Don’t Worry About Rankings. Worry About These Things Instead.

All of us in the agency SEO world have dealt with a client suffering from Rankings Monomania. The symptoms are always the same:

  • No matter how often you communicate, or by what means, they ask after their rankings in every exchange;
  • They’re usually laser-focused on how they’re ranking for one major keyword in particular, and usually for cosmetic reasons (in severe cases, this keyword will be a nightmarishly broad and competitive one, like “car insurance”);
  • They refract every strategic or tactical idea of yours through the lens of rankings, and in severe cases only buy in if specific rankings improvements are included among the expected gains in your pitch.

This obsession does not make RM-afflicted clients bad at their jobs. It’s not their fault. Rankings do matter. The anecdotal but universal understanding that Google search results beyond page one constitute a wasteland still holds true:

It’s just that rankings don’t matter as much as they used to, or, more accurately, that there are three things about search that have changed since the mid-2000s which have destabilized rankings as a trustworthy measure of success, and which the rankings-obsessed of today fail to recognize:

  1. that strategies explicitly engineered to improve rankings on specific keywords cannot be expected to work as well as they used to in the absence of a larger strategy;
  2. that “rankings” on a given keyword are not universal quantities anymore; and
  3. that even once earned, high rankings in organic search no longer guarantee traffic.

The reason for #1 is the vastly increased complexity of search algorithms, which in their efforts to better understand intent on the input side and to better satisfy intent on the output side have opened themselves to consider many more ranking factors than they ever did, or ever could, in the past.

Hummingbird in 2013 and RankBrain in 2015 revealed stunning new powers of contextualization that Google was able to apply to the queries it was tasked with interpreting, and the skyrocketing popularity of Google’s first-party Chrome browser — which overtook Internet Explorer to become the world’s most popular browser in 2016 — empowered them to bring scads of user engagement data to bear on their rankings as well.

Accordingly, luminaries in our field such as Rand Fishkin began to talk seriously about the core intention of Google Search shifting from “serve the results that give the best possible answer to the question” to “understand the user’s ultimate aim, and serve the results that help them accomplish it.” The upshot, for the sake of this discussion, is that the simple tricks we once deployed to try boost to boost one page’s ranking for one keyword — stuffing the page with that keyword, bolding and italicizing that keyword, building cheap links to that page with the keyword as our anchor text, etc. — just can’t be expected to work in the same way anymore, because we can’t be as sure as we once were of what Google thinks we want, or what it thinks we’ll need to see in order to get it.

The reason for #2 is that search rankings have been subjected to the same kind of extraordinary powers of personalization that have transformed online ad networks, so the search results page for a given keyword will never look exactly the same twice. The rankings on that keyword will change depending on the searcher’s location, device, and search history, and if the searcher is using Chrome and is logged into Gmail, then Google will bring everything else it knows about the person to bear on the query as well: their entire history of searches, emails, and locations visited, the ads they’ve clicked, the YouTube videos they’ve watched, and whatever profile information they volunteered on Google+, if anyone remembers what that is. So rankings are simply not a fixed objective measurement anymore.

Finally, the reason for #3, particularly in Google’s case, is because the appearance of search results pages has changed so dramatically in the past decade — with the additions of paid ads, Knowledge Graph panels, answer boxes, “people also ask” boxes, Top Stories carousels, local search map packs, featured snippets, and all manner of interactive embedded widgets — that there are now queries for which the top-ranked organic search listing exists below the fold:

and some for which Google’s objective is quite plainly to preclude anyone having to click on an organic result at all:

Sidebar: They would just as soon you forget, but a month ago, a smattering of bone-chilling SERPs hiding all organic results behind a “See All Results” button cropped up in response to the kinds of concrete research queries that typically fire answer boxes; these were flagged by SEO practitioners all over the world seemingly at once, including UpBuild’s fearless leader:

Danny Sullivan faced the music on this one, first defending it and then backpedaling. Google shuttered the “experiment”. But we will remember.

So Google doesn’t interpret your queries as narrowly and superficially as it used to, doesn’t rank organic search results as simple-mindedly as it used to, and doesn’t serve organic search results with anywhere near the same consistency and visibility as it used to. Therefore, rankings are more mysterious and mercurial than ever before, and they matter less from a reporting standpoint than ever before.

The bad news about this is that it means we all have to shift focus, starting… well, about five years ago. The good news is that back when it was so easily gamed, the rankings chase felt like a shallow pursuit, as did the tactics wielded in service of gaming it, so a shift in focus will do us good. With Google having gotten too smart for any of us to fool, we can finally dismiss that bag of cheap tricks that even the proudest SEOs once embraced as a necessary evil and commit ourselves to work of substance, that might help our clients while also making the larger web a genuinely better place to spend one’s time.

That’s the capsule history lesson that you’re more than welcome to share with any clients of yours who inquire as to why you insist rankings aren’t the core metric that they used to be. But what do you tell them when they respond by asking what metrics they should be chasing instead? And what kind of substantive strategic ideas might emerge from putting your focus elsewhere? Could a new perspective on SEO give you more than just a few talking points? I’m glad you asked!

Here are some things to start taking seriously, if you’re not already.

The Number of Keywords You Rank For

I’ll admit to some especial amusement writing this post in 2018 given that I can so vividly being asked, during the interview I had for my first-ever agency job in 2012, how I would respond to a Rankings Monomaniac (which was of course less widely seen as a problem back then but which the prescient leaders of that agency already knew from experience was a problem). My reply was something along the lines of “well, high rankings on major keywords are certainly nice, but they’re only nice insofar as they furnish traffic, and there is more than one way to skin that cat. Accordingly, I would tell this client that we’re interested in pursuing organic traffic on a broad range of keywords, such that rankings on any single one won’t matter so much”. I ended up getting that job, so let’s start there.

