# Writing Price Decimals as Superscript A/B Test

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/writing-price-decimals-as-superscript-ab-test

## Hypothesis

If the decimals are set small and in superscript, the eye may go to the whole-number part first and the price may seem lower at a glance. The purchase conversion rate might go up, but a display that makes the exact amount harder to read could bring the gain back later as disputes and returns.

## When to run this test

Sessions that stall on the price area and then leave, survey comments about cost, or a heatmap where attention clusters on the price block suggest visitors find the price high. If superscript decimals help, the price feels lighter and visitors move to purchase more readily. If visitors leave because of taxes or extra fees, make the total clear first, and do not test a display that blends those decimals into the main amount. In markets that regulate price display, check the applicable rules before changing the format.

## Primary metric

**Purchase conversion rate (CR)**: Compare purchases to see whether price formatting changes conversion.

## Other metrics

- **Average purchase value**: Read completed-order value alongside conversion.
- **Price comprehension**: If surveyed, ask the same brief question in both arms to check whether visitors can state the total payable amount.
- **Price-related disputes and returns**: Track recorded disputes and returns related to an unexpectedly high price over the same follow-up period.
- **Time to purchase**: Read time alongside purchase rate; longer time is not automatically negative.

## What to check before and during the test

- **Format**: Compare superscript decimals with standard-size decimals.
- **Readability**: Check that smaller decimals remain clear on mobile.
- **Price range**: Review whether the effect differs for lower- and higher-priced items.
- **Rounding perception**: Use the same method in both arms to ask whether visitors can identify the actual total.
- **Segments**: Define segments before display; do not infer price sensitivity from purchase behavior in the test.

## What to watch when reading the result

- A conversion gain alone is not enough; track price-related disputes and returns over a follow-up window of equal length and check average purchase value. If price comprehension drops or the decimals become hard to read, especially on mobile, the gain does not earn a launch, because the total payable must always stay readable.
- Credit the result to the format only if the price level stayed the same; if prices changed too, the difference may not come from formatting. Apply the format consistently across comparable products during the test, and read the effect within predefined price bands.
- Do not treat a longer time to purchase as negative on its own. When the conversion difference is small, check whether chance could explain it with the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test), and if it stays unclear, report it as inconclusive rather than as a small win.

## Frequently Asked Questions

### How can correct price reading be checked?

Use the same comprehension task in both arms. Preserve an accessible decimal separator and test screen-reader output. Session recordings alone do not reveal the amount understood.

### What matters if conversion increases?

Use purchase conversion as the primary outcome and inspect price-related disputes and returns over equal windows. Smaller fractional text must not obscure the payable total.

### Is longer time to purchase necessarily bad?

No. Interpret timing diagnostically. Examine predefined price bands rather than defining sensitivity from who purchased after treatment.