## Why SMS Metrics Deserve a Closer Look An SMS campaign produces a surprising amount of data for a message that is only a line or two long. Every text leaves a trail: it was accepted by a provider, handed to a carrier, delivered to a handset or not, perhaps tapped, perhaps answered with STOP, and in every case it cost money.
Those signals are the raw material for better campaigns, but only if they are measured consistently. The trouble is that two people can look at the same campaign and calculate different numbers. One divides clicks by messages sent, another by messages delivered. One counts every tap, another counts unique people. One includes the cost of failed messages, another leaves them out. When the definitions drift, comparisons between campaigns, countries, and routes stop meaning anything.
This guide focuses on four metrics that together describe most of what happens to an SMS campaign: - **Delivery rate**: did the messages reach phones? - **Click rate**: did people act on the message? - **Opt-out rate**: did the message cost you subscribers? - **Cost per click**: what did each engaged recipient cost? For each one, you will find a clear formula, the choices that change the result, common traps, and worked examples.
The numbers in the examples are illustrative only, chosen to make the arithmetic easy to follow. They are not benchmarks, and you should build your own baselines from your own sends, because results vary widely by country, audience, route, message type, and consent source. ## Start With a Shared Vocabulary Before calculating anything, agree on what each count means. Most disagreements about SMS performance turn out to be disagreements about definitions.
### Counting units: messages, segments, and recipients An SMS longer than one segment is split and billed as several segments, even though the recipient sees a single message. The limit depends on encoding: GSM-7 characters allow more per segment than Unicode, and a single emoji or accented character outside the GSM-7 set can switch the entire message to Unicode. Our guide to SMS character limits, GSM-7, Unicode, and segments explains the rules in detail.
For metrics, this matters because "messages" can mean three different things: - **Recipients**: unique phone numbers targeted. - **Messages**: one logical text per recipient. - **Segments**: billable units, which can be higher than messages. Engagement metrics such as click rate and opt-out rate should normally use recipients or messages. Cost metrics usually need segments, because that is what you pay for. Mixing the two is one of the most common sources of confusing reports.
### Status stages Each message passes through several stages, and your provider reports them through delivery reports (DLRs): - **Submitted** or **sent**: your platform handed the message to the provider. - **Accepted** or **queued**: the provider took responsibility for it. - **Delivered**: the carrier confirmed the message reached the handset. - **Failed**, **rejected**, or **undelivered**: the message did not get through. - **Expired**: delivery was attempted but the validity period ran out.
- **Unknown** or **pending**: no final status has arrived yet. Status names and their reliability vary between providers and destinations. In some countries and on some routes, delivery confirmations are not returned at all, or are generated by an intermediate network rather than the handset. For a full breakdown, see our guide to SMS delivery reports and DLR statuses. ### The measurement window Delivery reports can arrive minutes or hours after sending.
Clicks and opt-outs continue to trickle in for days. If you report a campaign one hour after sending and compare it with another campaign measured after a week, the comparison is unfair. Pick standard windows, for example a delivery window and a separate engagement window, and use them for every campaign. ## Delivery Rate Delivery rate answers the most basic question: how many of the messages you sent actually reached a phone?
### The formula The most widely useful version is: **Delivery rate = delivered messages ÷ messages sent × 100** "Messages sent" here means messages submitted to the provider, after your own pre-send filtering. Contacts removed by suppression, validation, or quiet-hours rules before sending were never sent, so they should not be in the denominator. Track them separately as "filtered before send," because they still tell you something about list quality.
### Illustrative example Suppose a campaign targets 10,000 numbers. Pre-send checks remove 400 that are on the suppression list or fail number validation, so 9,600 messages are submitted. After the delivery window closes, the reports show: - 9,120 delivered - 300 failed or rejected - 60 expired - 120 with no final status Delivery rate = 9,120 ÷ 9,600 × 100 = **95.0%** These figures are invented to illustrate the calculation.
The important detail is the 120 messages with no final status. Whether you count them as failures, exclude them, or report them separately changes the result. Excluding them gives 9,120 ÷ 9,480 = 96.2%. Counting them as undelivered gives 95.0%. Neither is wrong, but you must pick one approach and stick with it. ### Three delivery numbers worth tracking A single delivery rate hides useful detail. Consider tracking: 1. **Confirmed delivery rate**: delivered ÷ sent.
