# Adding an FAQ Section to the Product Page A/B Test

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/adding-an-faq-section-to-the-product-page-ab-test

## Hypothesis

If the installment, return, warranty and delivery questions shoppers get stuck on for this product are answered briefly on the product page, the last doubt before buying may go away. Purchase conversion rate may rise as a result, and support contacts on the same topics may fall.

## When to run this test

Support tickets that keep raising the same installment, return or delivery questions for one product, or visitors who go from the product page to help pages and never come back, are the cue for this test. When it works, visitors find the answer on the page, related support contacts drop and purchases go up. First confirm that every answer is current and consistent with the product's actual terms. Keep generic questions that are not specific to the product, campaign copy and sales messages out of the FAQ; they do not answer what visitors came to ask.

## Primary metric

**Purchase conversion rate (CR)**: Compare purchases per assigned visitor with and without the predefined FAQ section.

## Other metrics

- **Add-to-cart rate**: Track add-to-cart; by itself, this does not show how quickly a decision was made.
- **Support contact rate**: Compare support contacts per assigned visitor and changes in their reasons.
- **FAQ interaction rate**: Report answer opens within the variant; the control has no FAQ to open.
- **Return rate**: Track returns over the same follow-up period in both arms.

## What to check before and during the test

- **Change**: Add a predefined FAQ section to the product page.
- **Keep constant**: Keep FAQ content, product details, price and CTA structure unchanged.
- **Behavior**: Read FAQ interactions alongside purchase behavior.
- **Accuracy**: Keep answers current and consistent with the product's actual terms.
- **Separate tests**: Test placement or FAQ format separately.

## What to watch when reading the result

- Report FAQ interaction rate within the variant only, since the control has no FAQ to expand; frequent expansions alone do not mean the section works. Purchase conversion rate makes the call; compare support contacts and contact reasons per assigned visitor in both arms.
- Keep the answers short so the section does not grow into long, complex copy, and do not bury the FAQ so far down that visitors cannot find it. Leave price, images and buttons unchanged during the test so any difference comes from the FAQ alone.
- What you learn is tied to this FAQ content at the chosen position; measure the effect of position in a separate test with the same questions. Track return rate in both arms over matching follow-up windows.
- If support contacts and purchases are rare events, a small difference takes a long time to show. Work out the traffic you need up front with the [sample size calculator](https://alidemirbas.com.tr/en/calculators/sample-size-calculator), and if the result stays unclear, do not conclude that the FAQ has no effect.

## Frequently Asked Questions

### Which questions belong in a product FAQ?

Choose questions shoppers need answered before buying this product, including relevant delivery, returns and warranty questions. Avoid generic filler or campaign copy. Answers must match actual terms. Use purchase conversion as the primary outcome.

### Do frequent FAQ expansions prove the section works?

No. Report expansions within the variant only, since the control has no equivalent FAQ. Add-to-cart rate also does not establish faster decision-making.

### How should I measure the effect on support requests?

Compare support contacts per assigned visitor and changes in contact reasons. Relevant questions may decline if the FAQ answers them. Track returns over equal follow-up periods too.

### Can I put the FAQ at the bottom of the page?

Sometimes, provided visitors can find it. Critical information may need earlier access. This test measures the FAQ at the chosen location. Compare placements separately using identical questions.