and spending more money. So it’s an arms race to the top.
All of these problems affect every single company, brand or person trying to rank anything today.
And all of these problems are only going to get worse over the next few weeks, months, and years.
Over a decade ago, ranking was easy. Then, Panda and Penguin and a slew of other algorithm updates changed everything.
So now, we all have to navigate an ever-growing complex world of whatever AI + search monopolies dream up next.
Get the daily newsletter search marketers rely on.
How to reliably predict success
The primary benefit of working with hundreds of companies over the past decade in a variety of spaces and niches is pattern recognition – which is the ability to reliably see what works and what doesn’t to consistently produce results across time or context.
(As opposed to what works one time, for one company, in one very specific context, because they were able to exploit some temporary arbitrage to their advantage.)
And it led to the Planning Predictor framework that we now use for our companies to help de-risk SEO and content planning so that we're only focusing, right now, on what's mostly reliably expected to produce results.
Here's a high-level overview of the five decision-making criteria:
- Harvest demand: What are your customers searching for, when, and why? Perform both top-down and bottom-up analyses to build out initial keyword lists starting with relevance, volume, keyword difficulty (KD), and cost per click (CPC).
- SERP competitiveness: How difficult are the corresponding SERPs, and can you realistically compete based on the average domain rating (DR) and quantity/quality of referring domains – right now and in the years to come?
- Topical authority: Are you already seen as a topical expert in these spaces, or will you need to build it from scratch? (i.e., Are you already ranking for similar queries or hitting a glass ceiling?)
- Organic CTR: Can you realistically rank in the top ~3 for this given keyword now and in the future (assuming increased competition)?
- Payback period: What's your ideal payback period and risk tolerance? (Zero to six months, 6-12 months, 12-18 months, 18+ months, etc.)
This might sound complicated at first. But it isn't when you know how to analyze each one. So let’s get started.
1. Harvest demand
There's a saying in the SEO world: "You can't create search demand, you can only harvest it."
In other words, the vast majority of mortals out there (read: 99% of the people reading this) aren't going to be able to create a brand new category of searches from scratch. It's too difficult and costly.
Instead, most of our job is to identify what's already out there, what's already being searched, and how it lines up with both:
- Our customers.
- Our widgets or products or services.
Think of the typical buyer's journey or jobs-to-be-done process, where you analyze the decision-making process of customers when they:
- Develop need awareness for a problem in their life, to…
- Begin searching for potential solutions and their respective pros vs. cons, before…
- Deciding on purchasing the ideal solution (hopefully yours).
This top-down analysis leads to uncovering all the potential problems, pain points, alternatives, and solutions your customers might face.
Additionally, it should result in dozens to hundreds of potential keywords to target (or potentially thousands if you have multiple customer personas, segments, products, or services).
The next step is to break these down into varieties of variables, like:
- [Customer segment vertical] + [problem or pain point]
For example, [construction] + [project management]:
Then, you want to expand these initial keyword ideas for all the potential long-tail variations, weighing up the respective relevance, volume, KD, and CPC of each.
(Fun fact: CPC, while 100% focused on advertising, can help indicate keywords with higher potential buying intent.)
Another example for "best socks" then results in a giant list of related terms, that you can filter, to help cut through the signal-to-noise ratio for stuff you're most likely to rank for right now (relative to your own site strength).
In truth, this is part art and part science. You're looking to unearth gems. But it'll take sorting through a lot of junk, first.
2. SERP competitiveness
Psssst. Come closer.
I have a dirty little secret to let you in on.
All SEO tools (even the best ones pictured here!) lie.
Not on purpose necessarily, but out of necessity. Because what they’re trying to do – imitate and classify what Google's doing – is insanely difficult. Near impossible.
Case in point: the "keyword difficulty" metric.
It's helpful to understand the perceived difficulty of potentially ranking for a keyword. Nice in theory.
However, in reality (where we all live), it's badly skewed.
For example, a keyword can have a very low KD (like "6" out of 100), and yet not be easy to rank for. That's because it largely focuses on links as a proxy for strength (and page-level referring domains at that).
The average/median page-level referring domains might indicate the keyword is "less competitive," even though the domain-level strength (plus content quality) of the ranking sites are actually very competitive.
Like this example:
That's troubling because it can (and often does) lead people to make bad choices about which keywords to target – when and why – and ultimately overestimate how quickly they'll rank for it (i.e., not as soon as they think!).
As a general rule, if your own site-level strength is significantly below the ones already ranking for a given keyword (like the example below), then forget it.
Or, at least, add another XX months (and more $$$) to your estimated time to rank.
This is a giant problem because position 5 might as well be position 50 today. I'll explain why in the fourth section below on organic CTR.
But first, let's look at personalizing your keyword difficulty based on whether you already have topical authority on these terms.
3. Topical authority
Contrary to popular belief, HubSpot didn't invent "inbound marketing."
Seth Godin did "permission marketing" about a decade earlier. And the Michelin Guide did a century before that.
What's old is new again.
Take pillars or clusters or hub and spoke models or whatever your favorite marketing
guru charlatan comes up with next. All are semantics for describing content hubs that have literally been used for decades.
(OGs like Ian Lurie have been going on and on and on about these since most of you marketers today were still in diapers.)
These pillars or clusters or what-have-yous relate back to site architecture and content hierarchies, creating dense webs of information to help increase your topical authority on certain subjects.
In practice, you should be purposefully planning new keywords and topics in clusters like this, in bulk – vs. a bunch of one-off grab bags – so you can organize it with internal links to develop these silos.
For instance, the monday.com blog organizes topics around specific customer industries. Then, they use a parent-and-child hierarchy to help reinforce topical authority. (Disclosure: My companies have worked with monday.com.)
