How to measure an SMS campaign: 9 metrics that actually matter

Sent is not delivered, and delivered is not converted. These are the nine metrics worth tracking in every SMS campaign or flow, how to calculate them and what decision to make with each result.

Delivery metrics

They are the foundation: if the message does not arrive, nothing else matters. They come from the delivery reports (DLR) of each message.

1. Delivery rate = delivered messages ÷ sent messages. A sudden drop usually points to a base, content or specific operator problem.
2. Failures by cause: invalid number, landline, exclusion list, operator rejection, expired. Each cause has a different fix; we cover them in why an SMS is not delivered.
3. Delivery time: seconds between sending and the delivered report. Critical for OTP and alerts; less so for campaigns.

Response metrics

1. Click-through rate = unique clicks ÷ delivered messages. Measure against delivered, not sent, so you do not confuse a delivery problem with a content problem.
2. Reply rate (in two-way SMS) = replies ÷ delivered. Useful for appointment confirmations and surveys.
3. Opt-out rate = opt-out requests ÷ delivered. If it rises, you are sending too much or to the wrong audience.

Business metrics

1. Conversion rate = purchases, payments or confirmed appointments ÷ delivered messages. It requires links with UTM parameters or a unique promo code.
2. Cost per delivered message = total cost ÷ delivered. Remember that a long text or one with emojis may be billed as two or more segments; see bulk SMS pricing in Mexico.
3. Cost per result = total cost ÷ conversions. It is the number that lets you compare SMS with other channels.

For OTP flows, two metrics of their own

Verification has no clicks or sales, but two metrics say a lot:

• Verification rate = codes entered correctly ÷ codes sent. If it drops, check delivery, arrival time and code validity.
• Time to verify: seconds from requesting the code to entering it. An increase often anticipates user complaints.

How to analyze them

• Segment by operator, send time and message type: averages hide problems.
• Test one variable at a time (text, time, sender) with comparable groups.
• Keep transactional and promotional separate in your reports too.
• Use the same UTM scheme in every campaign to compare results month to month. If your links raise doubts, apply the rules in links in SMS without looking like fraud.

Sources

• Google, Analytics Help, “URL builders: collect campaign data with custom URLs”. support.google.com
• Google Analytics, Campaign URL Builder. ga-dev-tools.google

Frequently asked questions