Ask HN: Dos & Don'ts of Referral/Invite Programs
I've seen a lot of great referral and invite programs, they're definitely a great way to bring in new users and reward existing customers.
There are also a ton of really poorly executed programs that result in spam and angry/displeased customers.
What are the Dos & Dont's of Referral/Invite Programs? I'm asking specifically about non-free products. Gmail had a great invite system but it was totally free.
Two good examples I can think of off the top of my head are Linode and Dropbox.
# Linode "You refer someone and they use your referral code during sign up. They keep a Linode for 90 days. You then receive a $20.00 credit applied to your account."
# Dropbox "Invite your friends to Dropbox! For every friend who joins Dropbox, we'll give you both 250 MB of bonus space (up to a limit of 8 GB)!"
19 comments
[ 4.4 ms ] story [ 59.6 ms ] threadAlso, http://www.forrst.com has an invitation only approach going on and I think it works really great for a community like that which self-selects.
It really depends on the product you're working with. If the general excitement around it is high, you can do the invitation/referral to get more people interested like Dropbox and Gmail did. The Closed Beta idea also works because it gives some time to work out the kinks and have time to scale up (in the case with Greplin.com).
What do you think about the possible downsides of a two sided incentive? For example new users _always_ signing up with a referrer.
In Dropbox's case the reward is 250MB capped at 8GB, but if the same thing applied to the linode model of dollars in credit, it could spell trouble.
In a double sided referral (especially when giving money) you are splitting the discount between two people. This can mean that the new customer gets less of a discount.
If you have $10 to acquire each new customer, you have one option where you say "$10 off for new customers!" and hope people see it, or you split that amount "$5 off for both of you if your friends sign up".
The benefit of the latter is that you are essentially offering a deal to get high quality referalls. However, you might be offering less to the end customer.
The genius of giving away product like dropbox has (or giving store credit) is that the cost to you is alot lower so you can give away more for double ended referrals. Then, everyone wins. The deal is better, both parties get the deal, and the company is happy.
We think its a really interesting problem area that we're trying to solve.
Dropbox's referral program is on the spammy side of things though, I've seen people complained about it on Twitter.
I think the key in referral program is to reward your loyal customers who went out to sell your product for you. Make them part of your business, give them the power to give out freebies, but put a limit on it to create prestige... (a mix of Dropbox & Google)
The other thing is viral pressure... and nobody does it better (or worst depending the way you spin it) than Zynga. There's a public Google Doc named "What I learned from Farmville so you don't have to play it", really good read on its viral pressure mechanism! Not sure if it applies to your business but will probably get you thinking nevertheless.
I looked up the Google Doc, here's a link for anyone else who's interested: http://meidell.dk/archives/2010/11/25/what-i-learned-from-fa...
I'll give it a gander and see what I can find.
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
http://✂.ws/yX
We have a very small user base at the moment (about 100 regulars) and about 10 are really into the product and give us great feedback.
Of those 10 there is 1 that has really given the referral program a go with gusto.
He crafted up an email, sent it to just about everyone he knows - seriously like 100 emails, with a very in depth explanation of why he liked the service.
Zero response.
Why? Because the format of the email is so spammy. The email subject still says "You have been invited to ... " etc.
Our goal now is to try changing the program so that instead of sending out referral emails from our interface, each person gets a unique referral signup URL that they can send to their friends in whatever format they like.
http://www.quora.com/What-are-the-dos-donts-of-referral-invi...
Edit to elaborate: Since you specifically mentioned referral models that give credits, read what I just wrote about Linode's referral model above (http://news.ycombinator.com/item?id=2048246)
For example, you can do something like: "Get $10 back if your friend makes a purchase using this $10 coupon."
That way you aren't paying out until it results in a new customer.--much less risky than other forms of marketing where you are paying before seeing the results.
Dropbox and Groupon align everyone's incentives by offering benefits to both parties and acquiring customers at a much lower cost than traditional ads. They provide authenticity because the folks referring their friends are genuinely using their products.
Many consumers now frown upon affiliate marketing because they feel taken advantage of--the affiliates say whatever they think you want to hear in order to get you to buy, and some make an awful lot of money doing it.
Plaxo was perceived very negatively because they were inviting their user's contacts without their permission.
The book "The Ultimate Question" contends that asking "How likely is it that you would recommend this company to a friend or colleague?" is the single most important metric you should measure in your business. Thus, if you find that you have to offer too large of an incentive to your customers, you should instead focus on improving your product.
We're building a social referral platform right now with these factors in mind that will make it super easy to run different referral programs and find what's best for your business. We're focusing on eCommerce now, but are getting into SaaS and other services very soon.
Send us a note at hello _at_ curebit.com if you're interested in piloting with us.