Worked example
Check competitor backlink gaps before you decide they out-linked you
A rival passed you in the results and their link count looks bigger. That proves less than it feels like. Here is a worked example, clearly invented, of testing the claim.
The domains are placeholders and the counts are chosen to show the method. Nothing here describes a real site, and none of it is a result you should expect on your own data.
The setup: three pulls, same day, same depth
Imagine you run a small tooling blog, yourblog.example, and a rival, rivaldocs.example, overtook you on a group of how-to pages. Pull both with identical parameters, then ask where they overlap. Confirm how intersect takes its list of domains in the account page before you script it.
# your list and theirs, same sort and depth
curl -s "https://rankele.com/api.php?action=refdomains&q=yourblog.example&per=1000&sort=dr&order=desc&key=sk_live_..."
curl -s "https://rankele.com/api.php?action=refdomains&q=rivaldocs.example&per=1000&sort=dr&order=desc&key=sk_live_..."
# who links to both? 2 to 10 domains, 2 credits per call
curl -s "https://rankele.com/api.php?action=intersect&q=yourblog.example,rivaldocs.example,thirdrival.example&key=sk_live_..."What the invented numbers say
Suppose the rival shows 900 referring domains against your 400. Tempting. Wait. Domain count is a blunt number, and the useful question is how many of those domains are worth having and how many already link to you.
Now suppose 520 of their domains have a DR under 10. Odd, yes. The intersect call shows only 60 domains linking to you both. Most of the gap of 500 is small sites that probably do little for either of you, so a big headline gap and a thin real gap can coexist, as the table below shows.
Then check the other direction. If the third rival, who still ranks below you, also has 800 referring domains, the count cannot be what separates winners from losers on these pages. That is the quickest test we know for a story that sounds right.
How the example reads once split by DR band
| DR band | yourblog.example | rivaldocs.example | thirdrival.example |
|---|---|---|---|
| 0 to 9 | 210 | 520 | 490 |
| 10 to 29 | 150 | 260 | 240 |
| 30 to 59 | 35 | 100 | 60 |
| 60 and up | 5 | 20 | 10 |
What to do with the 20 strong domains
Look at them one at a time. For each, open the linking page, note what it links to on the rival's site and why a writer would bother. Some are directories, some are citations in a guide, some are partnerships you cannot copy.
Sort them into earnable and not earnable. Perhaps a handful are tool roundups that list several products, which you could pitch. The rest may be paid or personal and are not worth chasing. That short list is your real gap, and it is small enough to act on this quarter.
The traps that turn a gap into a false lead
Domain counts ignore which pages earn the links. A rival with twenty strong domains pointing at one how-to page can beat you on that page while looking small overall. Run toppages (2 credits) on both domains and compare the targets.
Index coverage is uneven. A tool that has crawled one site deeper than the other inflates the gap. Check the same pair in Ahrefs or Semrush, then trust only the pattern that survives both.
Rank has other drivers. The rival may have refreshed the content or won a SERP feature. Treat the link gap as one candidate, and read the suspect table before putting budget behind it.
Our reading of the example
A gap in strong domains is worth acting on; a gap in count is not
In this invented case the only difference worth chasing is the 20 domains at DR 60 and up. Find those 20, find why they link, and decide whether you could earn something similar. The other 500 are noise until proven otherwise.
- Same parameters on every domain you compare
- Check the control rival who ranks below you
- Re-run next month with the diff recipe to see who is still gaining
- Write the conclusion with its date, so next month's pull can confirm or overturn it