Hacker News new | past | comments | ask | show | jobs | submit | MattCutts's comments login

Hey Tim, your blog post mentioned the [recency microsoft word misspelled] query. The way I normally check if a word is valid is just do a query for the word, e.g. [recency]. In the top right we show the phrase "Results 1 - 10 of about 2,910,000 for recency" and "recency" will be a hyperlink to a dictionary definition if we recognize that as a valid dictionary word. Looking at the number of results is a good signal too.

Your "Red Room" query is hard in a couple ways. First, it looks like that root page used to have the words on the page: "The Red Room Doors open 6pm $18 Pre-Booked" And it's also tough because it looks like the name changed to the "2nd Degree Bar & Grill" at some point. The fact that you can type [red room] and get a suggestion for [red room st. lucia] is actually pretty helpful in my book because it leads you to the answer that the name changed.

Hey Matt -- thanks for investigating and replying, much appreciated.

I actually did a search on "recency" first, and Google didn't correct me and the first hits were a couple of dictionary entries, which is why I searched for the longer query. I don't know why, but I guess I figured there might be a blog post perhaps complaining about Word lacking common dictionary words and marking them as misspelled. In hindsight, including "recency" in the query was silly, but I'd prefer to get nothing back instead of Google assuming I want something else, which is the issue I was trying to highlight.

I agree that the suggestion of "red room st lucia" is a good one, and it helped me find what I was looking for! However the problem I was trying to show was Google's new approach at suggesting a first hit without search terms you've entered. As another example, I was trying to find my old (and embarrassing!) Geocities page recently, so I searched for [tim cederman geocities] and the first hit was my own blog.

I think the the results quality on Google has been the same or better over the last few months; one thing we have been looking at is helping less savvy users who might mistype a word or type extra words that they don't really need in their query. That can be a little more annoying for power users, but on the other hand the power users pick up tricks like "Use a '+' in front of a word to require Google to match that word."

Regarding the query [Linux asynchronous IO] returning older results, here's a tip. Above the search results click the "Show options" link to open up what we call "toolbelt" mode. From there, you can click to show only results from (say) the last year, or in a certain date range.

Toolbelt mode is really handy, e.g. if you search for a product, you can click "Show options" and then click the "Fewer shopping sites" link to get more reviews and manufacturer pages instead of comparison shopping sites.

Or the dangers of companies not allowing you to easily export/backup your data.

While this is true in principle, this is a bit of a Google talking point that has more PR cachet than practical implications.

Giving users the ability to make backups of their data certainly does not mean they will actually do so. How many Gmail users actually keep full backups of their email (using POP or IMAP or whatever)? And if Gmail had a case of massive data loss tomorrow, do you really think saying "well, you could've made a backup just fine" would have somehow made everything okay?

I don't deny that this sort of data loss is pretty awful, but "data liberation" is not the solution.

The reluctance of some people to back up their online data is a testament to the general dependability of online services. More screwups like this one, or like Magnolia recently, can only serve to scare folks from "the cloud", or teach them to back it up.


"Data Liberation", as in educated users, is the solution.

You back up your data because no cares as much about your data as you do (repeat...). It's like basic hygene. In the future, few people will be actual computer experts but everyone will know a few things. Backing up data will be one of them. A large enough density of people who understand this will mean those who will lack excuses - see main story!

I back up my Gmail by forwarding it to other providers (Yahoo, Comcast). If Gmail fails and loses all my data, it's in my comcast and yahoo account. If they both fail then ok Im screwed, but the Internet I hope would have to melt down if all three of these systems failed simultaneously.

I'm not sure how much is a tech feature or un-walled policy would solve the problem. I think it would fractionally minimize it. The current psychological barrier -- people not using a phone as a computer and its imperfections is profound.

Before, if a phone didn't work, it was one of a few things, usually the device or the carrier. Now, it's one of a dozen dozen or more products and services. People couldn't export before. They might be able to now.

A friend of mine once noted the subtle difference between how technology can protect you from failure, but a backup can help protect you from yourself.

It feels like we're moving towards a singularity. Expectations, responsibility and technology.

