I have received over twenty emails from Facebook, asking me if I know Tom Levesque, and to SEND him a friend request. Over and over, and over again. No, I don't know a Tom Levesque. I don't need to be asked the same question repeatedly.
A while back I logged into a Twitter account that had been dormant for a couple of years, since it appeared that some good friends had started using it more actively. As a result, I started getting email from Twitter once a week like clockwork. Clearly not a "our algorithms determined that something interesting happened" message, but a "our algorithm determined that you should get a message once a week, let's pick a random event from your social graph" one.
The main result is that I know better than to ever log in again and become a marginally engaged user worth spamming.
The problem for me is that those tactics don't seem to translate so well outside of the social-network environment. When you have customers who use your products, and there's no personal interaction between customers, then what can you use that's as effective to bring customers back to use your product?
Features is perhaps the only thing, but it's not even remotely as effective in my opinion. And features can be a double-edged sword as they are quite often not really important to the customer.
Don't call (email) me, I'll call you (visit your website when i need to).
Personally, I'd be much more likely to read and consider re-signing by way of less frequent messages which suggest how the product has been improved and why I might want to come back.
Incessant "FOMO" pressure just trains me to ignore that pattern.
For every other kind of service/app, I would send a reminder once in a while and/or showing the users what has changed on the service since their last visit.
Or much more crazy: ASK them why they are not coming back. If they are unhappy with your service. I think you might be suprised about how many valuable answers you will get.
I'm personally a fan of how well designed Twitter's re-engagement emails are.
Would you be able to point to this? Would love the reference.
