How to A/B test product photos, packaging or titles for free
You do not need a paid panel to learn which version people prefer. Six rules that make a free poll worth trusting.
You do not need a paid panel to find out which of two photos, packages or titles people prefer. If you have customers, followers or a community, you can ask them, as long as you set the test up so the answer means something.
- Ask one question. "Which would you click?" or "Which would you buy?" Pick the one that matches your decision. Do not ask people which they "like", because liking is not buying.
- Change one thing. If version A has a different photo and a different price from version B, you will not know which made the difference. Keep size, background and wording the same apart from the thing you test.
- Use neutral labels. Call them A and B. Do not tell voters which one you made or prefer, and do not post your own vote first.
- Ask the right people. Your customers and your niche community beat friends and family. A friend who would never buy the product is noise.
- Get enough votes. With 30 votes, a 70/30 split is unlikely to be chance (about a 4% chance). A 55/45 split with 30 votes means nothing yet; even with 100 votes it is still unclear. Look for a clear gap or keep collecting.
- Read the reasons. The "why" behind each vote often tells you what to fix in the losing version.
Do it free with AskThousand
Our free A/B poll lets you upload 2 to 4 photos or short texts, share one link, and watch real votes come in. The results page shows the chance that the leader is really the favourite of people like your voters, so you know when to stop. Photos are checked before the poll goes live, and the poll is deleted after 90 days.
The winner comes only from real votes, so the result reflects what real people chose, not a guess.
Questions
How many votes do I need for an A/B test?
It depends on the gap. A 70/30 split with 30 votes is fairly clear; a 55/45 split can need hundreds of votes. Our poll shows the chance that the leader is really ahead, so you can see when the result is clear.
Can I test packaging or logos the same way?
Yes. Any visual choice works: main images, packaging, logos, covers, thumbnails. Keep everything except the tested element the same.
Is the AskThousand A/B poll really free?
Yes. You can make up to 5 polls a day; each poll and its votes are deleted after 90 days.
Check your product page in under a minute.
100 simulated buyers matched to your product's audience read your title, price and description and tell you who would buy, who would pass and why. Free, no account.