> I'm curious [...] why Google doesn’t temporarily disable accounts so impacted until a human reviews activity.
Because Google doesn't have humans reviewing anything unless there's a direct link to marginal revenue/cost avoidance attached to that interaction that can be priced in. Their business model is to achieve scale through automation and machine learning; which means not doing things that would require manual intervention unless absolutely required.
Explicitly, this means that for free services like Gmail, humans aren't involved. Ever. Try getting support for a Google product and you'll see what I mean -- there's not even a phone number to call or an e-mail address unless it's a paid product (and even then, they've got a less-than-stellar reputation for support of paying customers).
So I've heard, that it's difficult to get a human involved if you have a problem with Google's free products. The article says "Eventually, with the help of Google’s customer support and some ex-colleagues who still work at Google, Bob was able to get his account back." For the average person who doesn't have ex-colleagues who still work at Google, or who's name isn't Linus Torvalds, it will be far more difficult.
I signed up for AdWords a long time ago and created a test campaign just to see what it looks like, but never completed the process, so I didn't have a payment method entered.
Fast forward till start of this year, I created an AdMob account to put ads in one iOS game I have, and apparently as soon as I entered my credit card info, the old AdWords campaign started running and was taking $200/day. I noticed after two days and immediately stopped it and contacted support. They quickly got back to me, a person called me and explained that the issue was an error from their side, and they refunded me the amount. It took three weeks for the refund to occur, and during that time the guy called me at least once per week to keep me in the loop. I was surprised with this after all I've heard about Google's support. But I guess that's just one data point in the pool.
AdWords is "different" because if your account has an issue, they aren't getting lots of your money. So their support is actually pretty good with AdWords. The rest of their services, the support is pretty bad (such as Google Apps), and their free services may as well have zero support.
My anecdote: I was brought in to help with an AdWords campaign for an ecommerce site. I did some keyword research, ran some small campaigns to get some CTR and conversion metrics, then kicked off a bigger campain, pruning keywords as needed to ensure CTR stayed up and we didn't get a low CTR slap on the account which results in CPC skyrocketing. It was chugging along at a pretty good ROI. At that point I turned it over to the person who is named as the contact on the account with instructions on how to keep it running. A couple days later they got a call from someone at Adwords offering to "help", and they recommended turning on display ads. A few days of that and the CTR had plummeted, CPC prices had skyrocketed, and ROI tanked and the account was ruined - couldn't get a decent CPC due to the CTR hole that had been dug.
Adwords is the only Google product to have dedicated support staff… and they're working on commission. Of course they're going to haunt you until the end of the world.
Yea, I've had a google voice number that I have used sporadically for my business for a few years and suddenly my account was just banned without any obvious reason. It was obvious to me that it was some bot gone wild on their end and was clearly a mistake. I tried contacting Google about it for months with no response. It was a number I actually needed for something important but was never able to recover it. I would have happily paid a monthly fee for the service if it meant I could at least get a response from a support email (ironically there's no number for google voice support). I even tried pleading with my adwords manager where I was spending a ton of money every month and she said she didn't known how to help.
This is why I won't rely on Google Voice for anything important - if they won't make it a paid service with some additional features, that means it's a cost center that could go away at any time.
Heck, I'd happily pay for Google Voice with the addition of fax capabilities (even if it was receive-only), particularly since my online faxing service was bought out by eFax (feh) and at the next renewal will jack up from $15/year to $85/year (bye!). I'd happily pay multiples of how much I was previously paying to be an actual /customer/ who gets that kind of services.
This right here is the quicksand you stand on with Google's free products. Starts out pretty innocuous. Hey, check out this cool new free phone product called Google Voice. You check it out, it's cool, don't really know how it might be useful but you play around with it and maybe use it for a side project or as a toy. Eventually, maybe years later, you find it may be very useful to your business. Great! You roll it out, slowly more and more things depend on it as you incorporate it more deeply into your services. Then suddenly, like a turkey on Thanksgiving Day, you're permanently cut off from your account without warning and without recourse. Now you're in a state worse off than had you never signed up for the service to begin with.
> So I've heard, that it's difficult to get a human involved if you have a problem with Google's free products.
Nobody claims it's difficult; it's impossible via official channels. You have to make a scene on Twitter, so obviously if you value your time and reputation you just switch providers instead of being "one of those people".
This is my #1 biggest gripe with Google as a company. if you have a serious problem that needs immediate attention... oh well! Too bad. You have to personally know someone that works at google, or post a really long article on HN to get whatever it is fixed.
You're a product, not the customer. How well do we treat most livestock?
I recently realized Windows 10 is going the same way and unfortunately it's much harder to not use your OS than Google or FB, and impacts your ability to use other purchased software.
> it's much harder to not use your OS than Google or FB
I think the exact opposite. Using Linux instead of windows 10 is easy. But not using any google service can be pretty difficult. Good news, there are some initiative to "ungoogle" the internet, such as Framasoft.
What particular google services do you rely on that can't be replaced? I'm using fastmail for mail & calendar services, duckduckgo for search, firefox instead of chrome, sync rather than google drive.
