> For the complete documentation index, see [llms.txt](https://docs.elevateab.com/elevate-helpcenter/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.elevateab.com/elevate-helpcenter/price-testing/troubleshooting-price-experiments.md). # Troubleshooting Price Experiments Symptoms, causes, and fixes for price experiments in Elevate A/B Testing that show the wrong price, show no price change, or collect no data. *** Find your symptom below. Each entry covers the likely cause, how to confirm it, and what to do. ### Everyone sees the control price The most common price experiment problem, and usually one of five things. | Likely cause | How to confirm | Fix | | --------------------------------------------- | --------------------------------------------------------------------------------------------- | ------------------------------------------------- | | Theme extension is off on your live theme | **Online Store > Themes > Customize > App embeds**. Elevate should be listed and switched on. | Switch it on and save | | The extension is enabled on a different theme | Check the theme you actually published, not a duplicate | Enable it on your live theme | | The experiment is not running | Check the experiment status in Elevate | Launch it, or complete review if it is your first | | You are looking at yourself | Your own browser may be excluded, or pinned to the control from an earlier preview | Open your store in an incognito window | | You arrived from a Google Ads link | Visitors from Google Ads are excluded from price experiments by default | Expected behaviour. Visit directly instead. | If none of these apply, contact support with your store URL and the experiment name. ### The price is right on the product page but wrong elsewhere Collection pages, search results, recommendation blocks, quick view, and cart drawers each display prices differently, and each needs to be configured for your theme. This is what Elevate's review of your first price experiment is for. If you are seeing an inconsistency after review, or on a page type that was not covered, contact support and tell them exactly which page type is wrong. They will correct it. Do not launch with a known inconsistency. A visitor who sees two different prices for the same product will not buy. ### The price on the page does not match the price at checkout Stop the experiment and contact support immediately. Include the product, the price shown, and the price charged. This is the one price experiment problem worth treating as urgent. ### Prices stopped changing partway through a run | Likely cause | Fix | | ------------------------------------ | -------------------------------------------------------------------------- | | A new theme was published | Enable the theme extension on the new theme. It does not carry across. | | The theme extension was switched off | Switch it back on | | The theme was edited or updated | Contact support so price display can be reconfigured for the updated theme | Note that switching the theme extension off does not stop an experiment. Carts already in progress may still be charged the test price. If you want an experiment to stop, stop the experiment. ### An experiment shows zero visitors | Likely cause | Fix | | ---------------------------------------------------- | -------------------------------------------------------------------------------------------------- | | Audience targeting excludes everyone | Open the experiment and review the audience filters. Combined filters narrow faster than expected. | | Traffic allocation is set to zero for that variation | Check the allocation slider | | Traffic isolation left no traffic available | Check how much isolated traffic other experiments have reserved | | The products are not being viewed | Confirm the products in the experiment get traffic | ### Another app changes the price after Elevate does Bundle apps, quantity break apps, subscription apps, and some page builders write prices to the page themselves and can overwrite the experiment price. Contact support with the name of the app. Most common cases are already handled and can be switched on for your store. ### A CSV upload failed Uploads are all or nothing. If any row fails, nothing is imported. | Message | Meaning | | ------------------------------- | -------------------------------------------------------------------------------------- | | Row level errors | The row number is shown, counting the header as row 1. All errors are listed together. | | Missing or unrecognized columns | Check the header row against the required columns | | Unknown product or variant ID | Confirm the IDs exist on your store, in numeric or GID format | | Unsupported currency | The currency must be one your store sells in | Download the sample CSV from the upload dialog and compare it against your file. Remember the control needs no rows, and Variation IDs run from 1 to 4. ### Autofill produced prices I did not expect | Symptom | Cause | | ------------------------------------------------- | ----------------------------------------------------------------------------------------------- | | Prices are whole numbers when you wanted decimals | Rounding is set to Up, Down, or Nearest. Set it to No round. | | Prices did not end in .99 | Rounding rounds to a whole currency unit. It does not produce price endings. Use CSV for those. | | The compare at price changed instead of the price | Change type was set to Compare Increase or Compare Decrease | | Only some products changed | Autofill applies to selected products only. Check your selection. | Use **Reset All** to clear everything and start again. ### Results look wrong | Symptom | Likely explanation | | ------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Fewer visitors than expected | Google Ads traffic is excluded from price experiments. Audience filters also reduce the pool. | | Numbers do not match Shopify | Elevate counts unique visitors and attributes revenue to the variation each one saw. Shopify counts orders and sessions differently. See [Analytics Discrepancies](https://docs.elevateab.com/elevate-helpcenter/analyze/tracking/analytics-discrepancies). | | A variation has orders at a price you did not set | A discount code or automatic discount applied on top of the test price | | Profit columns are missing | Product Costs are not set up. Enable them to see profit. | | One or two orders swing the result | Normal early in a run on low traffic. Wait for more data. | | The result reversed after a sale | If the sale changed catalogue prices, the experiment is invalidated. Run sales as discounts. | ### Still stuck Contact support through the chat widget in the Elevate app or at . Include your store URL, the experiment name, which theme you are on, the product where you see the problem, and what you expected against what happened. See [Help and Support](https://docs.elevateab.com/elevate-helpcenter/getting-started/help-and-support).