Most Lead Magnets Are Liability Magnets
3,000 downloads, 200 calls, 4 deals: the guide that anyone will download builds a list of the wrong buyers. Add friction, gain intent.

At one B2B software company I saw run the numbers: 3,000 downloads of the buyer's guide, 200 phone conversations, 4 closed deals. Marketing celebrated the download volume. Sales complained, correctly, that the list was unworkable.
Most lead magnets attract the wrong buyers. A buyer's guide behind an email field pulls in everyone curious about the category: competitors doing intel, students writing papers, consultants stocking their own decks, prospects a year or two from buying. The list grows. The list does not convert.
Low-friction content attracts low-intent readers, and that is the whole story. An email-only form filters for nothing, because an email address is cheap to hand over. The person who downloads is the person who would have read the thing for free had you left it ungated. The gate added a contact to the list without adding intent to the contact.
So add friction on purpose. Ask for company size, role, and the specific problem the reader is trying to solve. Make the form six fields instead of one. At that same company, download volume fell by roughly four-fifths, and the survivors were far higher-intent: they disclosed enough about themselves to signal they were evaluating, not browsing.
Marketing resists this because the download count drops and the list looks smaller in the board update. The board update is calibrated to the wrong number. The number that matters is pipeline value per lead, and by that measure the smaller list beat the larger one. This is the same discipline as firing the customers who cost more than they pay: fewer, better-qualified inputs beat volume you then have to service.
The deeper move is to build assets only an active buyer would want. A generic buyer's guide is too easy to grab; anyone curious about the category will take it. A pricing calculator that requires real inputs about the buyer's own business is too specific to download idly, so only someone modeling the purchase fills it out. The asset filters by self-disclosure, which is the same logic behind defining the client you want before you go chasing everyone.
Run it backward and the trap is obvious. Optimize for the biggest list and you build a list of people who will never buy. Optimize for the buyer who will pay, and the list gets smaller, quieter, and worth more.