That doesn't apply to Danger/Sidekick. I had one 4years ago and synced regularly on PC and Mac.

One nice thing of having your data on the server instead of on-device is that any syncing can happen between the server and your desktop, without even the need for USB cables.

I'm happy to ask the right team at Google about this and point them to the post on hizook.com.

As happy as I am that you're Google's unofficial ombudsman, Matt, you're a big billion dollar company who many of us are intensely reliant on for our businesses. There needs to be a support channel other than "make a PR issue for Google and Matt Cutts will swoop by to save the day".

I had a big issue with AdWords earlier this month. Google made it all but impossible for me to reach anyone about it. Earlier this year, I had an issue with my bank denying an AdWords bill (to protect me against possible fraud). It took one minute on the phone to resolve.

You make a hundred times what my bank does off me. Why is human contact only possible if we speak the secret password into the telephone disguised as a shoe which is stored in the unused broom closet marked Beware of Leopard.

Though it may not be a complete answer, it is encouraging that people from Google will reply on public forums. You'd never see an Apple employee posting on a thread about someone whose iPhone app had been sitting in the pipeline for months, or a Paypal employee posting on a thread about someone whose account had been summarily frozen.

On the other hand, the Google employee who has answered here is Matt Cutts. I doubt many lesser-known employees at Google have the luxury of replying outside of official company forums and sites in such a manner.

From personal experience, finding an Apple employee who can even find out what is going on with an iPhone app waiting for review for months can be difficult if not impossible without connections inside the company. However, if you have a development-related question or a bug report you want looked at, there's plenty of Apple employees on Twitter that would be happy to answer such as Chris Espinosa (a manager on the Xcode team and Apple employee #8, @cdespinosa) and Michael Jurewitz (developer tools evangelist, @jurewitz).

FWIW, I've seen "lesser-known" Google employees get people in touch with the right teams.

I'm not saying the poster is necessarily faking, but how should we know that someone who chose the name MattCutts is the Matt Cutts he claims to be? Do we assume he's not an impersonator because we like his answer? This comment is the very first activity on that account.

I like the way you think! I tweeted to you to prove this account is really me: http://twitter.com/mattcutts/status/3603379916

And of course anybody who hacked your HN account couldn't possibly have hacked your Twitter account (assuming they are both really u, and u'r the Google Matt Cubbs). :=)

"You'd never see an Apple employee posting on a thread about someone whose iPhone app had been sitting in the pipeline for months"

No, but you might get a call from Phil Schiller.

Startup idea: hack customer service into a product that scales.

The difference is that Apple is just being Apple while they try to figure out how to handle the App Store. I'm not saying it is right, but there's a certain intelligence and style behind the way Apple ignores people.

Google, on the other hand, is just incompetent.

(And PayPal is, too, but they seem to be slowly learning.)

I disagree. Google is just being Google. They've always been a company that does very cool (and non-evil) projects, but doesn't really bend over backwards to listen to customers. I always get the impression that they're too busy working on their latest game-changing scheme to be bothered with little details like keeping customers happy.

Not bending over backwards for customers is the action of an incompetent company. Whether by choice or not.

under which definition Apple is not just being Apple but is incompetent since they are doing the same thing you point to Google as doing. You can't have it one way for Apple and another for Google.

I'd wager that he's offering to help not because Google is such a friendly and benevolent business entity, but because it's in the spirit of Hacker News (and ultimately, in Google's best interest to do so).

The credibility of Don't Be Evil is taking a lot of hits these days. Some of us were never really convinced.

I fail to see how this is "evil". The evil one worries about in context of a company like Google is information privacy.

Google and PayPal both fall into the category of not putting enough profits into customer support.

This is what most fear from market consolidation and extremely large companies. They squeeze more profits from doing things that a company without monopoly-like characteristics could not get away with. To me, when you cross this line, you have become evil.

fyi, I spent more than 2 hours last night getting my PayPal personal account unlocked. It was an awful web and phone customer support and terrible security practices.

It's great to see this discussion here and especially to have Matt Cutts weighing in.