The conservative headline number. 2. **Failure rate**: failed + rejected + expired ÷ sent. The messages that definitely did not arrive. 3. **Unconfirmed rate**: unknown or pending ÷ sent. The messages you cannot judge. A rising unconfirmed rate often points to a route or destination where reports are unreliable, not necessarily to a delivery problem. A rising failure rate is a stronger signal that something is wrong with the list, the route, or the content.
### What lowers delivery rate - **Invalid or badly formatted numbers.** Numbers without a country code, with leftover trunk prefixes, or with typos. Normalizing to E.164 before sending prevents many of these. - **Disconnected and reassigned numbers.** Older lists accumulate numbers that no longer work or now belong to someone else. - **Carrier filtering.** Content, sender IDs, or traffic patterns that a carrier treats as suspicious can be blocked.
- **Route problems.** A provider route that is degraded or not suited to a destination. - **Sender ID issues.** Some countries require registered sender IDs or restrict alphanumeric senders. Failure reason codes are your best diagnostic tool. Group failures by reason, country, and route rather than looking at the total. A spike in one reason code on one route tells a very different story from failures spread evenly across the list.
## Click Rate Click rate measures how many recipients acted on the link in your message. Because SMS has no reliable equivalent of an email open, clicks are usually the first engagement signal you can measure.
### The formula There are two common versions, and the difference matters: **Click-through rate (CTR) = unique clickers ÷ delivered messages × 100** **Click rate on sent = unique clickers ÷ messages sent × 100** Using delivered messages as the denominator separates engagement from delivery. If a campaign's delivery rate drops, CTR on delivered stays comparable, while click rate on sent falls. Many teams report both.
### Unique clicks versus total clicks A single recipient might tap a link three times. Total clicks would count three, unique clicks would count one. For rates, unique clicks are almost always the right choice, because the question is how many people engaged. Total clicks can still be useful as a secondary signal, for example to spot links that people return to.
### Illustrative example Using the earlier campaign with 9,120 delivered messages, imagine the tracking link records 1,050 total clicks from 730 unique recipients. - CTR on delivered = 730 ÷ 9,120 × 100 = **8.0%** - Click rate on sent = 730 ÷ 9,600 × 100 = **7.6%** - Clicks per clicker = 1,050 ÷ 730 = **1.44** Again, these numbers exist only to show the arithmetic.
Your own rates depend heavily on audience, offer, timing, and how recently people opted in. ### Bot clicks and link previews Not every recorded click is a person. Link-preview services, security scanners, and some device or messaging app features can fetch a URL automatically. These fetches can inflate click counts, especially when they happen within seconds of delivery. Ways to reduce the noise: - Look at the time between delivery and click.
Large clusters of clicks within a second or two of delivery deserve scrutiny. - Check user agents and IP ranges for known scanners and preview fetchers. - Compare clicks with downstream events such as page views that run JavaScript, add-to-cart actions, or sign-ins. - Report "human clicks" separately once you have a filtering rule you trust. Many of the same techniques used for email apply here; our guide to email click tracking, custom domains, and bot clicks covers them in more depth.
### Link setup affects both clicks and trust Click tracking usually
Related Articles
Attribute SMS revenue honestly with click windows, last-touch vs multi-touch, holdout lift tests, and cost-per-conversion guardrails.
SMS has no undo. Use this pre-send QA checklist to catch segment surprises, broken merge fields, bad links, wrong senders, and mistimed sends.
Learn how AI intent scoring ranks inbound SMS replies so your team answers high-value leads first — without losing the human touch.
Stop over-messaging: set SMS frequency caps, cooldowns, kill switches, quiet hours, and alerts that pause sends—plus a rollout checklist.
The text earns the tap, but the page earns the result. Learn how to build SMS landing pages that match the message, load fast on mobile data, work with one thumb, handle late
Customers want relevant messages. They also want stronger privacy. That tension shapes SMS and email programs today.
Unlock the secrets of SMS character encoding and segmentation. Optimize your mobile marketing campaigns for deliverability, cost-efficiency, and impact.
Sent is not delivered. Learn what SMS delivery report statuses mean, why some reports never arrive, and how to use DLR data to fix routes, lists, and timing.
Explore SESender
SESender brings audience preparation, contact validation, sender and provider controls, scheduling, delivery tracking, and campaign reporting into one workspace. Review the current product and pricing information before deciding whether the platform fits your messaging workflow.
Explore the platform or review pricing.