I cannot imagine a business so narrow in its focus that it would have to live or die by the rankings it could earn on one keyword, or even one topical keyword group. That just isn’t a thing. There are always longer-tail variants corresponding to other funnel positions, keywords concerning support and usage of the product or service, keywords that situate the product or service in some kind of cultural or historical context, and many many more. To paraphrase every hack’s favorite nugget of election wisdom, it’s the traffic, stupid, and traffic depends just as much on how many keywords you can rank for as it does on how high you rank for the most popular ones.

How to get started. Expand your offering, start a blog, write some rich help pages, launch a forum or community feature to kickstart UGC, and/or figure out what gaps in your competitors’ service you can fill so you can speak to those on your core pages. Content strategies range far and wide, but the prevailing truth is that there is always more you can say.

Organic Clickthrough Rate on the Keywords You Rank For

In keeping with my adequately remembered self-quote from the paragraph above, rankings are only valuable insofar as they drive traffic. The point of the high ranking is to draw a click and land a person on your site, and a high ranking alone won’t make that happen. The appearance of your site in its ranking position — the search snippet — is what makes that happen.

The value of a good organic CTR figure was always evident, but the beautiful thing about focusing on this work now as opposed to ten years ago is that as your CTR climbs on a given keyword, your ranking on that keyword will likely climb too. As to why this is: it’s to do with Google’s newfound interest in task completion, as cited above. When someone chooses to click your snippet on a search results page instead of one ranked higher, that person is placing a bet that despite Google’s ranking, your page is going to do a better job satisfying their query. If you get that click and the user does not bounce back to the SERP soon after to try a different result, you’re going to be seen in Google’s eyes as having satisfied that user’s query the best of the choices it made available. If enough users make this choice, Google will revise its rankings to make the public’s preferred page more visible.

As to how this is: CTR became eligible as a new ranking factor owing to the aforementioned market dominance of Chrome and the wealth of engagement data it funnels to Google. Now more than ever, they know where you’re clicking and what you’re doing next. So they can start sculpting and refining their search algorithm around that knowledge.

How to get started. Arm all of your key organic search pages with:

  • first-rate metadata — i.e. a page title and meta description that are accurate, precise, interesting, and seductive;
  • structured data markup from Schema.org — if it at all applies to the page in question — for the sake of rich snippets;
  • page copy — and ideally richer assets like video or an interactive feature — that will keep users stuck there instead of bouncing back to the SERP to click elsewhere (this will boost your engagement figures among all logged-in Chrome users, which these days is a whole bunch of users);
  • easy access to a conversion point, the SEO value of which I will explain in the next section.

Wondering which pages to work on first? Here’s a sequence to follow:

  1. Look at your top landing pages in Google Search Console, sort them in descending order by impressions, select one at a time, and look at the keywords driving the impressions to each.
  2. Among the top twenty impression-driving keywords for that page, look to see which ones have the poorest CTR figures.
  3. Query each of those keywords in an incognito window and see what the pages just outranking yours look like. What do they got that you don’t got? A meta description with a compelling CTA, specifying a reward of some kind (e.g. free shipping, or 20% off to new customers)? Review/ratings markup?
  4. Learn from the success of the pages beating yours, or better yet, improve on what they’re doing so greatly that you make them irrelevant.

The Conversions Driven by Key Search Pages

Much as the best thing you can achieve from a high ranking in organic search is a site visit that you wouldn’t have gotten otherwise, the best thing you can get from that visit is a new customer. Accordingly, the best way you could get a click on a page in search results to serve your bottom line is to optimize that page for whichever conversion most closely aligns with the content.

And hey, if you’re an ecommerce site selling the product the searcher wants, or are providing downloadable assets containing the knowledge the searcher seeks, there might be reason to believe that improving your conversion rate on the page in question could improve that page’s ranking for that keyword. After all, the more people click your snippet in search and stay there to the end of their journey (the so-called “long click”), the better a job you did satisfying them, hence the better job you are likely to do for the next searcher, so by Google’s reckoning, your ranking should be revised upward. If nothing else, the more visitors from search hop into your funnel from a given landing page instead of clicking back to the SERP, the better that page’s engagement metrics get, so it can only help.

How to get started. Take a fresh view of all your key search pages from the conversion funnel perspective. Where is each one situated? Which ones are intended to raise initial awareness of your product or service, which ones are intended to enrich existing awareness, and which ones are intended to close the deal? How easy and enticing do you make it for a user to hop into that funnel from that page? Set up rich custom event tracking through GTM to track user journeys through your site. If they’re not entering the funnel, figure out where they’re going instead, and why. Get into CRO testing and see what you can change to optimize your pages for conversions without compromising the message or the user experience. This can be a long iterative process — in addition to being far too complicated to cover here in full — but the payoff is potentially enormous.

I hope this post has armed you not only with a few parries to deploy in conversation with clients suffering from Rankings Monomania, but the kernels for some deep, careful, and long-view SEO strategies. After all, our job isn’t primarily to react. It’s to lead.

Written by
Driven by a deep fascination with the philosophical and cultural implications of search, Will has applied his skills to improve visibility, traffic, and conversions for hundreds of sites in his 10+ years in SEO and analytics.

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *