Book recommendations are an underrated social skill. Done well, they deepen friendships, surface unexpected ideas, and create a small shared culture between people. Done badly, they are an awkward social transaction in which one person tries to convince another person to spend ten hours on a book the second person does not want to read.
I have done book recommendations both ways. I have also received them both ways. After a decade of trading recommendations with a small group of friends who read seriously, I have formed some opinions about how to recommend books without being annoying, and how to receive recommendations without feeling pressured.
This is what I have learned.
The problem with most book recommendations
The typical book recommendation fails for one of three reasons.
Wrong book for the person. The recommender recommends what they like, not what the recipient would like. The mismatch is invisible to the recommender, who has read the book and enjoyed it. The mismatch is obvious to the recipient, who can tell within ten pages that the book is not for them.
Wrong pitch for the book. The recommender describes the book in ways that misrepresent what it is. "You will love this" is rarely a useful description. "It's about a woman who…" is more useful. The pitch needs to be accurate to the book and relevant to the recipient.
Wrong timing for the pitch. The recommendation arrives at a moment when the recipient is not in a position to receive it. They are in the middle of another book. They are not in the mood for that genre. They are about to enter a stressful period and have no mental bandwidth. The recommendation that arrives at the wrong time is received as social pressure.
These three failure modes are not unique to book recommendations. They are the failure modes of any advice-giving. But books are the case I know best, because books are what I have spent the most time recommending and being recommended.
The first principle: know the person
The most useful thing I do before recommending a book is not think about the book. It is think about the person.
The questions I ask myself:
What did they last enjoy reading? Not what did they last read — what did they last enjoy. There is a difference. Most readers can name the last book they enjoyed, even if they have read several since.
What did they last put down unfinished? The book they stopped reading is information. It tells me what does not work for them, which is often more useful than what does.
What do they read for? People read for different reasons. Some read for plot. Some read for language. Some read for ideas. Some read to be told a story. The book that delivers what the person reads for is the right recommendation.
What is their current life context? A book recommendation that arrives during a period of intense work, grief, or transition is not well received. The book itself may be good. The timing is wrong.
How do they react to long books? Some people love immersive 600-page novels. Some people prefer 200-page novellas. The recommendation should match the recipient's tolerance.
The answer to these questions determines the book I recommend. The book itself is less important than whether it matches the answer.
The second principle: be honest about the book
Most book recommendations are oversold. The recommender describes a good book as a great book, a great book as a life-changing book. The hyperbole is reflexive. It is also counterproductive.
When I recommend a book, I try to describe it accurately. This means including the parts that might put the recipient off.
A real example. A friend asked me for a recommendation last year. I sent her "A Little Life" by Hanya Yanagihara. Before sending, I told her: "This book is about four friends in New York. It is beautifully written. It is also extremely long and extremely sad. The protagonist experiences trauma that is unrelenting. If you are not in a place to read something that heavy, do not read it. If you are, it is one of the best novels of the last twenty years."
She read it. She thanked me. She also said that if I had not warned her, she might have abandoned it in the middle, and she would have missed the ending. The honest description was the difference between a recommendation that was heard and one that was ignored.
The same applies to the books that are not great but are useful. "This is not a great novel. It is a useful book on X. You will probably get what you need from the first half." Honesty about the limitations of a book is what makes the recommendation credible.
The third principle: offer, do not insist
The hardest part of recommending books is the social pressure. The recommender has read a book they loved, and they want the recipient to read it. The recipient can feel the pressure. The pressure makes the recommendation harder to receive.
The fix is to offer, not insist. The form of the offer matters.
Wrong: "You have to read this book. It would change your life."
Right: "I read this book last month and thought of you. No pressure. If you want to borrow it, I have a copy."
The difference is the explicit permission to say no. The first form makes refusal awkward. The second makes it easy.
I have a rule about recommending books: I never recommend more than once. If the recipient does not pick up the book after I have offered it once, the recommendation is over. I do not follow up. I do not check in. I do not bring it up again. The cost of repeated recommending is the friendship, and the book is not worth it.
The fourth principle: read the books people recommend to you
The reciprocal of the recommendation is the reception. The way you receive recommendations is what makes people want to keep recommending to you.
