Mutual intelligibility is very, very hard to quantify.
There is an exceedingly crude measure, Lexicostatistics, that gets used in underdocumented languages, and that noone would dare used among familiar European languages. For what it’s worth (and it’s not that much), if two lects (= dialect or language, being agnostic about it) diverge in 20 out of the 100 words in the Swadesh 100 list of core vocabulary, they are considered different languages. It’s what you get for Ukrainian vs Russian.
Either Swadesh or myself (I honestly don’t remember!) ran the Swadesh list for Cretan and Cypriot against Standard Greek once. The result was 81% similarity for both. I did do Tsakonian vs Standard Greek, and came up with 70%.
Again: that number isn’t worth much. Cretan may have been subject to more assimilatory pressure than Cypriot, but I do think the combination of more phonetic change and intonation make Cypriot harder to understand than Cretan. Then again, I identify as Cretan rather than Cypriot, so I would say that.
I know I have been genuine difficulty in understanding heavier forms of the dialect, such as that spoken by my grandmother or my cousin’s husband Fotis. Be aware that there is a diglossic continuum in Cypriot, with people speaking on a spectrum between Standard Greek with a Cypriot accent, and what the locals call horkatika.
Are there any Standard Greek speakers who don’t understand what horkatika means? Good. Cypriot fortitions [j] to [k] after /r, ð, p/. In Standard Greek, that’s horjatika: “villager-talk”.