Delivery rate vs. reply rate: measure what happened after the send
Define SMS delivery, reply, conversation and appointment metrics with clear denominators and campaign examples for sales teams.
·3 min read
Two dashboards can report different reply rates for the same campaign because they divide by different totals. One uses sent messages, another uses delivered contacts, and a third counts every response rather than unique people.
Before comparing performance, define the funnel. A shared set of denominators lets agents and managers discuss whether the problem is delivery, relevance, handling or appointment conversion.
Distinguish messages, segments and people
One contact may receive several messages, and one long message may produce several SMS segments. Keep those units separate. Use segment totals for usage analysis and unique contact totals for most audience conversion questions.
A campaign that sends three texts to 100 people has 100 contacted people and up to 300 message attempts, with potentially more segments. Calling the latter '300 leads reached' overstates the audience.
Use an explicit delivery denominator
Define message delivery rate as finalized delivered messages divided by the comparable submitted message attempts, excluding or separately reporting pending records. Explain how failures before provider submission are treated.
Provider delivery statuses describe network handling; they do not prove that a person read the message. Telnyx's metrics documentation distinguishes sent, delivered and inbound activity. Keep that distinction in the agency report instead of labeling delivery as engagement.
Calculate unique-contact reply rate
For a sales campaign, count unique people who replied during a defined attribution window. Divide by the eligible contacted cohort or delivered-contact cohort, then label which denominator you used. Multiple replies from one enthusiastic prospect should not inflate audience response.
Illustration: 1,000 contacted people, 900 with at least one delivered message and 90 unique replies produce a 9% contacted-person reply rate or a 10% delivered-person reply rate. Both calculations can be correct; they answer different questions.
Separate replies from useful conversations
A wrong-number report, opt-out and qualified buying question are all inbound responses, but their business meaning differs. Classify responses without hiding negative outcomes. Track qualified conversations and appointments independently.
Using the same example, if 30 of the 90 responders book, the responder-to-booking rate is 33.3%, while the contacted-person booking rate is 3%. Do not present the larger figure without its denominator or imply the illustration is a platform benchmark.
Diagnose one funnel stage at a time
Low delivery calls for reviewing numbers, registration, errors, content and list quality. Healthy delivery with few replies calls for reviewing message fit and audience context. Many replies with few bookings may indicate slow handling, weak qualification or limited availability.
Compare like-for-like cohorts and consistent observation windows. A campaign launched yesterday should not be judged against one that has completed a two-week sequence. Preserve the definitions with the report so weekly changes remain interpretable.
Key takeaways
- →Keep SMS segments, messages and unique contacts separate.
- →Delivery is a network status, not proof of reading.
- →Label every percentage's denominator and attribution window.
- →Review reply quality and appointments separately from total inbound activity.
Put this into practice with Text2Sale
Import leads, create follow-up campaigns, organize replies, and connect texting, calling, and appointments in one workspace.
Frequently asked questions
What is a good SMS reply rate?
There is no useful universal number without audience, consent, message purpose, cadence and denominator. Establish a baseline from comparable campaigns, then measure improvements at each funnel stage.
Should opt-outs count as replies?
They can be included in total inbound response counts, but report them separately from interested conversations. A higher reply rate caused by more opt-outs is not improved sales performance.