A rising tide lifts all boats.
Topical authority helps you rank higher and faster for related terms.
But the inverse is also true: little to no authority means it's going to be more difficult (and costly and take longer) to rank vs. those that do already possess it.
So ask yourself:
- Are you already ranking for similar terms? If so, you probably already have some level of topical authority. The more stuff you have ranking higher in a single cluster, the more likely you'll rank faster with brand-new stuff in this space, too.
- You can also use tools like the MarketMuse Inventory feature to analyze "personalized difficulty," which helps predict your own personalized topical authority (and therefore the difficulty of ranking for it) vs. relying on vague, general aggregate scores a random SEO tool spits out.
Nuance and context are everything.
A good keyword for your competitor might not be good for you right now.
And what's good for you right now, when starting out, might not be what's good for you in two years' time when you're trying to make bank.
4. Organic CTR
Position 5 on a SERP might as well be position 50. Why?
Because you can virtually expect the same level of traffic (read: zero, zilch, zip!). Why?
Because based on SERP CTR studies (the number of people who click on different positions), the top three results routinely net around ~70% of the clicks, eyeballs, and therefore credit cards.
(You know, like that whole Pareto thing?!)
A quick glance through Advanced Web Ranking’s CTR studies will bear this out.
So, what does this mean?
- Ranking in the top 3 is all that matters at the end of the day.
- It's often better to pick less popular and less competitive and more relevant keywords so that:
- You have a better shot at success (ranking top 3).
- You can rank more stuff, faster, and shrink the time to value (traffic, leads, sales).
And, it means that to compete for the most lucrative keywords in your space, you need to get big ASAP or go somewhere else and feed off the scraps of the competition.
Got it? Good.
Then why is all of this important?
Because SEO and content marketing comes down to a simple cost vs. benefit analysis:
- What is it going to cost you to rank in the top three for a given keyword?
- And what's the potential payoff if / when you do?
The "international SEO" keyword is a perfect example. (Plus, a solid meta-reference you can't pass up.)
Weigh up the potential volume, traffic potential, KD, number of links to rank, and CPC of this term below:
Now. Ask yourself, can you break into the top 3 within the next ~12-24 months?
To answer this questions, we must look at the SERP competitiveness:
Especially when you consider the loooooooow traffic you’ll get even if you did rank in the top 3. Plus, there's virtually zero buying intent, too.
Even if you thought you had a shot at ranking for this term in the next ~12-24 months (based on topical authority), why would you even bother?
Unless you literally sold "international SEO" services or similar, it's most likely not worth it for 99% of sites out there.
So adjust expectations accordingly.
- Pick a new keyword and topic to target.
- Or, adjust your patience and budget and timeline to build a giant site and enough topical authority it might take to rank for this keyword one day.
5. Payback period
Paid search gets the most love (and $$$) because:
- It's relatively easy to see which specific keyword leads to which specific conversion and how much money you made in return.
- And because marketers + C-suites are lazy. (OK, maybe just "impatient" as well.)
Organic SEO and content, on the other hand, also have a defined payback period. But it's often harder to track because it might take 6, 12, or even 18+ months to see the true ROI.
Exhibit A: the best converting keywords – even for the web's biggest brands – might take years to rank in that coveted top 3 position that drives all the pesos or yen or Bitcoins or whatever:
So, your job is to figure out:
- Which keywords are "worth" the "risk" or investment (literally the first four criteria listed in this article)?
- Will the potential ROI make up for the investment of time, money, and resources?
- Can you really wait that long to see the payoff?
This leads you to a simple decision-making tree:
- If the answer is "Yes," then full speed ahead! Go big and bold, doing hundreds, if not thousands, of new content pieces to dominate your space in the years to come (~18-24+ months).
- If the answer is "Maybe," then it's time to get creative. Realistically, which new keywords and content can you prioritize in the next ~12 months to deliver enough results that provide enough of a positive ROI to double down in the future?
- And if the answer is "No,” then it's time to shift focus onto what you already have – typically updating existing content ranking bottom of page one, top of page two, or adding new content only in spaces you have topical authority – so that you can generate some tangible momentum within ~6 months time.
Because this last part is entirely subjective.
A lot depends on your role, the company you're in, internal stakeholders or champions, the broader market conditions, and more factors that can be outside your control.
In truth, after working with hundreds of companies over a decade, I've seen that most companies fall into #2 or #3 above. Only Unicorns and public companies live in the rarified air of #1.
(Trust me, the air really is better up there!)
Predicting future SEO success is doable
Ranking today is difficult.
And unfortunately, it's only going to get more expensive and complicated over time.
That means most outside of the web's top 1% need to prioritize keywords and content to show results within 6, 12 or 18 months' time.
Showing some short-term results often buys you more time and money to double down later, whether that's from your boss, clients, spouse, or your own gut.
This typically means going back to the drawing board, starting again at the top of this article, and then:
- Adjusting your keyword selections based on potential traffic and buying intent filters down to target less competitive stuff,
- Scrutinizing SERP competitiveness to make sure the domains and content you're up against are as "easy" as the link-biased KD makes it out to be,
- Prioritizing areas you already have existing content ranking or some validation of topical authority in these spaces, and
- Can reliably break into the top 3 by publishing enough, building enough links, and/or increasing content quality faster than everyone else, otherwise
- Readjust your payback period expectations to stick it out for the long haul or compete for more realistic, short-term targets.
It's not easy. You can't "guarantee" anything.
But it is doable. And you can predict future success if you weigh each category carefully.
The post How to reliably predict SEO success before publishing content appeared first on Search Engine Land.