The principle is simple: read the books people recommend to you. Not all of them — the recipient should still be honest about what they can and cannot engage with — but most of them. The people who take your recommendations seriously are the people whose recommendations you should also take seriously.
In my close reading group, we have an informal rule. If someone recommends a book and another person reads it, the reader owes the recommender a 200-word response, even if the response is "I did not like it." The response is the social contract. It says: I took your recommendation seriously. I read what you gave me. Here is what I thought.
This rule has deepened the friendships and improved the recommendations. When you know the recommendation will be read and responded to, you recommend more carefully. When you know the response will be received, you respond more honestly. The system makes both sides of the recommendation better.
The fifth principle: do not weaponize recommendations
The worst book recommendations are the ones that are not really about the book.
The "you should read this because it would improve you" recommendation is usually a moral judgment disguised as advice. The "you should read this because it would help you understand what I care about" recommendation is a bid for emotional connection disguised as a book recommendation. The "you should read this because everyone in our field reads it" recommendation is a status signal disguised as a recommendation.
All of these are valid things to say. None of them should be disguised as a book recommendation. The disguise is what makes them annoying.
When I recommend a book, I try to be clear about why I am recommending it. "I think you would enjoy this" is a different pitch from "I think this would help you with X" is a different pitch from "I want to share this with you." Each pitch is appropriate in different contexts. Mixing them is what makes the recommendation feel loaded.
The genres that travel well
Some genres travel better through recommendations than others. The genres that work best:
Literary fiction with universal themes. Books like "A Little Life," "The Remains of the Day," "Olive Kitteridge," "A Gentleman in Moscow." These have broad appeal and work for most readers.
Strong-voice nonfiction. Books like "Educated," "The Soul of an Octopus," "Crying in H Mart," "Between the World and Me." Memoirs with distinctive voices and specific subject matter.
Genre fiction done well. Mysteries, thrillers, science fiction, romance — when the book is exceptionally good, it can travel to readers who do not usually read the genre. The trick is finding the exceptional books.
Short story collections. Underrated in recommendations because they are less famous. The right short story collection for the right person is one of the best gifts you can give.
The genres that travel poorly:
Highly specialized nonfiction. The reader has to share the specialization. Most people do not share most specializations.
Avant-garde or experimental fiction. Works for the people it works for. Does not work for most people. Recommending it broadly fails more often than it succeeds.
Books that are "important" rather than good. The classics that are respected more than enjoyed. The books that people should read but rarely want to. The recommendation fails because the recipient senses the obligation.
What to do when a recommendation fails
Sometimes a recommendation does not work. The book was wrong. The recipient did not engage with it. The relationship absorbs this. The next recommendation can still work.
What to do:
Do not apologize excessively. A short "Sorry it was not for you" is enough. Excessive apology makes the recipient feel they have to comfort the recommender.
Do not defend the book. The defense usually makes the recipient feel they have to defend their non-enjoyment, which is exhausting.
Do not take it personally. The book did not work for this person at this time. It does not mean the recommender's judgment is bad. It does not mean the recipient is unsophisticated. It means the match was wrong.
Ask what did not work. This is the useful follow-up. The answer tells you what to recommend next time.
Move on. The friendship is more important than the book.
The honest summary
Book recommendations are a small social ritual that, done well, deepen friendships and surface good books. Done badly, they create the kind of awkward obligation that most readers have experienced and few enjoy.
The skill is not hard. It requires knowing the person, being honest about the book, offering rather than insisting, reading what people recommend to you, and not weaponizing the recommendation. The first two are about accuracy. The middle two are about social sensitivity. The last is about honesty.
Most recommendations fail because they confuse enthusiasm for accuracy. A recommender who loved a book has a hard time describing it as it is. The description drifts toward what the recommender wishes the book were, not what the book actually is. The recipient, who experiences the book as it is, is disappointed. The mismatch is the failure.
The fix is small. Describe the book honestly. Recommend to the right person. Offer, do not insist. Accept the failures as part of the practice. The next one will be better.
Most of life is this. The art is in the small adjustments.