I've had terrible experiences with duckduckgo. I really want to switch, but I've never been able to justify the costs. Every now and then I give it a shot but the results from DuckDuckGo just aren't there. Sometimes they aren't even in the ballpark of what I'm looking for, while Google nails it - what I want specifically is in the top 3, and the whole rest of the first page is related.
I don't know if this is because Google and DuckDuckGo require different phrasing of terms to do well and I'm just highly adapted to Google, or if it's something I can't help.
Doesn't that just redirect you to Google's search results (encrypted.google.com), and thus makes the point of using DuckDuckGo in the first place unnecessary?
I have a domain and a paid-for email hosting service that I use for my personal email. When I tell people this I often get looks of incredulity: Pay money?! Why not just use google? I must be mad!
My response is that I do not want an advertising company to handle my most personal correspondence. In my mind it really is that simple.
>they've got a less-than-stellar reputation for support of paying customers
I've actually had surprisingly good support from the $150/month plan for their cloud products. They get back to me quickly and give good advice (with custom code samples when needed).
Have had multiple issues with Google Apps and GCloud over the years and very promptly received phone calls from support (that I didn't expect). I live in Fiji, and have a tiny small business account (~$30/m)
Like how I have more than enough information to get one of my old accounts back but their Automated Response Forms reject everything. Even the moderators on the gmail support forums rejected my claims as well and pretty much said "deal with it".
Yeah, I can't get into two of my Gmail accounts even though I have all the information it asks for. It simply doesn't work. I just want to delete them too (not even sure if that is possible), because I refuse to use Google for anything now days.
Same happens to me. I have the original creation email, my email is (was?) the backup email originally created. I know the date and time of creation of the account, and they still reject it.
If you don't have a recovery email address actively assigned, then you should be locked out.
That's what adding a recovery email is for: to deliberately weaken your account's security, in exchange for convenience.
If you disable recovery, then -- no surprise -- you won't be able to access your account if you lose access.
This post may seem unkind, but keep in mind that some people deliberately do not add recovery methods, in order to prevent secondary-takeover access; and that the above poster's request would fundamentally drastically weaken the security of those users.
> Because Google doesn't have humans reviewing anything unless there's a direct link to marginal revenue/cost avoidance attached to that interaction that can be priced in.
Google would obviously start losing money though if people perceived Gmail as easy to hack.
Right; so instead they've built their security model around assumptions that other companies' (namely telcos) account processes are secure. That way Google can say "it's not our fault; we can't do anything about it!"
It's called externalizing your costs, and Google is exceptionally good at it.
No one is perfect, but IMO when it comes to security, Google goes to great lengths to make stuff in the web secure, probably the most from all the bigcorps out there.
They are very aggressive about fixing TLS/SSL-related issues (finding and fixing vulnerabilities, deprecating old ciphers, promoting new stuff like cert pinning and HSTS, lowering SERPs in Google for non-https websites; they've also put in place means to detect and report fraudulent HTTPS certs, and many more).
Many of the top security researchers work for Google, and many of them even not on things directly related for Google's daily businesses.
If I'd have to say about some bigcorp that they don't care about security, Google would be the last on this list.
Google customer support is totally another pair of shoes though.
> Right; so instead they've built their security model around assumptions that other companies' (namely telcos) account processes are secure. That way Google can say "it's not our fault; we can't do anything about it!"
No, they've built their model on the fact that for the vast majority of their users this will never be an issue. Maybe a celebrity, maybe an important target but the average user, never going to matter.
No, they've built their model on the fact that for the vast majority of their users this will never be an issue. Maybe a celebrity, maybe an important target but the average user, never going to be an issue.
Email is the gateway to almost everything else. It's used for account recovery as well as (wrongly) passwords and other sensitive information. Google Drive and docs would also be compromised.
It's actually a bigger issue that you would think.
I am not doubting that. I am just saying that the vast majority of people that use gmail will never have a case where someone wants what they have so much to go to the effort to social engineer their wireless company. Ntim but to mention that there is a bigger risk of getting your house broken into or a host of other things or identity theft. Plus this requires certain bad behavior and failure as well on the part of the phone company. [1] It's not clear how easy this would be for a normally determined hacker to pull it off.
This is why last weekend I moved to FastMail. I've filed two support tickets since, and both were responded to in an hour or two. For the trivial cost of half a Netflix subscription, the most vital thing I have on the Internet is supported by real people.
It's hard to justify Gmail these days other than the frustration of migrating off of it.
As a side perk, Australia has no equivalent to a National Security Letter, and FastMail is able to hence notify me of any government requests for access to my data.
> As a side perk, Australia has no equivalent to a National Security Letter, and FastMail is able to hence notify me of any government requests for access to my data.
Unfortunately we've got something pretty close: ASIO can designate any of its activities to be part of a Special Intelligence Operation at will, which makes any disclosure of the operation or its components into a crime (10 years jail if it 'endangers' the operation).
Interesting. Can other countries invoke it? Because I know the US can ask the Australian government for assistance, but I wonder if that assistance can qualify for being an SIO.
By all indications, use of this provision is extremely rare, nothing like the number of NSLs issued in the US. So it would be a big ask, but it could happen if the consequences of wide disclosure were 'extremely grave'.
