Too Good to Leave Too Bad to Stay — How to Finally Know for Sure
Too Good to Leave Too Bad to Stay — How to Finally Know for Sure
If you are looking up Mira Kirshenbaum's book right now, you are almost certainly sitting with one of the hardest questions in adult life: stay, or leave? The book's 36 diagnostic questions have helped readers think through that question for decades, and its search popularity in 2026 confirms it still resonates. But if you finished it and are still not sure, that is not a failure of your judgment. It is a function of what the book can and cannot do.
What the Book Actually Does
Kirshenbaum's approach is built on clinical observation across her work with couples. The 36 questions are designed to surface "diagnostic moments" — patterns in a relationship that, based on that research, predict whether a relationship is workable. Each question functions as a threshold: if you can answer it one way, the implication is to stay; if you cannot, the book points toward leaving.
This is genuinely useful. It gives vocabulary to feelings that are hard to articulate and structure to a decision that can otherwise feel like an endless loop.
Why Readers Still Feel Uncertain After Finishing It
The 36 questions are not scored. They are not weighted by your specific history, communication patterns, or compatibility dynamics. Two people in very different relationships might answer the same question the same way — one correctly for their situation, one not — because the question cannot account for individual context.
Kirshenbaum is clear that she is offering a framework, not a verdict. The book works through pattern recognition across clinical populations. Your relationship is one data point it was never designed to individually analyze.
This is why many readers finish the book with more clarity than they started — but still short of certainty.
What a Compatibility Score Adds
CupidsLogic's PRISM model does something the book cannot: it computes an exit-probability score based on the specific combination of compatibility factors present in your relationship — not generic clinical thresholds drawn from a different population.
Rather than asking whether you can answer a diagnostic question the right way, PRISM calculates how your actual dynamics interact across multiple compatibility dimensions that predict long-term fit. The output is a personalized exit-probability reading — a data point that no checklist or clinical framework can produce for your specific situation.
The PRISM Founders Report is available at $47 (regularly $297) and covers your full compatibility breakdown and exit-risk profile. The $19.99/month coaching access adds an AI coach to help you work through what the data means for your next step.
How to Use Both Together
The Kirshenbaum book and a PRISM assessment address different parts of the same question. The book helps you identify what you already sense and feel. PRISM surfaces what the compatibility data says about your specific dynamics — patterns you may not yet have language for.
If you read the book and still feel stuck, a scored assessment is the logical next move. A framework tells you how patterns across clinical populations tend to resolve. A score tells you where your particular relationship sits.
FAQ
What is Too Good to Leave Too Bad to Stay about?
Too Good to Leave Too Bad to Stay by Mira Kirshenbaum is a guide to the stay-or-leave relationship decision. It presents 36 diagnostic questions drawn from her clinical work with couples. Each question is designed to identify whether a key relationship quality is present or absent — and her argument is that certain patterns indicate a relationship cannot work regardless of love or effort. #### Is Too Good to Leave Too Bad to Stay still worth reading? Yes — for what it does, which is structured self-reflection and pattern identification drawn from clinical experience. Its limitation is that it was written before compatibility modeling and relationship data science existed in their current form. Readers who want a thinking framework will find real value in the book. Readers who want a personalized score will need a tool the book was never designed to provide. #### What is the difference between Kirshenbaum's diagnostic questions and a relationship assessment like PRISM? Kirshenbaum's questions are threshold-based: you answer according to a clinical pattern observed across many relationships. PRISM is score-based: it calculates how your relationship's specific compatibility dimensions interact to produce an exit-probability estimate. The book identifies whether a known danger signal is present. PRISM quantifies the overall risk profile of your unique dynamic. #### I finished the book and I'm still not sure what to do. What is the next step? Persistent uncertainty after reading the book usually means the diagnostic questions landed in gray-area territory — which is exactly where a scored assessment adds the most. PRISM's exit-probability output is specific to your relationship, not drawn from clinical population averages. It does not replace your judgment, but it gives that judgment a concrete data point to work from.