I think that you might want to reevalute e-hawk. The price is really 
competitive: http://www.e-hawk.net/pricing/

That said prefiltering out known-by-you bad ones is always smart.

Tom



On Feb 21, 2015, at 6:35 AM, Gil Bahat <g...@magisto.com> wrote:

> From Magisto's perspective, having publicly announced reaching 50 million 
> users 2.5 months ago, such costs can rack up very, very quickly to tens of 
> thousands of dollars per month. This applies not just to e-hawk but also to 
> akismet, to email verification/validation services and to transaction fraud 
> services. at least the latter can be directly linked to revenue...
> 
> At minimum to make this economically viable, we will have to wrap any such 
> service with our own rudimentary implementation to filter out / downrank the 
> blatantly bad stuff.
> I didn't do this research for signup fraud yet, but for e-mail address 
> verification there is little point in getting an inflated bill from your 
> vendor for doing something which you can do in-house cheaply and efficiently, 
> like maintain a DEA blacklist, good domain whitelist and do some parked 
> domain detection. And then let the vendor handle the more complex remaining 
> stuff.
> 
> Gil
> 
> 
> On Sat, Feb 21, 2015 at 12:13 PM, Paul Kincaid-Smith <p...@sendgrid.com> 
> wrote:
> I'm glad you found the UGC blog post useful, Gil.
> 
> To make sure I understand your comment on pricing, were you referring to 
> E-Hawk's pricing or the costs to integrate other "blocklists and bot 
> detection mechanisms?" 
> 
> What orders of magnitude are you seeing for signups on your service? Seems 
> like an interesting problem to solve.
> 
> On Thu, Feb 19, 2015 at 2:00 AM, Gil Bahat <g...@magisto.com> wrote:
> Excellent stuff, comprehensive and yet very straightforward. e-hawk in 
> particular seems a great alternative to rolling something on our own and 
> starting to evaluate various blocklists and bot detection mechanisms. can get 
> very very expensive though at our signup rate.
> I find the idea to apply akismet to outgoing UGC a great one, because 
> 'standard' outbound spam filters don't aim UGC sharing systems, rather they 
> aim at ISPs, ESPs and big webmail providers.
> 
> Thanks a lot, extremely useful.
> 
> Gil Bahat,
> DevOps/Postmaster,
> Magisto Ltd.
> 
> On Thu, Feb 19, 2015 at 11:20 AM, Paul Kincaid-Smith <p...@sendgrid.com> 
> wrote:
> Hello Gil,
> 
> I am the postmaster for magisto, an app centered around user generated 
> content (UGC). we enjoy some popularity, and with popularity comes abuse. 
> There are users who utilize magisto to generate content to be used for 
> spamvertisement and/or other unsavory content. they will then "invite" users 
> to see this content, in an unsolicited fashion, using the built-in content 
> invite mechanism.
> <snip, snip>
> in short - recommendations are most welcome, reference to best practices or 
> case studies from applicable examples.
> 
> Here is a blog post that consolidates my Delivery and Compliance teams' tips 
> for safely emailing user generated content:
> https://sendgrid.com/blog/how-to-safely-email-user-generated-content/ 
> I hope it's helpful to you.
> 
> --Paul
> 
> 
> 
> 
> _______________________________________________
> mailop mailing list
> mailop@mailop.org
> http://chilli.nosignal.org/mailman/listinfo/mailop

_______________________________________________
mailop mailing list
mailop@mailop.org
http://chilli.nosignal.org/mailman/listinfo/mailop

Reply via email to