A/B Link Testing

A/B link testing splits traffic from one short link between two destinations — turning every distributed link into a controlled experiment with no additional setup required at the destination.

---

Definition

A/B link testing is the use of link rotation as the mechanism for split testing. A single short link is configured to distribute incoming traffic between two destination URLs according to a configured percentage split. Both destinations receive real visitors from the same link — enabling a controlled comparison of two variants without creating separate links or distributing different URLs across channels.

A/B link testing in Cuttly is available from the Single plan ($25/month).

A/B and A/B/C Testing in Cuttly

Cuttly supports both A/B testing (two destination variants) and A/B/C testing (three destination variants). The available variant count depends on your subscription plan — check your plan's features at cutt.ly/pro-pricing.

The traffic split between variants is set using a slider — you can configure any proportion, not just equal splits. For example: 70% to A and 30% to B, or 60% to A, 25% to B and 15% to C.

Click counting for alternative destinations (B and C) follows the same counting method — total or unique — as configured for the main link. Counting is consistent across all variants.

How to Set Up A/B/C Testing in Cuttly

1. Go to the list of short links in your dashboard
2. Click the ABC test settings icon for the short link you want to test
3. Depending on your subscription plan, you will see the A/B form or the A/B/C form
4. Enter the alternative destination URL(s) — B for A/B, or B and C for A/B/C
5. Set the traffic split proportions using the slider
6. Save — the rotation becomes active immediately

Full setup guide: How to Set A/B/C Test for a Short Link.

Link-Level vs Page-Level A/B Testing

Both methods are valid; they answer different questions. A/B link testing answers: "Which destination page or offer performs better?" Page-level A/B testing answers: "Which version of this specific page element performs better?"

What to Test with A/B Link Testing

• Two landing page variants. Different headline, different layout, different social proof — complete pages built and compared via link-level split.
• Two offers. 20% discount vs free shipping offer — same product, different promotion mechanics.
• Two content formats. Long-form article vs short video — which format drives better engagement from the same link click.
• Platform migration testing. Existing page vs newly rebuilt page — validate new page performance before full cutover.
• Two CTA destinations. "Book a demo" page vs "Start free trial" page — testing which ask drives better conversion from the same top-of-funnel link.

Statistical Considerations

A/B tests require sufficient sample size to produce statistically reliable results. Common errors:

• Ending tests too early. A few hundred clicks is not enough sample for most conversion rate tests. Determine the required sample size before starting — based on the minimum detectable effect and expected conversion rates.
• Testing multiple variables simultaneously. Changing headline AND layout simultaneously makes it impossible to attribute performance differences to either change specifically. Test one variable at a time.
• Ignoring seasonal or timing effects. A test run over a period that includes unusual events (holidays, news cycles, promotions) may produce results that do not represent typical performance.

Related Terms

• Link Rotation
• Conversion Tracking
• CTR
• UTM Parameters
• ROI of Short Links