Shopkeepers don't pick up at the start of the day's rush. Salaried people may answer at lunch and after seven. A list rung from the top each morning treats them all the same, and then your report calls the ones who did not answer uninterested.
Best time to call tries to correct this in a modest way. It works out when each contact tends to answer, and rings them first when that window opens. It changes the order, and it never changes who gets called.
How it learns
A completed call, in either direction, counts towards its India-time hour. Weekdays and weekends are kept apart. Once a night, it finds the two-hour window with the most answers for each contact.
Until someone has answered a couple of times, there is not much to go on. So it falls back: first on contacts from the same source, then on your whole workspace, ranked by answer rate. On a contact, this shows as a line saying when they usually answer, and if it is borrowed from contacts like them, it is marked as such.
What happens in a campaign
Switch on Call people when they usually answer, on a schedule. Schedules made before this feature existed have it off, so check your old ones. Then, each time the dialer picks its next batch, people whose window is open come first. Everyone else follows in the same pick. Nobody is skipped or held back.
Here is a small example. Your campaign calls from 10 am to 7 pm. A customer who usually answers between 6 and 8 in the evening still might get rung at 10 am if there are lines free and nobody with an open window ahead of them. The feature does not keep them back until evening. That is by design, because nobody is held back.
What it cannot do
- It cannot call outside your hours. Your schedule, your own window and the 9 am to 9 pm IST limit all still decide when a call may go out. If a contact's best hour is 10 pm, they are still rung inside your hours.
- It learns only from answered calls. A new list starts on fallback hours, and a brand-new workspace has little to go on.
- It cannot tell you why someone answers. It sees only that they did.
- It updates once a night, so today's calls will not change today's order.
Your own dialling can mislead it
This is the limit worth understanding. If your team only ever dials at 11 am, most answers come at 11 am. That tells you when you dial, and it says little about when people would have picked up. The whole-workspace fallback inherits that habit.
So if you want it to learn anything useful, spread your calling across the day for a while. Run your schedule across a wide window for the first few weeks, including the evening hours your staff may not love. Analytics keeps outbound and inbound apart and shows the hours people answer. It names a best hour for your workspace only when one hour has 20 outbound calls. Before then, treat any pattern with suspicion.
How to tell whether it helped
Compare answer rates, not your impression. Switch it on for one schedule and leave a similar campaign without it. A similar campaign means a similar list from the same source, in the same week, and ideally the same agent. Before-and-after comparisons on different lists tell you very little, for the same reason they fail in A/B testing.
Remember that it may change little if your campaign finishes its list within a couple of hours, because everyone gets picked in the same stretch anyway. The re-ordering matters when your list is bigger than your lines can reach in a day. That is also the case where a better order has most to offer.
Other places it shows up
Telecallers see the same line on a contact, saying when that person usually answers. It is a small thing, but a good caller planning their afternoon's callbacks will use it. And a second round of no-answers is a natural place to try it.
Cost and availability
It has no charge of its own and is on every plan, including the free trial. The calls it orders are billed as usual, which you can read about on the pricing page. If you are just starting out, read run your first campaign first, and come back to this when you have a few weeks of calls behind you.