5 ms·

An interesting paper, but some of the other numbers reporting how these catalog sales work are also interesting. Apparently mailing a catalog to 60,000 people

An interesting paper, but some of the other numbers reporting how these catalog sales work are also interesting. Apparently mailing a catalog to 60,000 people will only generate single-digit purchases for each item, with only about 1000 sales overall. That's a shockingly inefficient result for all the advertising and junk mail we all put up with, but also not particularly surprising. It does mean that these kinds of businesses need far smaller stocks than I would guess off the top of my head of each item.

madaxe_again 2y ago
I used to build ecommerce presences, many of which were for old catalogue merchants making the jump to ecommerce. It was eye-opening. Even in the heyday of catalogue shopping, which this paper is from the tail end of in the early noughts, conversion rates were abysmal - and they were entirely acceptant of that, and didn’t believe it could be improved upon. Don’t even get me started on print and distribution costs. You’re looking at double digit CPAs on sales. Margins had to make up for it, which usually meant cutting everything else to the bone and really, really pushing finance options. An interesting part of this was that when they went online they often really, really, really wanted to replicate the catalogue experience on their websites - page flipping, etc. - took a lot of dissuasion, and in one case where they were particularly obstinate, we went ahead with their desires, with the proviso that they would allow us to build a competing site on more traditional ecommerce lines at no additional cost, with the proviso that we got a cut of sales. They dropped their catalogues entirely a year into the experiment. Catalogues suck.
- alexchamberlain 2y ago
  > conversion rates were abysmal Is that surprising? Email campaigns have similarly low open and click rates, which are also accepted as the norm.
  - madaxe_again 2y ago
    Yeah, but on email it costs you $0.002 to hit a target - catalogues, you’re looking more like $3-$15 depending on scale, catalogue size and quality. Those things cost money.
hermitcrab 2y ago
Typically only about 1% of people who visit websites for commercial software buy a licence: https://successfulsoftware.net/2009/04/23/the-truth-about-conversion-ratios-for-software/ https://successfulsoftware.net/2009/04/23/the-truth-about-co...
- codetrotter 2y ago
  I sometimes have a look at something but don’t pull the trigger yet, then come back multiple times to look at it again before I buy. I wonder if that kind of behavior could mean that the real number is a bit higher. For example say 2% of visitors are buying a license but it is counted as 1% because people visiting without buying and then coming back later and buying are counted as different “unique visitors”.
  - hermitcrab 2y ago
    Hard to say as this sort of data is very 'dirty'. Basic analytics will generally take account of returning visitos, but someone might be counted as multiple visitors because they looked at a website on multiple devices.
MrSkelter 2y ago
I used to work for one of the largest tech catalog companies in the US. The largest? Won’t say. Anyway we made about $30 per catalog and we sent 3 a month to millions of people. Our prices ended in 9s but I am not sure there was research behind it, just convention.

madaxe_again 2y ago

I used to build ecommerce presences, many of which were for old catalogue merchants making the jump to ecommerce. It was eye-opening. Even in the heyday of catalogue shopping, which this paper is from the tail end of in the early noughts, conversion rates were abysmal - and they were entirely acceptant of that, and didn’t believe it could be improved upon. Don’t even get me started on print and distribution costs. You’re looking at double digit CPAs on sales. Margins had to make up for it, which usually meant cutting everything else to the bone and really, really pushing finance options. An interesting part of this was that when they went online they often really, really, really wanted to replicate the catalogue experience on their websites - page flipping, etc. - took a lot of dissuasion, and in one case where they were particularly obstinate, we went ahead with their desires, with the proviso that they would allow us to build a competing site on more traditional ecommerce lines at no additional cost, with the proviso that we got a cut of sales. They dropped their catalogues entirely a year into the experiment. Catalogues suck.

alexchamberlain 2y ago
> conversion rates were abysmal Is that surprising? Email campaigns have similarly low open and click rates, which are also accepted as the norm.
- madaxe_again 2y ago
  Yeah, but on email it costs you $0.002 to hit a target - catalogues, you’re looking more like $3-$15 depending on scale, catalogue size and quality. Those things cost money.

hermitcrab 2y ago

Typically only about 1% of people who visit websites for commercial software buy a licence: https://successfulsoftware.net/2009/04/23/the-truth-about-conversion-ratios-for-software/ https://successfulsoftware.net/2009/04/23/the-truth-about-co...

codetrotter 2y ago
I sometimes have a look at something but don’t pull the trigger yet, then come back multiple times to look at it again before I buy. I wonder if that kind of behavior could mean that the real number is a bit higher. For example say 2% of visitors are buying a license but it is counted as 1% because people visiting without buying and then coming back later and buying are counted as different “unique visitors”.
- hermitcrab 2y ago
  Hard to say as this sort of data is very 'dirty'. Basic analytics will generally take account of returning visitos, but someone might be counted as multiple visitors because they looked at a website on multiple devices.

MrSkelter 2y ago

I used to work for one of the largest tech catalog companies in the US. The largest? Won’t say. Anyway we made about $30 per catalog and we sent 3 a month to millions of people. Our prices ended in 9s but I am not sure there was research behind it, just convention.