It’s such a blessing when the YT algorithm does it’s job and shows me this type of brilliant channel with such small subs. Love the content already. Subbed
Thanks! I should add that since I made this video I tried out Bing and actually prefer it for this. The work flow is the same but bing pages start at 1 and increase by 10 (so start=1 then 11 then 21 etc)
Then stay tuned! Tomorrow I’ll be dropping one (unless England surprise me by playing fluent attacking football leading to a hangover and a wasted day. But that seems 99% unlikely to happen)
Good to hear! I made this version very very basic but I’ve been involved with much more interesting flows that really get into verification, AI calling and ads. All cool stuff!
Hi Georg! That was absolutely best tutorial video i've ever seen on RU-vid man! By the way, i would like to ask, do you have any suggestions as roadmap for a self learner like me? I appreciate!
Great video! Lots of inspirations, thanks! Just a quick comment re your last statement. you don't need to run it "over and over again" :) in n8n you can "pin" the output data of particular node and it makes everything a whole lot easier :D
My favourite comment! I never got to play with n8n like I have with make.com. I needed to build something specific and so just focused on that without the other aspects of learning! But this makes like that much better
Unsatisfactory answer: I did this based on the api limits at the time (The rate limits are: 60 calls per minute, 5 calls per second (burst allowance)). It’s a bit of an old video though and I’ve since updated the flow and moved to findymail for this task. It should be a video from maybe a month ago if you want to check it out!
I'm stuck at the part where I need to identify the string indicating the "next" button in the second field below "does not match regex" in the initial IF node. I've tried -pasting in the entire data content from the previous node -"inspecting" the next button in the actual google/bing page to get the code It still however doesn't seem to working. (I.e., it's returning true and therefore updating the airtable as "complete") Any help would be greatly appreciated. Thank you for sharing this incredible flow.
It’s very hard for me to troubleshoot without seeing the workflow. Feel free to set a call with my on my tidycal from my bio page or in the product description
@@gorgeousgeorgous Hi there! I've bought the example you've listed on Gumroad. Thanks for sharing it! However, I'm a bit puzzled about how the 'Start Page' is being determined. From your video, I understand that 'Start Page' is a value coming from Airtable and it seems to be set or updated within N8N. What I'm not quite grasping is the exact method by which the search query begins from page 10 upon making the HTTP request. Could you clarify this part for me, please?" To clarify: what is within the HTTP request itself that would allow a generic query of the search to occur at a paginated page?
@@christianclark4612 In your airtable base for managing the campaign you have a column called start page and, as a default, when you create a campaign you give it a value of zero. then when the automation runs it takes that start page and adds it by 10 each time, thereby moving to the next page of search results
@@gorgeousgeorgous That makes sense, but what in the search itself is defined as the start page? If the airtable is a value, and the code in N8N is incrementing it, then what is actually informing the start page? I don't see it in the search query or the search operator itself?
Yep! All my gum roads have a step by step guide set up guide document to accompany them. Plus my calendar is public (link in bio) so you can reach it to a face to face whenever you like
Hi George, awesome video here! Is this process relatively similar to do on make? And is the code all easy enough to simply replicate into the automation? Thanos
Yep @thewizardtower. Here is a video of it in action going through a google result and collecting new people on every run. It’s a video I made to help a client who had trouble implementing it just this morning! www.loom.com/share/891cda6f8d304729a5e2884263095035
@@rainerlummerstorfer2425 Yep! This is the basic template because I can’t predict the exact usage of each person. But to escape the 429 you just need to 1) set a filter after the http module to only pass through 200 code results 2) set a repeater to reprocess the 429 calls after a delay
N8N lets you run a much higher number of operations. So if you were to start scraping in high volumes it’s the way to go. As a note too, I’ve since adjusted this flow to work in bing (just need to change the regex and add a deduplication) module as it seems to work better
@@gorgeousgeorgouswow! Can n8 also be used to cold email those 80,000 emails? Was originally thinking of buying instantly but this seems like that on steroids
For the bing search, what exactly did you enter under the "does not match regex" field? Was it the entire data from the search, or did you have a criteria by which you were able to identify the code that manifests the "next" button?@@gorgeousgeorgous
It blocks maybe 1 in8 requests which is why 1) the scenario should only run every 5 mins 2) there is a filter to pause and delay it if the request is unsiccessful