2026.05.19

Until now, the questions in your studies were "fixed". You could influence them through your product name and description, but you couldn't change the wording itself, and if something read awkwardly, there was nothing you could do about it.

That's now changed. When creating a study you can now review each survey question, and if the wording sees off, choose from three AI-generated alternatives before you launch. Every option preserves the original intent and structure of the question, only the phrasing differs, so you get the flexibility to pick what fits your product best - without risking the validity of your results.

On top of that, all questions now go through an automatic semantic check at creation, so even if you don't customize anything, the wording that goes out to respondents will be natural and contextually accurate for your product and language.

Writing a product description from scratch just to set up a study is busywork. Now you can paste your product page URL directly into the description field. Priceagent fetches the page, pulls the relevant content, and uses it consistently across your entire study — intro, questions, and all — in any language you run it in.

The result: less setup time, fewer errors, and studies that accurately reflect your product from the moment you launch them.

The accuracy of your study results depends partly on how well the target audience is defined relative to the overall market. We have improved backend settings for locating your audience, based on your product description and your past research, to improve accuracy, resulting results rolling in faster. You don't need to configure anything, it happens in the background every time you launch.