People are often demonstrably incapable of deciding whether results are poor, misleading, or wrong. For example, they routinely delegate that decisionmaking to

People are often demonstrably incapable of deciding whether results are poor, misleading, or wrong. For example, they routinely delegate that decisionmaking to trusted advisors who they have picked in a not-strictly-speaking-rational manner (e.g. most middle class people abominably overpay for investment advice from people who give them screamingly bad advice. This is the nearly universal finding of decades of research. Evidence for "If you're paying 2% to a broker to buy a mutual fund with 2% fees you are making a bad decision" is substantially more certain than "Cigarettes: not that great for you!" and yet people make that decision every day. They will make it, in controlled laboratory settings, for things as frivolous as $250 of airline miles.) A specific example of this with regards to information search problems: not only are people incapable of telling whether a SERP is a good SERP or not, you can create objectively bad SERPS and then sell them as good by, e.g., mimicking either a) the design, b) the branding, or c) the design and branding of, e.g., a Google SERP. There exist studies of what happens when you e.g. dress up Yahoo results in Google CSS (people will prefer them over Google results in Yahoo CSS or, for that matter, Google results in Google CSS). With specific relevance to the giving of money, I have access to anecdotes from the SEO world that occasionally curl my toes. A fairly common trick is a review site with the review stars awarded by "sort by affiliate commission, descending", which obviously bears no relationship to quality results. It doesn't just work, it works astoundingly well. And not only does it work astoundingly well, it works astoundingly well on people who one would presume are fairly savvy about quite-involved purchases where the dollar amounts involved will, frankly, swamp the total lifetime value of all paid-for HN searches. (e.g. Obscuring particulars a little bit, would you expect that anyone would make a decision for a seven-figure software deployment at a $FINANCIAL_INSTITUTION on the basis of a starred review site with a name spiritually similar to www.best-bank-software.biz? cough Demonstrable evidence of success.)