3 ms·

When Freemium Fails

27 points by cwan 14y ago

anigbrowl 14y ago
Goddamit Clement :) One of the problems with the freemium model is that many people sometimes seem to pick a revenue target arbitrarily and then backfill their projections with the necessary conversion rate. Obviously, few firms or entrepreneurs want to expose their pricing models publicly in forums like HN, but startups need to be able to answer the question of what marginal utility they provide to their potential customers. The 'less than a cup of coffee!' model of app pricing, for example, assumes that since coffee is such a small thing users will be just as willing to throw down $3 for an app. But the coffee has physical, psychological and social value for the consumer that pays off on a predictable time horizon - and pays off not just the $3, but the time spent standing in line or buying and brewing your own coffee. Consumer preferences in for this commodity are sufficiently complex to support a large and extremely competitive market. If it was 'just coffee' then it would cost the same everywhere and vendors would be limited to normal rather than economic profits. A better approach for vendors might be to assume your target consumer does not have endless disposable income. What value does your product or service provide that would make a coffee consumer go without their favorite beverage for a day? Or if not coffee, a competitor's offering or some other part of the consumer's status quo? How does your offering compare to the composite bundle?
Lasher 14y ago
Interesting article but a little misleading at the start. That first example (billing management software) touts 900 paid users @ $65 per month after moving away from the freemium model, but how many of those 900 users would have never become users at all if the software was not initially free? It's much easier to start charging a customer once they are locked in and familiar with the software. Maybe a "first year is on us then it's paid" model is a good compromise between freemium and paid software, with the option to upgrade sooner if you want support and/or some advanced features.
hatu 14y ago
Aren't most companies always under 50 employees? That sounds like a terrible idea. How often does a company grow massively, and when they do they often get their own dedicated accounting teams.
luckydude 14y ago
I can relate to this article though I was even less smart than any of the people interviewed. I choose to give away BitKeeper because I wanted to help Linux. There was absolutely no business model, freemium or otherwise. We eventually stumbled upon the privacy costs money, out in the open is free model but it was always sort of so-so. In retrospect, what I should have done is make 2 differentiated products, free and paid. Free is basic, no guis, just enough to get the job done for the kernel and paid has all the bells and whistles. We actually thought of that at the time and I rejected it because at the time, before Linus was using BitKeeper, I felt like if he didn't get the good stuff he wouldn't use it. Pretty stupid from a business point of view, kinda makes sense if you look at it from a help Linux not fragment point of view. Kinda.

27 points by cwan 14y ago

Goddamit Clement :) One of the problems with the freemium model is that many people sometimes seem to pick a revenue target arbitrarily and then backfill their projections with the necessary conversion rate. Obviously, few firms or entrepreneurs want to expose their pricing models publicly in forums like HN, but startups need to be able to answer the question of what marginal utility they provide to their potential customers. The 'less than a cup of coffee!' model of app pricing, for example, assumes that since coffee is such a small thing users will be just as willing to throw down $3 for an app. But the coffee has physical, psychological and social value for the consumer that pays off on a predictable time horizon - and pays off not just the $3, but the time spent standing in line or buying and brewing your own coffee. Consumer preferences in for this commodity are sufficiently complex to support a large and extremely competitive market. If it was 'just coffee' then it would cost the same everywhere and vendors would be limited to normal rather than economic profits. A better approach for vendors might be to assume your target consumer does not have endless disposable income. What value does your product or service provide that would make a coffee consumer go without their favorite beverage for a day? Or if not coffee, a competitor's offering or some other part of the consumer's status quo? How does your offering compare to the composite bundle?

Lasher 14y ago

Interesting article but a little misleading at the start. That first example (billing management software) touts 900 paid users @ $65 per month after moving away from the freemium model, but how many of those 900 users would have never become users at all if the software was not initially free? It's much easier to start charging a customer once they are locked in and familiar with the software. Maybe a "first year is on us then it's paid" model is a good compromise between freemium and paid software, with the option to upgrade sooner if you want support and/or some advanced features.

hatu 14y ago

Aren't most companies always under 50 employees? That sounds like a terrible idea. How often does a company grow massively, and when they do they often get their own dedicated accounting teams.

luckydude 14y ago

I can relate to this article though I was even less smart than any of the people interviewed. I choose to give away BitKeeper because I wanted to help Linux. There was absolutely no business model, freemium or otherwise. We eventually stumbled upon the privacy costs money, out in the open is free model but it was always sort of so-so. In retrospect, what I should have done is make 2 differentiated products, free and paid. Free is basic, no guis, just enough to get the job done for the kernel and paid has all the bells and whistles. We actually thought of that at the time and I rejected it because at the time, before Linus was using BitKeeper, I felt like if he didn't get the good stuff he wouldn't use it. Pretty stupid from a business point of view, kinda makes sense if you look at it from a help Linux not fragment point of view. Kinda.