Rankings move for many reasons. If you add links, change a page, and a core update lands in the same fortnight, you will not know which one did what unless you set things up first. This is a short method for judging link work without telling yourself a story.
Set a baseline before anything goes live
Record where you stand today:
- Positions for a fixed list of keywords, tracked from one location and one device type.
- Clicks and impressions for the target page in Search Console.
- The number of linking domains your link tool reports for the target page.
- The date of the last meaningful edit to the page.
Save screenshots or exports. Memory is a poor record. If your site is new or gets very few visits, expect noisy data and be more cautious about conclusions.
Change one thing at a time
If you rewrite the page and add links in the same week, the result tells you nothing about either. Where you can, finish the page edits first, wait, then add links. If you are testing, pick two similar pages and add links to one only. It is a crude control, and it is better than none.
Check that the link exists and is indexed
A placement that is not live, not crawled or not indexed cannot be expected to do anything. After delivery, open the page, confirm the link and its anchor, and check that the host page is in the index. Re-check after a few weeks. Hosts get edited, and links get removed.
Use a long enough window
Daily movement is noise. Look at weekly averages across several weeks, and compare against your baseline and against competitors who did not change. A page that rises for three days and falls back has not shown you anything. Compare like with like: the same date ranges, the same tracking tool and the same location. Switching tools halfway through produces a jump that has nothing to do with links.
List the other explanations
Before crediting a link, write down what else changed: algorithm updates, seasonality, competitors gaining or losing links, your own content and technical work, changes in how the results page is laid out. If two or three explanations remain, be modest about which one you pick.
Watch for risk signals too
Good measurement includes the downside. Check these on a schedule:
- The Manual Actions and Security issues reports in Search Console.
- A sudden drop in impressions across many pages, not only one.
- A host page leaving the index, which you can spot with a site: search.
- Anchors changing or the link disappearing.
A drop shortly after new links is a signal to pause, not to add more. If it happens, our post on what to do if a link or host is devalued covers the next steps.
Decide in advance what counts as success
Write down, before you order, what you would consider a good result and a bad one, and over what period. People who set the target afterward almost always find that the result met it. A written target also makes it easier to stop spending when the plan is not working.
Keep the log
One spreadsheet is enough: date, host, link type, target URL, anchor, term, and notes on what else changed that week. Add what each placement cost, because a result only means something next to its cost. Review it monthly. Over a few months it will tell you more than any single ranking snapshot, including whether the whole approach suits your site and your tolerance for risk.
If you want a plan that includes reporting of live URLs, message us on Telegram or use the request form. Reviewing a request is free.
Search guidelines and tactics change, and no link or plan can be promised to produce a result. Treat everything here as general education, test carefully and decide with your own risk in mind.