3 ms·
> the takeaway message for most people reading this is "Should we implement lifecycle emails or not?" rather than "What exact words should I use?" or "What freq
by memset 13y ago
> the takeaway message for most people reading this is "Should we implement lifecycle emails or not?" rather than "What exact words should I use?" or "What frequency is appropriate for a zero engagement user?"
I think (since we're both conjecturing about the takeaway message for the hypothetical "most people") the most interesting thing to come from this discussion would be how to do this sort of stuff tastefully.
It's hard to know how to engage! I think there is value in picking apart the techniques used in this campaign, the same way there is value in picking apart how CPU scheduling works.
After a bit of digging, it is easy to know the "things one should do for marketing." Engage, measure, test, iterate, funnel, A/B test, NPS, etc.
But what's not as widely known are what specific copywriting techniques one should use. Or how to measure the effectiveness versus email frequency. How do you decide when - or whether - or how frequently - to send a customer survey?