It does seem like this is an endemic problem in Google and comes right from the top. Some startups focus on customer service and use it as a differentiator. Others focus on product and believe they can build their way out of support.

Google is and always has been in the second category. The mindset is from what I understand firmly held by Larry and Sergey and manifests itself through the whole company. Until they change, it seems unlikely the support teams will change.

Yes, this does seem to be a top-down problem and not a result of incompetence.

I'm guessing they won't change until they see their profits effected. As to PayPal, I have chosen other providers. So I only have to deal with them if some vendor only accepts PayPal and I really want to use that vendor.

As for Google, for now, I don't need to serve up ads. If and when I do, I will spend time to see if alternate ad brokers can give me equal or better returns. Google knows that their hold on the ad market stays as long as buyers and publishers do not move in sync. That is, its very hard for a publisher to use a different brokerage when all the buyers feel they need to use Google (and vice verse). This movement is probably best done by market segment. Specialty brokers ("we only do blogs", "we do sports very well", etc) should be the ones to disrupt this space. I guess this is already being done, but it certainly isn't happening fast enough. Ideas?

This comment is obviously an utter and complete waste of space, but I couldn't resist pointing out the irony of someone called "lucifer" failing to see how something is evil.

Irony? Denying that something is evil seems pretty typical for the referent of that symbol.

Actually, the root meaning of the name is 'Light', from which one can arrive at the notion of 'Sight' So what is really ironic is that any sentient would deny their inner lucifer and presume for itself an objective view of truth ...

Avoiding this discussion was why I danced around the name issue by pointing to its referent.

There is a support channel for appeals, and it sounds like hizook.com submitted an appeal via that channel earlier this week. My guess is that the Google team is already looking into it; I just wanted to let people know that I'd ask the relevant team from my side as well.

Given the hundreds of millions of users, tens of millions of webmasters, and many advertisers and publishers that Google interacts with, it would be difficult to have one-on-one conversations with all of them even if everyone at Google did nothing but try to provide that support. Historically, we try to tackle that challenge by reducing the need to talk to a person at all (e.g. our webmaster tools let people diagnose lots of issues without talking to a webmaster expert at Google). We try to pursue scalable communication methods such as forums and keeping an eye on the blogosphere too. I wouldn't claim that we're perfect--I'd be curious to hear more about the AdWords issue that you had--but our first instinct is to look for ways so that people don't need to telephone Google in the first place. I think a lot of companies do that; I can't remember the last time I needed to talk to Amazon or Facebook or Netflix on the phone, for example.

But I take your point, and while we do have support channels, I think it's good to keep an eye out for blog posts because that feedback can help us improve so that future people interacting with Google don't get as frustrated.

A part of that frustration has to be from that email that gets sent out. I haven't been on the receiving end, but I've read it in a couple of blog posts about this sort of thing, and honestly: that letter is awful.

It contains exactly one piece of "useful" information, the link to the disabled account FAQ. The rest is just simply terrible. It starts with a very serious accusation, intoning "significant risk" and "financial damage", but Google wasn't "compelled" or "forced" to act, it "decided" to, implying not that some mandatory threshold was reached but that someone sat down and worked it out and reluctantly agreed that this was the way to go, or worse, that sometimes similar situations might go the other way.

Immediately after accusing the person of being an evil bastard, it turns on the smarmy fake politeness. "Please understand" that we think you're an evil bastard. "Thank you" for understanding that we think you're an evil bastard. And thanks for "cooperating", even though you actually have no choice in the matter at all. If you have any questions, kindly fuck off and don't use the same method of communication we just used to reach you, here's a link to more standardized dehumanizing copy.

It reads like Google is breaking up with you while checking its phone, acting like you've done something horrible enough to merit no discussion on the matter, trying to act nice enough so you don't think it's a bitch, but making it clear it doesn't give a shit and really never did.

calambrac, I personally agree with you about the wording/tone of the current letter. I would like them to take a look at changing that, although no one will ever be delighted to get that letter.

