Loren, thanks for sharing, first and foremost. I appreciate this thread (and the original) a lot.
I have a related idea and I want your opinion (and others fellow HNers')... Say, you sell the T-shirt for $16 each including shipping. The twist is that if there are over a fixed number of orders then everyone gets a discount, with the earlier buyers get a deeper discount. i.e. the more people buy, the more group discount, and the earlier you buy, the more discount. Do you think, based on this exercise and your experience, whether this will help make it more viral? Or would it be too complicated and hence destroy the conversation rate and virality?
I think we could rule out refunds and go with the Groupon model of delayed payment. And what's the risk in that? Collect the payment information (credit card or paypal) and hit them with the charge when the shindig is up and you know the final costs. I'm not seeing the risks.
I didn't know about the PayPal delay payment, so this is even better (more plausible).
I was actually thinking about giving a credit (charge $14 originally, final cost is $12, earn a $2 credit) which can be used for the next purchase. i.e. retention.
He doesn't have any server side code running. He used PayPal for payments. The scheme seems more plausible to me now, but it would definitely take more work.
I am curious whether it is effective or not, but pyramid/MLM is not something I'm worried about. Let's say the price schedule is well defined (and visible), and that people are paying money in exchange for goods, with various level of discount due to quantity and timeliness to encourage sales. In any case, from a good<->evil scale, I don't think what I suggested is even in the league of Zynga, Groupon, and Twitter promotion (tweet this for a chance...), let alone MLM scam. Whether it could be effective or not, that's my question.
To clarify this is what it would look like, using customer #10, #212 (if exists), and #1350 (if exists) as examples under different scenario
(1) #customers < 100
customer #10 pays $16
(2) 100 <= #customers < 1000
customer #10 pays $12, customer #212 pays $14
(3) 1000 < #customers
customer #10 pays $8, customer #212 pays $10, customer #1350 pays $12
The numbers are made up (the actual numbers may not make sense, so don't try to find flaw about total profit based on these numbers). So basically customers are rewarded for (1) being an early adopter, (2) propagating the viral effect, and (3) having good taste.
Comments
Loren, thanks for sharing, first and foremost. I appreciate this thread (and the original) a lot.
I have a related idea and I want your opinion (and others fellow HNers')... Say, you sell the T-shirt for $16 each including shipping. The twist is that if there are over a fixed number of orders then everyone gets a discount, with the earlier buyers get a deeper discount. i.e. the more people buy, the more group discount, and the earlier you buy, the more discount. Do you think, based on this exercise and your experience, whether this will help make it more viral? Or would it be too complicated and hence destroy the conversation rate and virality?
Flat pricing is predictable and transferred immediately. Your scheme would involve refunds or delayed payment. Sounds risky for questionable benefit.
I think we could rule out refunds and go with the Groupon model of delayed payment. And what's the risk in that? Collect the payment information (credit card or paypal) and hit them with the charge when the shindig is up and you know the final costs. I'm not seeing the risks.
I didn't know about the PayPal delay payment, so this is even better (more plausible).
I was actually thinking about giving a credit (charge $14 originally, final cost is $12, earn a $2 credit) which can be used for the next purchase. i.e. retention.
He doesn't have any server side code running. He used PayPal for payments. The scheme seems more plausible to me now, but it would definitely take more work.
That sounds dangerously close to a pyramid/MLM scheme ...
I am curious whether it is effective or not, but pyramid/MLM is not something I'm worried about. Let's say the price schedule is well defined (and visible), and that people are paying money in exchange for goods, with various level of discount due to quantity and timeliness to encourage sales. In any case, from a good<->evil scale, I don't think what I suggested is even in the league of Zynga, Groupon, and Twitter promotion (tweet this for a chance...), let alone MLM scam. Whether it could be effective or not, that's my question.
To clarify this is what it would look like, using customer #10, #212 (if exists), and #1350 (if exists) as examples under different scenario
(1) #customers < 100
customer #10 pays $16
(2) 100 <= #customers < 1000
customer #10 pays $12, customer #212 pays $14
(3) 1000 < #customers
customer #10 pays $8, customer #212 pays $10, customer #1350 pays $12
The numbers are made up (the actual numbers may not make sense, so don't try to find flaw about total profit based on these numbers). So basically customers are rewarded for (1) being an early adopter, (2) propagating the viral effect, and (3) having good taste.
Now that I think about it some more, this is similar to Amie Street for indie music (didn't quite strike it big) http://en.wikipedia.org/wiki/Amie_Street
What do you think?
Will it work? Who knows, but it sure sounds like it could. So here's my advice to you: try it and share the results with us.
I agree but was hoping to pick your brain. It is more of a thinking out loud.
Except everyone knows what they're paying, with $x discount.