3 ms·

I doubt many people would pay someone $50/hour to optimize their next $9.99 kindle purchase. The one-on-one approach does not scale. We already have the many-

by finnh 9y ago

I doubt many people would pay someone $50/hour to optimize their next $9.99 kindle purchase. The one-on-one approach does not scale. We already have the many-to-many form of this (goodreads, for example).

IanCal 9y ago
The hourly cost sounds high, but as an honest question, how long would it take? You likely didn't speak to the proprietor of the book store for an hour every time you visited to get a single book recommendation. Would you pay $5 for 5 minutes and a set of recommendations? It's worth noting that what we had previously was effectively a store manager working part time as what I'm suggesting. Your hourly cost there is also (if full time) equivalent to well into the top decile of earners in the UK. > The one-on-one approach does not scale. Does it need to scale? The previous setup was already a one on one approach. The advantage of it now is that location is not important and I could talk immediately to a recommender pretty much anywhere rather than only the one at my local book store.
- finnh 9y ago
  > Would you pay $5 for 5 minutes and a set of recommendations? You'd probably need more than 5 minutes for the first set of recommendations, right? If they can give you 5 recs in the first 5 minutes of talking to you, then those recommendations aren't necessarily so personalized (or, at least, the nominal lift above crowd-sourced recs is dubious). > what we had previously was effectively a store manager working part time as what I'm suggesting. Good point. Let's call it $20/hour, then? Starts to sound a little better. (but, of course, without a physical footprint & actual books to sell, you need to spend more on marketing, so you need to charge more). > The advantage of it now is that location is not important and I could talk immediately to a recommender... I get the appeal. If I was provided reliable, way-better-than-crowdsourced recommendations, then that would be worth a lot to me - easily enough to justify a one-time fee (say, $30) and then $1 per book or $5 per series. I doubt that's possible, though =( Even really close friends with really similar tastes to mine, with whom I have regular long conversations about books, still have occasional misses when the recommend books to me, and vice-versa.

by finnh 9y ago

IanCal 9y ago

The hourly cost sounds high, but as an honest question, how long would it take? You likely didn't speak to the proprietor of the book store for an hour every time you visited to get a single book recommendation. Would you pay $5 for 5 minutes and a set of recommendations? It's worth noting that what we had previously was effectively a store manager working part time as what I'm suggesting. Your hourly cost there is also (if full time) equivalent to well into the top decile of earners in the UK. > The one-on-one approach does not scale. Does it need to scale? The previous setup was already a one on one approach. The advantage of it now is that location is not important and I could talk immediately to a recommender pretty much anywhere rather than only the one at my local book store.

finnh 9y ago
> Would you pay $5 for 5 minutes and a set of recommendations? You'd probably need more than 5 minutes for the first set of recommendations, right? If they can give you 5 recs in the first 5 minutes of talking to you, then those recommendations aren't necessarily so personalized (or, at least, the nominal lift above crowd-sourced recs is dubious). > what we had previously was effectively a store manager working part time as what I'm suggesting. Good point. Let's call it $20/hour, then? Starts to sound a little better. (but, of course, without a physical footprint & actual books to sell, you need to spend more on marketing, so you need to charge more). > The advantage of it now is that location is not important and I could talk immediately to a recommender... I get the appeal. If I was provided reliable, way-better-than-crowdsourced recommendations, then that would be worth a lot to me - easily enough to justify a one-time fee (say, $30) and then $1 per book or $5 per series. I doubt that's possible, though =( Even really close friends with really similar tastes to mine, with whom I have regular long conversations about books, still have occasional misses when the recommend books to me, and vice-versa.