That's true, it's always going to be a hard letter to get, and I'm sure that most of the time it's being sent to people who deserve to get it. I think a rewrite should simply dispense with the dire accusations, the fake politeness, and the implication that there's a significant human presence behind the individual decision. Don't waste people's time or patronize them, be direct about what's happening and acknowledge that there may be some kind of recourse if you want to work through the process. 'We're disabling your account because our algorithm told us to. We regret if this is an error, please see this page for more information if you feel this is the case.'

Yes, it's a nasty, no-easy-solution dilemma: Shugging off the spammers yet not mistreating the "false positive" decent customers that your algorithm bans.

What I find distressing (and would find even more so had I been the victim) is that Google refuses to to identify which part of TOS have been violated. Police and courts have to specify which parts of criminal code a defendant is accused of violating. Can we not get similar fairness from corporate America?

Are you afraid that such clarifications would give too much useful info to actual spammers?

Well, maybe a slight edge. But I might compare this to the dilemma that clarifying legal defendants their constitutional rights ("Miranda" in the U.S.) deprives law enforcement of some opportunities to tap some useful self-incrimination: It's part of the cost of valuing Constitutional rights. Would Google dare to offer its customers some Consumer Bill of Rights?

Google is in a contractual business relationship with its partners, not an hiring or governing relationship. The contract may suck, but does anyone sign up for AdSense because the contract is great?

I'm sure that most of the time it's being sent to people who deserve to get it

Bingo. It's optimized to stop spammers, not be inoffensive to false positives.

I love your characterization.

Except it's not just breaking up with you, it's breaking up with you -- and stealing your TV, your sofa, and your cat!

Not really. In Google's eyes (or Google's algorithm's eyes) you "earned" your Adsense credits in a way that was unfair to the advertisers. So basically Google's returning the TV, sofa and cat that you "stole" from advertisers in the Adsense network in the first place. Obviously the process by which Google decides what accounts are fraudulent is imperfect, so some of the time legitimate accounts will get flagged.

I do agree though that this is a big enough problem that Google should devote more time and effort to improving the systems they use to flag Adsense accounts, and certainly improve the letter and general customer support experience once your account has been flagged.

"So basically Google's returning the TV, sofa and cat that you "stole" from advertisers in the Adsense network in the first place."

I haven't heard either way but if you're an advertiser, does goog credit your account - "we believe these x number of clicks you paid for were fraudulent, so we're not charging/crediting you for them."


Here's the short version of my recent experience:

Half of my business comes from AdWords and I'm pretty happy with it. Actually, you could go substantially farther than that: I'm literally a case study for it, recommend it any chance I get, and despite what I'm about to tell you continue to recommend it.

One day, for no reason I could be certain of, AdWords just stopped serving my ads. Google's automated diagnostic said that you had no credit card on file for my account. That was contrary to reality, as you had already successfully billed me for $300 that week and about $12k total.

I respect that you would rather I use the self-help automated diagnostic, FAQ, and whatnot. But either the diagnostic was broken or your data was, and the FAQ didn't address "What to do if our diagnostic tells you untrue things". So I tried to contact support.

After digging through about 5 separate redirects to get to "contact an AdWords representative", I wrote up a detailed bug report (or as detailed as the 512 character limit would accept) and sent it in. As for what I got back, you can read the above blog post, but it was non-responsive and appears to be computer generated for me.

In the wake of the above blog post and further attempts to use the "scalable communication methods", I was contacted by an AdWords representative. This is now several days after I had first contacted Google. She successfully summarized my issue (which proves that she was not a computer -- an oddly comforting notion to me by this point) and said she would escalate it to a specialist to investigate.

That was a month ago. I still haven't heard back.

On the plus side, the symptom which was of most immediate concern to me (my sales dropping by half because my AdWords ads were not showing) recovered before I missed my entire busy season, and things are more or less normal for me now.

I'm glad that the ads are running again. Reading through your blog post, it seems entirely possible that the @welcome variable not being defined could cause an issue with conversion tracking (although I don't have any first-hand knowledge). I wonder if the AdWords rep worked to make sure that the issue got escalated/resolved, but was working under the impression that you didn't need to be contacted back afterwards?

