How do you deal with spam / malicious accounts as you grow?
Hey everyone, as I've been getting more users to my tool, I have been noticing an increase in spam accounts or accounts created to use for common scams (especially since its around the Instagram space where a lot of these exist). Most of these are against the TOS of the site as well.
I was curious how people approached these types of accounts on their platforms. Most of them are free accounts, although I've even had some paying customers oddly enough who strictly use the tool for spam / scams. Currently my approach has been to leave it if it wasn't too bad, but if it seemed extra scammy / spammy, I have been removing the account and refunding payments if applicable. Would love some insight from others who may have experienced this as well!
Early on, we built in a facility in our SaaS app to easily block or disable accounts that we thought looked suspicious.
From our 'God mode' dashboard that the founders can use, there are a couple of options to instantly disable a user account, and/or send them an email explaining why (we have a couple of default email templates that we can send).
We routinely get competitors, or bug bounty hunters log in and poke around the system. If they are being problematic (constantly trying to access restricted content etc.), then we hit the 'ban' button to quickly disable them and send them an email (and in app message) explaining why.
Previously this was a process that took up to 15 minutes to do, as we had to manually go to the database to disable their account, manually send an email etc, but now we can do it in about 3 seconds.
I would suggest building a dashboard for your app that only company employees can access, and from here, have a facility to automatically disable user accounts or delete inappropriate content etc.
Also, since we implemented a sign up validation for non corporate email addresses (i.e. Hotmail, Gmail, Yahoo addresses rather than mycompany.com domain addresses), we've seen the level of abuse and nefarious activity go down. We use the ZeroBounce API to detect whether the email address is a free or throwaway account, and act accordingly.
Really helpful advice, thank you! Right now I do most of it dealing with the DB directly, but I think building out a quick Admin portal would go a long way. I also haven't heard of the ZeroBound API so I'll look into that as well.
I'd be careful, trying to police stuff like that actively can often times become a full time job, and it's a pretty heartbreaking gig as more and more time can be invested in trying to stop those things, and it takes away from your legitimate users and positive growth of your platform.
Does IG reach out if they notice sketchy activity from your app, or are you running the risk of just being shut down out of no where if you don't mitigate it yourself?
If things get too out of control on your free plans, I'd say you may want to venture into offering a paid trial instead of a free plan. The tax on new users just kicking the tires will usually weed out a TON of the riff raff, and also leave you with warmer leads since somebody is willing to toss you a few bucks to check out what you're doing (likelihood to convert goes way up).
Thanks for the feedback! I think you're right in that it can take up a lot of time, I suppose my only worry is not allowing the platform to be used for illegal stuff, but I suppose that's where the TOS comes in to a degree. Also, Instagram has been known to actually block certain URLs from the bio if its the source of enough spam, so part of the reason for the policing is also on that end. From some quick investigation, I think it's really only 5% or so of users who are abusing it in some way, so it might not be a big deal at this point. Appreciate the insight!
All good... after nearly 10 years of running a social network with a ton of UGC, I know how some of this can be. In the early days where I felt like my livelihood was being attacked in the form of Adsense violations I tumbled down the dirty road of attempting to moderate any and everything I could, pre-emptively. At it's worse, were reviewing a stream of images that were being uploaded to help catch things sooner than later. It's no way to live.
You probably are already doing this, but I would keep a list of the markers on the accounts that you do run into that are spam/scammers to help build up a persona that could potentially be used to identify bad accounts. Often times we are reactionary and just remove accounts w/o actually recordings things that we can leverage in the future.
Best of luck with things!
Definitely, I can see how easy that would be from what I already found myself doing already. Great idea about building a persona as well, I think that could help a lot and is something I could setup pretty quickly I think. Thanks again for the advice!