I think the hardest part is the one SkillNexus raised at the start, because a ranking change in the same month as a content edit tells you very little on its own. What has helped me is writing down the date each link actually went live, not the date it was ordered, and keeping that list next to the target page. Then I look in Search Console at the queries for that one page in the weeks before the link and the weeks after it, and I note whether new queries appear or the same queries move, rather than looking at the whole site. That matters because sitewide traffic can rise for other reasons and hide what one page did. I also agree with Abco_digitals and rudolp that changing one thing at a time makes the test cleaner, but in practice that is not always possible. When it is not, I compare the linked page with a similar page on the same site that had no new link and no edit in the same period. If the linked page moves and the similar page stays flat, that gives me more confidence than a ranking chart alone. And as amberthread said, the linking page needs to be indexed first. I check that before I count the link at all, because an unindexed placement cannot be credited with a later gain. If after two or three months there is no change for that page while the control page behaves the same way, I treat the link as not proven, even if the site as a whole improved.
