Built for on-call SEOs
Monitor backlinks and get paged when they vanish
Rankift is a backlink API for in-house SEOs who would rather get a Slack message than open a dashboard. Snapshot a domain, diff it on a schedule, and read only what changed.
- Snapshot referring domains with one call
- Diff against last week in twenty lines
- Free
domain,overviewandstatsactions
$ curl "https://rankift.com/api.php?action=overview&q=nytimes.com&key=sk_live_..."
- domains in our own link index
- 181,132,097
- domain-to-domain links
- 2,774,625,913
- backlinks indexed, refreshed about monthly
- ~798.5 billion
- free credit on sign-up, no card
- $100
What it is for
How to monitor backlinks when nobody is watching a screen
Most link tools are built for people who look at charts. This one is built for the person who gets paged. Four jobs cover almost every alert we have heard about.
A lost link is an incident
A referring domain that was there last snapshot and is gone now needs an owner and a timer. The lost links runbook shows who checks what, in which order.
A new rival link is a signal
When a competitor lands a link from a site you also want, you should hear about it within days. See competitor backlink alerts for the rules and the severity table.
A traffic drop needs a post-mortem
Core updates arrive without warning, and the first question is always the same: did our links move first? Work through the core update post-mortem.
A weekly diff is the whole product
Pull, store, compare, post. The weekly backlink diff recipe is a single Python file and a cron line.
The smallest monitor: one call, one file
Run this from cron on Monday. Save the output next to last week's file and compare them. That is the entire idea.
curl -s "https://rankift.com/api.php?action=refdomains&q=example.com&per=1000&key=$RANKIFT_KEY" \
> snapshots/example.com-$(date +%F).json
# 2 credits per call = $0.0002, whatever the row count (max 1,000 rows)How it works
From sign-up to the first page-out
- 1
Sign in with Google
No card is asked for. The account starts with $100 of credit, which is 1,000,000 credits.
- 2
Make a key and test it
The
overviewaction is free, so a bad key or a typo costs nothing. A wrong key returns HTTP 401. - 3
Snapshot your own domain first
Pull referring domains for the site you are paid to protect and store the JSON with a date in the filename.
- 4
Add rivals and a schedule
List three to five competitors, run the same pull weekly, and post only the differences to Slack or email.
Compare
What the monitor can and cannot see
| Question | Rankift API | Where to look instead |
|---|---|---|
| Did a referring domain disappear? | Yes, between monthly index refreshes | Your own crawl for same-day proof |
| Did a rival gain a link this week? | Only if it reached our index | Search Console links report for your own site |
| How fast do changes show? | About once a month | Ahrefs or Semrush for fresher crawls |
| Keywords and rank tracking? | No | Ahrefs, Semrush, a rank tracker |
| Price of one pull | 2 credits, $0.0002 | Ahrefs API units start at $129/month |
Who it is for
Three questions the diff answers, and one it cannot
A monitor earns its place when it answers a question you would otherwise ask by hand. Most of the questions on-call SEOs bring to us are narrow, and a weekly diff handles them well.
Have they lost links recently? Compare two referring-domain snapshots and read the rows that left. If none left, that is a real answer for this index, not a shrug.
Does Search Console show a penalty? It shows manual actions in its own report, and that is the place to check first. The diff cannot see a penalty, only the links that moved.
Have they lost links recently?
Pull the donor list twice, a few weeks apart, and read only the rows that disappeared. Confirm each one on the live page before you call it lost, because a snapshot gap and a removed link look the same in a diff.
Is the junk growing, or is it normal churn?
Count the new rows per week over a few months. A steady trickle of low-quality donors is normal for a popular site. A sudden batch from one network is the pattern to read first, and the spam links triage page shows how.
Who it is for
Where a link diff stops being enough
A diff answers which domains moved. It does not answer why rankings moved, and it does not show what a rival changed on its own site. Treat it as one input to a review, not as the review.
Some questions need another tool. Keyword rankings, rank history and AI answers are outside this product, and so is a same-day crawl.
Be clear about the monthly refresh with anyone who reads the alerts. A link built last week may not appear until the next refresh, and that delay belongs in the report, not in a footnote.
Should I disavow the junk the diff shows?
Rarely. Read the donors first, then decide whether a removal request or a disavow is the right step. Most junk is ignored by search engines, and disavowing it changes little unless a manual action or a clear pattern points you there.
What if the drop came right after a core update?
Start with the timeline, not the diff. A diff that shows no lost links rules one suspect out, and that is useful. The core update post-mortem lists the other suspects in the order to check them.
FAQ
Questions on-call SEOs ask before wiring this up
How do I monitor backlinks without paying $129 a month?
Use prepaid credit instead of a subscription. One referring-domain pull costs 2 credits, so $0.0002, and returns up to 1,000 rows. Store a snapshot weekly, compare it with the last one, and a hundred domains cost about a cent a week. New accounts get $100 free.
How fresh is the link data?
The index refreshes roughly once a month, so a link built last week may not appear yet. A weekly job still works, because it runs the diff and stays quiet until the index moves. For same-day proof, check the live page yourself.
Can it alert me when a competitor gets a new link?
It cannot send alerts itself. You schedule the pull and post the difference wherever your team reads. The rules for deciding what is worth a message are on the competitor alerts page, with a severity table.
Do I need to know how to code?
A little. The recipe is about twenty lines of Python and one cron line. If you cannot run cron, the free explorer and the domain action still answer one-off questions, and a developer friend can schedule the rest in an hour.
Will this tell me why my traffic fell?
No tool can promise that. It tells you whether your link profile changed before the drop, which rules one cause in or out. The core update post-mortem lists the other checks to run.
Should I disavow spammy links it finds?
Rarely. Google has said it ignores most junk links, and the tool exists mainly for manual actions. Read the spam links triage page before you upload anything to Search Console.
What happens when my credit runs out?
The API returns HTTP 402 and your job fails loudly instead of silently. Top up with USDT (TRC-20) from the account page. Every authenticated response includes the remaining balance, so a script can warn you early.