For context, one of the motivating leaks was when there was a planned 5am counter terror raid, and someone in the state police leaked the who/what/where to a major newspaper, who had papers coming off the presses at 3am about people who hadn't been raided yet.
Yeah, I procrastinated on this for years. What I did was:
- Created an email address at my own domain and forwarded it to Gmail.
- For the last year or so, started using it as my email address and changing accounts to use it as it came up.
- Signed up with FastMail and set up my domain's MX records to point to it.
- Still checking Gmail for stragglers to change over to my new address.
- Eventually will set Gmail addresses to forward, but didn't want to rush that so I remember to just make stuff not go to Gmail.
The hardest thing is getting used to folders again vs. labels. But FastMail has everything like filters and aliases and two-factor auth to an extent that... honestly makes Gmail look kinda basic.
The advanced features of Gmail were the most compelling reason to switch to it, but FastMail takes them a lot further.
Eventually will set Gmail addresses to forward, but didn't want to rush that so I remember to just make stuff not go to Gmail.
When I did this (from @gmail to @mydoman on google apps), I set up a filter to star/label anything forwarded from the old address, so I'd see an update was needed at the sending service/person. Perhaps the same could be done with FastMail.
It definitely can, but I feel that if I make it easy to automatically handle email on my old address at my new address, I'll get lazy about actually changing it. So I'll wait a bit to do that until most of my email is properly changed over.
I'm in a similar boat. I've used Gmail since its first year in operation. I started transitioning off it earlier this year, keeping it running as I've slowly migrated the vast number of web accounts that use my Gmail address as an ID, to use the new email address. With the benefit of hindsight I see that I should have registered my own domain and used that for all web registrations. Oh well. The final step will be to migrate my archive. I hope to be done by Christmas at which point I will strip all details from my account and then delete it.
For some people, email is not a big part of their lives, and it's probably fine to have a free service. But I'd argue almost anyone for whom email is a daily necessity, you should be paying for a certain quality of service. I mean, your email is the core of your online identity, and if it vanishes, such as the dozens of stories we've heard of Google terminating people's account without warning, it can be devastating to put everything back together.
> For some people, email is not a big part of their lives
Email is a big part of everyone's lives, who made an online account somewhere. Not to mention if they paired real identity and/or money to those accounts.
Search seems adequate for my needs. And it works great with any IMAP email client.
Actually the strangest thing I find amazing, is FastMail's endless scrolling being basically magical. I can scroll through over 2500 messages in a folder of mine, top to bottom, instantly. It's not paginating, it's not locking up at the bottom of the first 100 to stop and load more. It :just works: in their UI.
Not if we can't get past the he-said-she-said politics. People are voting for personalities over issues more than ever, and I don't see that trend reversing.
Dwayne Elizondo Mountain Dew Herbert Camacho 2020.
I get responses to my Google Apps support questions pretty quickly and I'm just using this for my personal family accounts, so I'm not paying them tons of money. I don't think it's really fair to say they should provide support for their free tier which hosts mail for a billion people and compare it to a service that hosts, what, maybe a million?
The problem is that the burden of responsibility Google carries for those accounts is huge. The amount of money, private information, treasured photos go over their systems... with no support?
You know, you can see articles where people report social engineering attacks on Amazon customer service and extract a great deal of information from them.
Having a human involved is not necessarily a solution, can be another attack vector.
It's a great counterpoint because with a large staff you can never solve the human element. It's also corollary to the hacking approaches we see regularly employed: Send an exploit payload via email to everyone in the company and somebody is bound to open it. In fact in the security space they've moved away from even saying you can protect the front door and are more focused on detection and correction once an intrusion occurs.
Yeah, maybe human review is not the most scalable solution; if data analysis shows that certain patterns of behaviour are highly predictive of an account takeover, there are almost certainly product solutions for them.
I guess the real question is what the data actually show
How about "we'll review your request for $20". For recovery of email accounts and such Google could absolutely make a profit from that I feel and enable people to recover their accounts; they can surely scale a support system on that sort of funding.
Sure I can see problems with that. My initial feeling is people will thing that such support is a scam, but that at least shifts the position of Google from "we can do such support" to "we don't want people bad mouthing us so we're going to refuse to do that support even if it were prima facie profitable".
Interestingly, one of the features announced for the new Google Pixel phone is some kind of free live support, though they didn't say which Google products are covered. It is an expensive phone but if the support includes help with other consumer cloud products it might be worth the price of admission.
If I permanently lost access to my facebook account I'd probably be mildly relieved that I'm finally done with it. My google account though, I'd be in panic mode.
Because Google doesn't have humans reviewing anything unless there's a direct link to marginal revenue/cost avoidance attached to that interaction that can be priced in. Their business model is to achieve scale through automation and machine learning; which means not doing things that would require manual intervention unless absolutely required.
Explicitly, this means that for free services like Gmail, humans aren't involved. Ever. Try getting support for a Google product and you'll see what I mean -- there's not even a phone number to call or an e-mail address unless it's a paid product (and even then, they've got a less-than-stellar reputation for support of paying customers).