The three systems
Clarivate’s Journal Citation Reports ranks journals by Journal Impact Factor within Web of Science categories, and publishes annually in June. SCImago ranks by SJR, a prestige-weighted metric over a three-year window in which a citation from a highly-ranked journal counts for more, within Scopus categories. Scopus itself publishes CiteScore quartiles, computed over a four-year window with no prestige weighting.
None of the three is more "correct" than the others. They answer slightly different questions, and an institution choosing one over another is making a policy decision, not a technical one.
Why the same journal lands differently
Three structural reasons, all independent. First, the metrics differ: a two-year unweighted impact factor, a three-year prestige-weighted SJR and a four-year unweighted CiteScore will not rank a set of journals identically. Second, the category systems differ — Web of Science uses 254 categories, Scopus over 300 — so a journal is being compared against a different set of peers in each. Third, the underlying populations differ, because a journal can be indexed in one and not the other.
The practical upshot is that "Q1" without a system, a category and a year attached is not a checkable claim.
What changed recently
Web of Science coverage widened significantly, and a lot of guidance online has not caught up. From the 2023 Journal Citation Reports, all Core Collection journals including AHCI and ESCI became eligible for an impact factor, adding more than 9,000 titles. From the 2024 release, ESCI journals entered the category rankings and quartiles for the first time.
That second change moved quartile boundaries. Because ESCI journals typically rank lower, some journals shifted up a quartile without their own impact factor changing at all. If you are comparing a journal’s quartile against a figure quoted a few years ago, this alone can explain the difference.
A journal’s quartile can change without its metric changing, purely because the category population moved around it.