I hear that Jarvee is really bad when it comes to the manual scraping of targets so I’m interested in seeing where the benchmark stands. How expensive is it really? Additionally, does the change in source have any real impact on the longevity of the scrapers? Say 3 scrapers to extract 1k from followers of @ as opposed to 5 scrapers to extract 1k usernames from “users who interacted with target post”.
How many scrapers do you typically burn through to gather 1k usernames from followers of source?
Its’ the amount of Api call the scraper does that causes issues. IE the more filters you use the more burnout you will get. This is why people use specific users with no filters (as they scraped externally with filters)
I’m not sure as to why jarvee kills scrapers more than external methods. But i’m guessing its something to do with the old IG api they use. Wheras coders use selenium etc and no api
So does that mean if you were to use selenium, you can avoid making any API calls? Will a singular scraper do the job?
No there still are limits and you need to be able to hide the fact you are using selenium also otherwise itl get blocked fast.
The limits are much higher with external scrapers. Just not unlimited. Ie. Sometimes I can scrape 5-8000 profiles before captcha externally with aged scrapers
How many filters do you usually set for those operations? Or is it just 5000-8000 followers of a specific source?
Depends on the client or what I’m doing. Can be non bussiness, non private, english speaking, posts etc.
Varies from what I’m doing.
to get 1k from followers of @ user I don’t burn any scraper just using one scraper I can get more than 2k but I guess you are asking how many 1k of users can you get before the account is burnt.
Oh wow that’s pretty good actually. And that’s no filters on correct?
What do you mean they scrape externally? As in, using Jv just to extract all followers from one source and then manually applying all filters? Or using another app besides Jv to filter those?
some scrapers have few filters other don’t but those numbers are consistent so scraping using JV work fine for me
I use a script to gather all my sources then add to jarvee. Saves scrapers being murdered