I did that with Sifter. Depending on the amount of issues they had created I sent a modified message. For instance, with zero issues, I offered help or more time to try it, with only a couple of issues, I’d ask if there was something I could help with. You could really go pretty far with this and based it on when they last logged in, how much activity they had, etc.
I also had an automated plain text email from me personally that would auto-adjust the message based on their level of activity. It was automated, but it was sincere. Anytime someone replied to these, I would usually reply in about 10 minutes or less. In my experience, that’s the key with these kind of messages. If the email purports to be from the founder but then a customer service rep replies, it seems incredibly disingenuous.
I sold Sifter earlier this year, but I’d say that it definitely helped to tailor the messages to their level of engagement. Unfortunately, I don’t have any specific numbers because it was something I did several years ago, but I can objectively say that it helped on several levels. I received more questions and had more requests for trial extensions.
Based on my experience, I can anecdotally say that the more narrowly the messages are focused, the better the response rate. For example, with zero engagement, keep it simple. “Was something wrong? Do you need more time to try it out?” or if really active, “It looks like you’ve got the hang of it, so I just wanted to say hi and let you know I’m here if you ever need anything.”
Tangentially related, I just created this best practices guide for trial expiration emails. Not sure if it can help inspire some ideas, but it’s essentially a collection of ideas and lessons I learned over the years running Sifter.