Small changes in wording, offer or timing can double the response to an SMS. A/B testing tells you which version works, instead of guessing.
How it works
- Pick one thing to change.
- Split a random sample of your list into two groups.
- Send version A to one group and version B to the other, at the same time.
- Measure responses (codes, replies, clicks, sales).
- Send the winner to the rest of the list.
What to test
- The offer: 20% off vs KES 500 off.
- The opening: leading with the offer vs the customer’s name.
- Urgency: “this weekend” vs “today only”.
- Call to action: “Reply YES” vs “Call now”.
- Timing: 10am vs 6pm — see best times to send SMS.
- Sender ID: brand name vs branch name.
Sample size
Small tests give noisy results. As a rough guide, use at least a few hundred recipients per version, and more if response rates are low. If the difference between versions is tiny, treat it as a tie and test something bolder.
Example
- A: “Weekend offer: 20% off all shoes at BRAND. Visit our Moi Ave branch. STOP to opt out”
- B: “Grace, your 20% off shoes is ready — this weekend only at BRAND Moi Ave. STOP to opt out”
Give each version its own redemption code so you can count results.
Keep a testing log
Record each test, the result and what you learned. Over a few months you’ll build a playbook for your audience.
Related: measuring SMS ROI and the SMS marketing guide.
Get started
Create a Connect Media account — SMS from KES 1.0, no minimum top-up, credit that never expires, and one account for Safaricom, Airtel and Telkom. Call +254 707 339 945 or email info@connectmedia.co.ke.
