Use Chat GPT to Find Top Correlating Backlink Factors

Here’s a quick way to find the best backlinks pointing to your competitors based on correlating metrics.

2 Quick Notes:

1. Top backlink factors vary between industries, search volumes and user intents. So what may work in one arena may not work against another set of competitors. This is why finding a universal approach to this is great. In this example we’re looking at a fiercely competitive home decor & improvement niche with all competitors being multi-million dollar corporations.
2. Factors to rank for a keyword is different than factors to obtain grander visibility. In this example, we’ll be looking at the top correlating factors to rank for the most keywords within the niche.

Okay, let’s dive into this.

For several years, we would supplement our backlink research and analyses with a tool output that looked something like this:

Top correlations (over 0.995):

1. Domains – RefSubNets: 0.9981098226
2. IPs – RefSubNets: 0.993787876
3. RefDomains – RefSubNets: 0.9909022456
4. RefIPs – RefSubNets: 0.9938083682
5. Total Est. Taffic – Branded KWs: 0.9962825406
6. Branded KWs – Est. Brand Traff.: 0.9594640803
7. RefDomains – RefIPs: 0.9958321272

Lowest correlations:

1. Nofollow links – IndexedURLs: -0.3523219202
2. ExtBackLinks – IndexedURLs: -0.366448293
3. RefDomainsGOV – IndexedURLs: -0.001009296609
4. Frame links – IndexedURLs: -0.1834524275
5. Total KWs – IndexedURLs: -0.2845432494
6. Form links – IndexedURLs: -0.08146025318
7. Image links – IndexedURLs: -0.2891032575
8. Text links – IndexedURLs: -0.228580923
9. Follow links – IndexedURLs: -0.2330692509
10. Backlinks – IndexedURLs: -0.2693296251

This was great because it helped us hone in on specific areas of our competitors’ backlink profiles and sift through millions of backlinks. It essentially eliminated wasting time on link types that might have been important in another niche but isn’t for this specific client (ie: gov links). For this niche, it’s a total numbers game.

We also used a chart like this to identify specific topical categories or to identify which competitors have the most “weightiest” backlinks. (ie: ‘Look at Competitor X’s dofollow image backlinks.’)

But nowadays, we can use something like Chat-Gpt, Llama, Grok, etc., or my favorite S.E.Poe and (depending on the prompt & AI model used) you will get something like this in less than 2 minutes:

---

Based on the data, I’ve calculated the correlation coefficients between different metrics. Here are some interesting findings:

• Domain score is strongly correlated with the number of backlinks (r = 0.83) and referring domains (r = 0.78).
• Topical trust flow categories are also correlated with domain score, particularly “Business/Construction and Maintenance” (r = 0.65) and “Shopping/Home and Garden” (r = 0.56).
• Branded KWs and Est. Brand Traff. are strongly correlated with each other (r = 0.92), indicating that these metrics are closely related.
• Total KWs and Total Est. Traffic are also correlated with each other (r = 0.85), suggesting that these metrics are related to the overall traffic and keyword performance of the websites.

Identifying key factors

Based on the correlation analysis, we can identify some key factors that contribute to the domain scores and traffic estimates:

Branded KWs and Est. Brand Traff.: These metrics are strongly correlated with each other and are likely to be important factors in determining the brand visibility and traffic of a website.

Number of backlinks and referring domains: These metrics are strongly correlated with domain score and are likely to be important factors in determining the online presence and credibility of a website.

Topical trust flow categories: The relevance and authority of a website in specific topics, such as “Business/Construction and Maintenance” and “Shopping/Home and Garden”, are also important factors in determining domain score and traffic estimates.

---

Isn’t AI Amazing???

Basically, we need to increase branded search volumes and accumulate a copious amount of backlinks from the home improvement & shopping sites, ideally from ones linking to our competitors.

Easier said than done. But, nothing new.

We still have to verify the responses. But with a table and a couple filters you can have a complete list of effective backlinks that will help your site rank for the most amount of keywords, in your niche. This shaves off a ton of time combing through links and contacting sites that wont really move the needle.

For our client’s – little additions like this save them $$$ & helps them get results faster.

Data used: Majestic SEO Backlink Analysis & Clique Hunter

This task can be used with any tool output (spreadsheet). The more data you can give it, the more informed your response will be.

Let’s let this sink in. We’ll revisit this soon…

Chat with S.E.Poe-SEO-Pro for free here. He’ll give you the perfect prompt to give him, to give you great insights. Don’t forget to use URLs!