3 ms·

I went down this road — the economics don’t work unless you can get your acquisition costs very very low. Good conversion rates and margins but we are talking s

by gfodor 7y ago

I went down this road — the economics don’t work unless you can get your acquisition costs very very low. Good conversion rates and margins but we are talking small numbers, and large capex per user to print cards, esp if you give some away for free (which is a great acquisition channel)

daliusd 7y ago
Have you actually tried creating postcard sending service? Could you share what you have tried to acquire new users?
- gfodor 7y ago
  Yup, I spent 6 months or so trying to find product-market fit for two different postcard apps, one focused on creating memes and the other focused on sending baby/kid photos to grandparents and family. The latter had much more potential, but the economics were tough due to user acquisition costs. I suspect you could make it work if you had a better marketing strategy that I did. I was trying to get traction through display ads by doing highly targeted social networking campaigns, which seemed like it should be fruitful, and was getting low CPIs ($1-$2) -- however even at that level, and with significant viral loops, growth hacking, and a recurring revenue subscription based model, I just couldn't get the numbers to work. Ultimately, users seemed to need to see the product before they spend money, so having a trial period of one or two cards seemed essential. Without it, conversions flatlined I suspect since you were asking people to pay for something they couldn't even see the quality of first. These one or two free cards greatly added to acquisition costs, and led to a long, complex acquisition funnel, and so you had to more than make up for sending a lot of free cards with your paying customers while the new leads converted. When you consider the free cards, you were now talking about $3-4 per lead, and with a generous conversion rate of 3% that is approx a $100 cost per customer to acquire them. Surprisingly, over the course of a year or so of a subscription this could actually work if churn was low. But it was very risky and cost intensive to subsidize the acquisition costs for me personally, so I stopped working on it after sending approx 9k postcards and getting approx 250 customers, who almost entirely churned out after a few months. I learned a lot from this exercise, one thing was anytime you hear anyone doing this idea, especially if they are VC funded, it's a good idea to bet against them surviving unless you see they have some kind of cheap, high volume, high quality marketing channel.
  - daliusd 7y ago
    Wow. Thank you for such long, detailed and insightful comment. I look at this now as marketing challenge as well and I am trying some alternatives ideas.
z3t4 7y ago
Instead of targeting end users you can target intermediates like tourist shops. Where the tourist picks a photo from his/hear camera/phone via bluetooth. Or a photo from a touch screen. So instead of buying a postcard, a stamp, writing a message, and then finding a post box, you just have to go to the service in the hotel lobby. Imagine if you are going to send five post cards, you would probably save one hour of work. And the postcards could be guaranteed to come in time due to magic (they are not actually sent from the same country. That part is faked)

by gfodor 7y ago

daliusd 7y ago

Have you actually tried creating postcard sending service? Could you share what you have tried to acquire new users?

gfodor 7y ago
Yup, I spent 6 months or so trying to find product-market fit for two different postcard apps, one focused on creating memes and the other focused on sending baby/kid photos to grandparents and family. The latter had much more potential, but the economics were tough due to user acquisition costs. I suspect you could make it work if you had a better marketing strategy that I did. I was trying to get traction through display ads by doing highly targeted social networking campaigns, which seemed like it should be fruitful, and was getting low CPIs ($1-$2) -- however even at that level, and with significant viral loops, growth hacking, and a recurring revenue subscription based model, I just couldn't get the numbers to work. Ultimately, users seemed to need to see the product before they spend money, so having a trial period of one or two cards seemed essential. Without it, conversions flatlined I suspect since you were asking people to pay for something they couldn't even see the quality of first. These one or two free cards greatly added to acquisition costs, and led to a long, complex acquisition funnel, and so you had to more than make up for sending a lot of free cards with your paying customers while the new leads converted. When you consider the free cards, you were now talking about $3-4 per lead, and with a generous conversion rate of 3% that is approx a $100 cost per customer to acquire them. Surprisingly, over the course of a year or so of a subscription this could actually work if churn was low. But it was very risky and cost intensive to subsidize the acquisition costs for me personally, so I stopped working on it after sending approx 9k postcards and getting approx 250 customers, who almost entirely churned out after a few months. I learned a lot from this exercise, one thing was anytime you hear anyone doing this idea, especially if they are VC funded, it's a good idea to bet against them surviving unless you see they have some kind of cheap, high volume, high quality marketing channel.
- daliusd 7y ago
  Wow. Thank you for such long, detailed and insightful comment. I look at this now as marketing challenge as well and I am trying some alternatives ideas.

z3t4 7y ago

Instead of targeting end users you can target intermediates like tourist shops. Where the tourist picks a photo from his/hear camera/phone via bluetooth. Or a photo from a touch screen. So instead of buying a postcard, a stamp, writing a message, and then finding a post box, you just have to go to the service in the hotel lobby. Imagine if you are going to send five post cards, you would probably save one hour of work. And the postcards could be guaranteed to come in time due to magic (they are not actually sent from the same country. That part is faked)