I want to get an output of companies that have a pricing page with both conditions are met: their pricing page has at least one plan with a specific price, and there is at least one other plan that has no price and has a CTA contact sales. The output is just No even if the output should be yes. I even tries to simplify the ask so it will only add yes if there is a pricing page, but the output is always no. when i began this flow it was working well but i made some changes now its completely off and i spend many credits just to test it but each time is giving me the wrong answer.
Hey @danitbensimon - Could you please share the run links for this from the https://www.gumloop.com/history page? The run links will have the “run_id” in the URL.
Also, please make sure to set the access to “anyone with the link can view”
Thank you. Could you share the run link as well from the history page so I can take a look at the error? The run links will have the “run_id” in the URL.
Hey @danitbensimon - If you view the inputs/outputs of the AI node you’ll see the context that is being passed onto AI. In both the cases I don’t see any pricing information on those two websites. Even if I view the sites manually I didn’t find that information. If it is hidden somewhere you might have to scrape multiple pages and improve the prompt within the node.
Thanks but I don’t thinks that’s the issue.
Even in pages with pricing pages the AI couldn’t find them. It did find them at the first few runs but I changed something and now it can’t. All I want is the scraper to tell me if there is a pricing pages and it doesn’t and again I know it can because it did. I’m referring to websites where the pricing is written in the header of the website
this company has a pricing page https://www.jotform.com/ but the AI can’t find itsee the run here:
https://www.gumloop.com/pipeline?run_id=M5SoYVPdDW7wo3PiqaxQT6&workbook_id=eEJ5LJ4QsVUgfEfHx2WCqi
The run link you shared is for greyorange.com where there is no pricing page. Could you share the one for Jotform?
sorry see it for https://www.docker.com/
https://www.gumloop.com/pipeline?run_id=Pu4nurNRoQUN2oPbmRoX75&workbook_id=eEJ5LJ4QsVUgfEfHx2WCqi
Docker has a pricing page and the response from the AI was also ‘YES’, what seems to be the issue here?
I think the issue might be that you’re seeing mismatched data on your Google Sheet. You should use the Google Sheet Updater node here which updates the existing rows instead of the writer node which adds new rows at the end of your sheet.
amazing thanks but the nodes cant connect
https://www.gumloop.com/pipeline?workbook_id=eEJ5LJ4QsVUgfEfHx2WCqi
Disable loop mode on the sheet updater node since you’ve ‘Update Multiple Rows’ set as writer mode already.
Thank you for the quick response. Now how can I only test a few rows now that I disconnected the list trimmer?
ok i tested it and i get no for a company like with does has a pricing page:
https://prezi.com/
https://www.gumloop.com/pipeline?run_id=iVQtX2fe5WW2ZQ4YVZmPph&workbook_id=eEJ5LJ4QsVUgfEfHx2WCqi
The pricing is in the header menu and it didnt came through in the website scraped content, you can check that by clicking view inputs/outputs of the scraper node from the run log.
You can try screenshotting the page and passing the content to an AI Analyze Image node, here’s an example with Prezi: https://www.gumloop.com/pipeline?workbook_id=mJXN2HR3bSxXqESksHLf3D&run_id=eYh4FdChpAewghF89ubjCS
there was an error in the image scraper:
https://www.gumloop.com/pipeline?run_id=ZiYcKs5ZtnkaeTdEdKeeny&workbook_id=eEJ5LJ4QsVUgfEfHx2WCqi
This topic was solved and automatically closed 4 days after the last reply. New replies are no longer allowed.