Ask HN: Is it crazy to offer a freemium package in a "lo-fi" startup?
I've been testing out a "pay as you go" plan on SnailPad (www.snailpad.com/plans) where you get 2 free stamps each month (you get charged for stamps you use above that amount).
It's only been a few weeks but it's been working pretty well -- the majority of the initial users on that pay as you go plan have already either abandoned the service or upgraded to a monthly plan.
I've been tracking my path on a simple excel spreadsheet just to make sure I'm earning more than I give away (so far, so good). Is there anything else I should be watching out for (such as potential abuse or anything else)?
(
15 comments
| 19 points
| on HN
)
There is nothing wrong with giving people free things that actually cost money, as long as your COCA (cost of customer acquisition) is substantially lower than your LTV (lifetime customer value). It is no different, conceptually speaking, from doing something like paying from AdWords for a user who might convert into the trial but never into a paying user. I'll probably pay Google close to $1,000 this month for that: this is safe because I have historical data and a failsafe (discussed later).
It has also been my experience that most users go up-or-out very quickly on my service. If that continues to be true for you, you have very little to worry about.
If you speak to an accountant, who are generally conservative types, they will recommend you to do something to cap your maximum liability via this. For example, you can clarify that unused free stamps expire at the end of the month. (That might already be the policy. If so, smart.) You could also reserve the right to yank that offer at any time. I'm not a lawyer, but I think that is likely to be as simple as merely yanking the offer at any time.
I'd put a failsafe in the system to prevent you from spending over $X on free stamps in a day, for reasons similar to why Google only has authorization to charge me $X in a day: if the process goes suddenly out of control, have a human think things through before saying "Oh, sure, 1500% growth is totally normal under these circumstances. Spending approved!" It might happen because you ended up featured on the front page of the NYT. Congratulations. It also might happen because you ended up on 4Chan when some enterprising individual suggests using your service to send a million pictures of genitalia to some public figure who has made statements they do not approve of.
Now that I'm starting to get confident about things, it's just a matter of blocking off a few hours on Saturday morning and turning it into code.
And yes, all unused stamps expire at the end of each month -- it's in the TOS. :)
I love the idea of the failsafe... this is definitely going in.
Anyhow, the actual implementation is just a web page (or similar) which has stats, graphs, etc for whatever metrics you want to display, designed to be quickly comprehensible.
If you give me a few hours to get back to my apartment (and Paint.NET) I'll screenshot mine, block out the private customer data, and show you.
I'm not the one you replied to, but: Will be interesting, thanks in advance!
See here:
http://news.ycombinator.com/item?id=1112500
On an unrelated note: why isn't there a nice big picture of your gorgeous-looking paper and envelopes on the site?
As for the picture of the paper/envelope, (this is probably going to sound dumb) but I can't figure out how to take a decent shot without making the stuff look terrible. Any tips or thoughts on how to get past that?
I'd love to get an idea of where you think that picture ought to be and any other feedback you have.
I equally like to a point how your tracking it via excel, instead of building a system for hours. Once proven it might aid growth of revenue, then i see the system being built.
Even if the above is not correct for you, it has given me food for thought and helped solidify a few liquid thoughts.
I figured that you need to be able to send yourself at least one sample -- it takes one stamp if you live in the US and two stamps if you're anywhere else on the planet.
> If someone uses your free service for three months, or six, or twelve, what are the chances that they will upgrade to a paid service?
Ask me again in 3, 6 and 12 months. :)
> Why would this be more profitable than a plain pay-as-you-go plan with no free stamps, or one or two initial trial stamps?
I originally had no free stamps included in the pay as you go plan. This idea of including a few free stamps started up just a few weeks ago so I don't have any long term data. The theory here is that I "hook" you buy giving you an easy way to send your first letters... then you'll (hopefully) see the value in upgrading.
> If I choose to sign up with one of your monthly plans, will you offer me a refund of my unused credit if I decide not to use it?
Technically, no (it's in the TOS, etc). (Though, I'd probably refund the money if someone asked for it. I'd rather keep some good karma going...) When you pay your phone company (or 37signals, or anyone else) for a full month and then decide to quit, I doubt any of them will easily give you a refund. :)
There is a trick to make people pay, the same that Microsoft and other software companies are using. Give your service for free and start charging when your customers can't afford to stop using it, then they afford to pay!
Evil, but works.