Overall, analyzing 200 sites in my niche was the best SEO training I’ve ever had. No courses, blogs, or articles can replace the real data sitting in the top rankings. I advise everyone to take 200 competitors, export them into Ahrefs Batch Analysis, and start reviewing them site by site—their content, backlinks, traffic, and so on. It took me about 2 weeks, but it was well worth it. I don’t want to expose anyone’s sites or delve into every site’s specific tricks; this article covers only the general parameters of the sample.
I think very few people will actually do this; 99% will scroll past to read other people’s opinions on what works and what doesn’t. That’s why I’m writing this—competition isn’t scary. Very few actually put in the work. Analyzing each site individually is painstaking labor.
Baseline Data
Now let’s move on to analyzing the aggregated data for the 200 sites.
Vyacheslav Varenya wrote a really cool post on Google’s June update — https://pro100blogger.com/2019/06/htobnovlenie-algoritma-google-iyun-2019.html
I strongly recommend everyone read his post.
I got very interested in his methodology, and huge thanks to him for helping me launch AnswerMiner and load the data into it. The data was loaded in the following format:
A total of 200 relevant domains were loaded. I exported the domains via Semrush; it displays competitors better and more completely than Ahrefs.
The sample included all kinds of sites—e-commerce stores, blogs, etc. I didn’t filter by site type.
SEO efficiency is calculated as the ratio of traffic divided by keywords. So if you have 5,000 keywords and 10,000 traffic, the ratio is equal to two.
Ideally, one should evaluate the ratio of queries in top 3, top 10, and top 100 for a more precise analysis, but so far I haven’t seen an export format that allows analyzing this.
keys.so has such a tool, but I have English/international market data, so our efficiency calculation is slightly rough.
The more queries in the top 5, the better optimized the site is, and squeezing extra performance out of it is very tough. It has reached its peak. Niche leaders have a 25% efficiency score according to keys.so.
We export all the data into AnswerMiner and start analyzing the correlations.
Ahrefs DR
Efficiency vs. Ahrefs DR. Notably, DR<43 shows better efficiency. There’s a suspicion that niche-focused sites rank better than massive portals covering everything. Medium, for instance, has low efficiency despite an off-the-charts DR.
Domains
The more referring domains according to Ahrefs, the higher the efficiency:
Domain Authority
The higher the DA according to Moz, the better the efficiency
Citation Flow
If CF<38, efficiency is higher than if CF>38. Could it again be topical niche sites? Just a hypothesis.
Trust Flow
Kinda unintuitive. If a domain has TF<12, it is more efficient than a domain with TF>12.
Study Conclusions
- When analyzing drop domains, it’s best to look at the following metrics: Ahrefs DR, Majestic CF, Moz DA, and total referring domains count.
- Links rule more than ever. Pump the site with links and watch it grow on Google. Cliché? Agreed, but now I have my own mini lab for data analysis.
- If you stuff your site with low-quality garbage like generic «service crowd links,» you might not get any growth in DR, CF, or DA metrics and will later claim links don’t work.
- If you pick a drop domain with weak metrics, you’ll end up saying drop domains don’t work and don’t pass any link juice.
- If you use weak PBNs and place links from them without seeing results, you’ll later complain that PBNs are an ineffective tool.
- We need to somehow extract top 3, top 10, and top 100 rankings for a deeper efficiency analysis. For 200 sites, I don’t know how to do that right now.
- There are questions regarding TF, CF, and Ahrefs DR. TF definitely behaved strangely.
- We should take a group of hyper-niche sites and see how everything looks.
UPDATE:
CF correlates directly with domain count.
TF vs. domain count
TF vs. traffic volume
I think the issue here lies in ROI. Small niche sites will outperform multi-topic sites in terms of efficiency and will pay off faster. That’s my personal take on why TF, CF, and DR can be high, yet return on investment remains low. I could be wrong.
Here’s another cool chart. No links = low efficiency:
