A book can be admired by critics and resented by readers, or adored by readers and ignored by the press. The disagreement is systematic and follows from what each kind of review is for.
The two are answering different questions
A reader review answers whether the experience was satisfying: did the story deliver what it promised, was the ending earned, was it worth the hours.
A critical review answers where the book sits among other books, what it attempts, and whether the attempt succeeds on its own terms.
A novel can fail the first test and pass the second comfortably, and the reverse happens just as often.
Both are legitimate questions. They simply produce different verdicts, and neither reviewer is obliged to answer the other's.
Expectations are set differently
Readers usually arrive through genre. A romance reader expects specific structural promises to be kept, and a book that breaks them has failed regardless of its prose.
Critics frequently treat the breaking of convention as the interesting part, and reward exactly what a genre audience experiences as betrayal.
Neither position is wrong. They are measuring against different reference points.
The audiences are not the same size or shape
A critic writes for a general readership that mostly will not buy the book. The review is a piece of writing in its own right, judged as such.
A reader review is written for people deciding whether to spend an evening on it, and consequently deals in pace, character and payoff.
This is why reader reviews are often the better guide to whether you personally will enjoy something, and critical reviews the better guide to what the book is doing.
The formats reflect that split, with critics writing at length about intention and readers writing briefly about effect.
Selection effects run in opposite directions
Critics receive books chosen by publishers as worth attention, so their sample is skewed towards the ambitious and the well-supported.
Reader ratings are dominated by whoever bought the book, which skews towards the popular and towards existing fans of the author.
Each sample is therefore reliable about its own population and misleading if read as a verdict on readers in general.
Where the two agree
Agreement clusters at the extremes. Genuinely incompetent books and genuinely exceptional ones tend to be recognised by both, though for different stated reasons.
The disagreement concentrates in the middle, where a book is competent but unusual, and that is exactly the territory where reading both kinds of review is worth the effort.