This is my first post on this forum after days spent on reading, getting familiar with it and getting hooked on that search function. First, want to say thanks to all of you for the quality community you have built around here and the knowledge that you provide. I, of course, would try to give back my personal findings and knowledge when I see fitting opportunities.
Now, on to my question:
I’ve been testing out different proxy providers recently and have had some good and some bad results, as expected. Now, the natural thing you would do is to stick with the one that has shown best results and scale your accounts using them, right? However, one thing I have been thinking about is the necessity of Plan B in case of proxy provider failure. Namely, I was wondering is it better to put all of the eggs inside one basket (when you find that one really good proxy provider) or split them in couple of different providers that have the same (or slightly worse) quality as your main provider. I’ve seen in some posts that @HenryCooper advocates of sticking to the proxies that shown the best performance for you, however due to dynamic nature of Instagram and the rapid changed they introduce, it can happen that something that has worked with you for months, maybe years to stop working all of the sudden.
So, by knowing that, in your experience, how important is it to search for new proxy providers continuously, and have couple of them as Plan B, even when you have found something that works for you? Do you think this is just me overthinking this, or is this something some of you veterans commonly do?
Thank you all for giving so much into this community!
Rule number one of business never put all eggs inside the same basket create plan A, B, C
This plans, don’t create regular plans, tweak method’s, think strategies, don’t create plans just for having backup’s
Start with one plan, think something outside of box, learn, fail, try again. And when you are comfortable with this plan, things are going good, sketch plan B and so on
Okay, so, do you think this structure would be a viable approach:
Tier one: 50% of accounts on the proxy provider that performs best for you
Tier two: 35% of accounts on the provider that performs second best
Tier three: 15% of accounts on the provider that performs third best
And then having 2 more providers that you test with couple of accounts (10 on each for example). Then if one of the tiers starts under performing or causes constant issues you replace them with some of the testing ones. Tiers can change according to their performance and can upgrade or downgrade accordingly.
I find one flaw in such design, How do you measure a proxy provider’s performance?
Issues might also be related to account’s trust, settings, creation, and Zuckerber’s mood.
So choose your main wisely, if two providers performs the same, I would go 40% 40%, and rest 20% for testing and backup…
Yep, those numbers were most definitely just approximations. Two equally used providers also seems like a reasonable way to go. I might go that route when I think about that.
As for the performance and quality measurement, I take couple of factors such as Action Blocks and Phone Verifications. Although, the occasional PVAs I get seem to be more related with the quality of the accounts rather than the proxy providers. But yeah, when I test the proxies I tend to test them with same type of accounts using the same settings and action scaling (which is crazy conservative), so if the accounts on one of the proxies go bad, it couldn’t be anything else except the proxies themselves.
Incorrect actions (all accounts start executing actions in the same timeframe) might get your proxy flagged. Same batch of accounts on the same proxy for a long time might also cause that.
Proxies are the easiest to blame, It’s not always the case. (Not that i’m blaming you for anything, I usually blame myself, but I think you got my point…)
Everything you say is sound as it can get. But I tend to think that the level of randomness and non-overlapping of the settings is pretty much as high as it can get (accounts get multiple days per week not working at all, they all have different nightmodes multiple times per day and work for total of 8-12 hours top with high pauses between operations, each account doesn’t do more than 600 actions daily and I never use more than 1 account on 1 proxy, not counting mobile of course). But then again, with so many parameters there is always the chance of human error, so your concerns are just on point.
Still, when all of these things are taken into consideration, and I split test two proxy providers where the proxies from one of the providers gets action block on all (or at least the majority) of the accounts and the accounts on the proxies by the other providers don’t get any, you can pretty much narrow the source of the issue.
I mean, this is how it worked for me for the time being. I’m not too experienced or anything, but plain narrowing of factors in split testing brings consistent results for me.
All I said was with thinking of mobile proxies, when it comes to DC, I would even go with higher number of providers, there are so many!! You can also try and probably find providers that match your geolocation which will allow you to login without challenge
When it comes to mobile, I’m just starting out with them, so I’m not competent to share my opinion. But I’m familiar with the way they work, so I will try to play it right by setting them to work on different intervals by using the nightmode option as efficiently as possible, so the accounts overlap on one IP address as little as possible.
Although I haven’t had problems with DC (again, testing, testing, testing, as some can most definitely cause issues), it seems that problems will certainly arise on pretty much all of them sooner or later, so that’s why I’m trying to adapt to the mobile proxies.
create plan A, B, C
