Last year I asked myself a question, could I scale SEO traffic by targeting thousands of long-tail keywords? The answer turns out to be: yes.
April 1, 2021 UPDATE: Originally, this post was about 100K visits over 4 months. Data is updated to show performance over 5 months, which shows flattening of growth. I’ll report back again in the next month or two.
If you read up on SEO strategy, the majority of guidance is to write a good blog, get some good backlinks, and build your brand.
While this is sound advice, it’s not the only way to approach SEO, and it excludes other areas of opportunity.
According to a study done by Ahrefs, 92% of keywords in their dataset of 1.9B search queries classify as long-tail, which accounts for 39% of total searches.
Here’s how Tim Soulo from Ahrefs describes long-tail keywords.
Long-tail keywords are terms with low search volumes. Despite what many people think, it has nothing to do with how long or short a keyword is, how well it’s likely to convert, or how specific it is.
While Tim is correct about long-tail keywords, the strategy I will discuss here targets long-tail keywords that are indeed specific.
Search engines like Google group similar search terms together, such as “how to lose weight quickly” and “fastest way to lose weight”.
Google DOES treat specific search terms differently. If you’re living in Toronto, the searches “best restaurant” and “best restaurant new york” are going to give you very different results.
Since each long-tail keyword delivers a tiny bit of traffic, you will need to target many keywords with a lot of pages.
While not all businesses may be structured to use this strategy, my business RecoRank has been able to do so.
If you’re not aware of RecoRank, it’s an application that finds reviews and relevant information for consumer electronic products every day and presents it in a consumable format.
With a large dataset of content that people search for on places like YouTube and Google, I discovered RecoRank was a good fit for building thousands of static pages that all individually targeted specific long-tail keywords.
If you look at RecoRank’s robots.txt, you can see a variety of experiments I have tried. Right now, the VS’s keywords have paid off. The rest, not so much.
Here is what a RecoRank VS page looks like:
I’ll admit, the page isn’t mind-blowing. However, the average session duration for these pages is about a minute, which isn’t bad.
Nevertheless, RecoRank is looking at smarter ways to analyze and present this kind of data, such as using Open AI.
Even without any fancy AI-powered content analysis, the results so far have been pretty amazing. Pretty much all of RecoRank’s traffic goes to these comparison pages, which has resulted in 152K visits and 2.16M impressions from Google in 5 months.
What’s also worth noting is the CTR and average position being really high. This is an effect of ranking for mostly low/no competition keywords.
Google loves original content. But what is original content? A 2,000-word article is. So is a company website. But what about presenting information in ways that are relevant to someone’s search?
Let’s take a look at these keywords:
Here’s what happens when you click on the link to Product Hunt when searching “apps like honey”.
This is an automatically-generated page. Product Hunt has thousands of these types of pages, and they’re all in their sitemap.
Posted March 06, 2021
Help Tavis get to 40 followers on Twitter.