Randomizing email send times can be very powerful for a number of reasons.
First, it applies to Einstein's Theory of Insanity; "Doing the same thing over and over again and expecting a different result". If you always email Sarah at 9:00AM on Tuesday and she happens to be a late night person, you'll likely get buried under the countless other emails she received that day / week. By randomizing your send times, you're sampling her to see when she'll engage. So, it's creating unbiased data on the individual recipient's habits.
Second, it can help with email deliverability because you're not flooding receiving servers.
Third, we've done extensive testing on this and as part of those tests, even randomizing everyone on a list over 24 hours will slightly outperform sending to everyone at once (blasting).
With Seventh Sense, anyone who has historically never engaged with you, the system will pick a randomized send time for these individuals within the provided delivery window.
If you're not scheduling the delivery over at least 24 hours and someone has never engaged within the delivery window regardless of time of day or day of week, the system will pick a randomized time for those individuals. By default, the system will also weight the randomized times based on the overall engagement of your entire audience which provides another advantage in getting these people to engage for their first time.
For example, if your scheduled email delivery window is from midnight on Tuesday to midnight on Wednesday and you have 100 people that have never engaged with you, the system will weight how many people are scheduled per hour based on the overall engagement patterns of your entire audience. So if 3% of your overall audience tends to engage at 6am, only 3 people who are to have their delivery time randomized would be scheduled during the 6am time window.