Free tool
A SAGARIS calculator: what does another line buy?
Nothing here is a benchmark. Every figure below is computed from the inputs you set.
Run it on your own numbers
Illustrative inputs| Lines | Attempts per hour | Expected connects per hour | Handled conversations per hour | Share of hour talking | Connects with nobody free |
|---|---|---|---|---|---|
| 1 | 86 | 4.3 | 4.3 | 28.6% | 0.0% |
| 2 | 135 | 6.7 | 6.6 | 43.8% | 2.5% |
| 3 | 168 | 8.4 | 8.0 | 53.3% | 4.9% |
| 4 | 193 | 9.7 | 9.0 | 59.7% | 7.3% |
| 5 | 214 | 10.7 | 9.7 | 64.4% | 9.5% |
| Published ceilings at other vendors | |||||
| 6 | 231 | 11.5 | 10.2 | 67.9% | 11.7% |
| 10 | 285 | 14.3 | 11.4 | 76.2% | 19.7% |
Model assumptions
Independent lines at one connect rate; a round is n simultaneous attempts costing c seconds; the rep dials again only when free; no per-contact, per-number or calling-window limits; no abandonment governor. Attempts per hour are a model quantity; the product caps each number at 50 dials a day and holds contacts outside their calling window. Nobody free is the model's output, not a measured abandonment rate.
SAGARIS dials 1 to 5 lines, a count the rep sets for each wave. Rows 6 and 10 model published ceilings: CloudTalk 10 (cloudtalk.io/pricing), JustCall 10 (justcall.io/product/parallel-dialer), Elto 6 (eltodialer.com), read October 8, 2026.
How the model works
Throughput
Rounds per hour x the chance that at least one line connects.
The overflow expectation
n times p, minus the probability that at least one of the n connects.
What the model leaves out
Number caps, calling windows, the governor. Exact only under its assumptions.
How many lines does SAGARIS dial?
Up to five, suggested for each wave; the rep sets the count. Five is the ceiling.
Run it on your own list, inside the dialer.
Up to five lines at once. Pre-launch: $499 per founding seat per month.
Because a rep can only be in one conversation at a time, and every conversation takes minutes out of the dialing half of the hour.
It is what happens when two people on the same dialing round pick up at once: the rep can take one of them, and the other is held. That second call is the cost of running more than one line, and it is the number the calculator on this page works out for you. The figure on this page is exact under the model's assumptions rather than a rule of thumb: for a round of n lines at connect rate p it is n times p, minus the probability that at least one of the n connects.
Your own, measured from your own list. We do not publish a benchmark here because we do not have the call volume to support one honestly. Vendors that do run large dialing platforms publish theirs, and their numbers are worth more than a figure we would have to guess at.
Up to five, and the count is suggested for each wave rather than fixed for the whole campaign, and the rep sets it: more lines when a pickup looks unlikely, fewer when it looks likely, and a single dedicated line when the contacts on the wave are likely to answer. Five is the ceiling. The table above shows why a higher number buys less than it seems: the share of connects with nobody free climbs faster than handled conversations do. To see where five sits against what other vendors publish, Nooks vs Orum and JustCall vs Kixie carry their dated line counts.