Then that support rep needs to go on a very basic, low level training course, to find out the basic truth that "if you've done the work, you should at least make a token effort to get the credit for the work".

I never understand people who go to all sorts of lengths to get something done and then don't get back to you to let you know that it's been done. It always seems suspicious to me (though I like to remind my self not to ascribe to malice what can be perfectly well explained by incompetence).

I specifically requested contact after they had determined the cause of the issue, and $REP_NAME assured me that I would be contacted with results of the investigation. That contact was on August 5th, roughly a week after my initial bug report, five days after my blog post, and a day after the problem had begun to rectify itself (and I posted the blog update).

Despite $REP_NAME's assurance that the matter was being looked into and that I would be contacted again when the investigation was complete, I have not heard anything about this matter since August 5th.

I value my business relationship with Google. So much so, in fact, that I am increasing my ad spend -- you have the best solution for advertising for small businesses, bar none, warts and all. However, Google's recent performance has not caused me to think that Google values its business relationship with me.

The point in the above comment is that adwords customers with large accounts pay such large amounts each month because

When you do get to speak to someone (via above hidden shoephone method) it's still like speaking to a clumsy CIA agent. In fact I spent most of this morning doing that. It seems very clear that the googlers on the other end of the shoe phone aren't allowed to disclose much. They allude. They read out the relevant policies to hint at what might be going on. [me]'Has a trademark complaint been made?' [Googler]'All I can say is that your ads have been manually disabled at some point last week?' [me]'Would it help if I got the trademark owner on the phone?' {Trademark owner has already told us that they don't know what is going on. Google has been impossible to get on the phone and answer emails extremely vaguely} [googler] 'reads some policy that hints at a recent complaint' [me] 'Can you email me the procedure for trademark exception.' [googler] 'Google it'

challenge Try & find the form you need to have filled in order to get approval to advertise using someone's trademark.

We are talking about customers that have spent hundreds of thousands with google over time. Google allows itself to unceremoniously damage their business as if they are doing something shady. They then make it impossible to contact them and when they are contacted, well they make phone companies look good.

Answering blog posts is great. Writing your blog is great. It isn't instead of normal customer support though.

I'm so tired of big companies pointing to their huge userbase as a reason for bad customer support. I've heard it from Google, Skype, and HostGator.

I've also specifically tried to contact AdSense via the contact form, email, and the forum about a missing payment. Never heard a peep back, so I stopped running AdSense.

I'm happy to read the FAQs, search a knowledgebase, and communicate via email or a ticket system, because I know its not feasible to talk to all customers on the phone. But you can't just leave people in the lurch when you want to be their business partner.

Successful companies see customer service as an opportunity to right a wrong and impress their customers, not as a nuisance.

Matt, thanks for the explanation and acknowledging the fact that it is not a perfect (or even a good) system.

Lets say I was in the OP's position and I have legitimate reasons to believe that Google is at fault here and they banned my account by mistake - but after weeks of trying they didn't get back to me (I only received automated e-mail, which are vague at explaining what going on). Assuming that scenario, my best option to get someone noticed is to make a blog post and submit it on HN and/or e-mail you about it or hope that you will notice it here?

Surely, there are better way to handle this?

This is not the only case, there are literally 1000s of such stories for many years all over the blogoshphere and forums.

For instances like these, I would happily pay $x per minute to talk to a real person.

A high call charge would prevent random crazies from calling (e.g. "why isn't my website top of the search rankings!?!?"), but would at least give the chance for people for whom Google's service has financial implications the chance to talk to a real person who can get appropriate answers from within the company.

And if random crazies do end up calling it, then maybe that's a new revenue stream for Google?

Perhaps they should be responsible for lost revenue in cases where their incorrect assessment hurts the business. That would at least make it somewhat important to them to implement a working solution, scalable or not.

While I'm willing to pay $x for a call, if I'm in the right I want to be reimbursed for it (MS is actually quite good about this).

As has been pointed out here before, Google are essentially a customer in this situation, therefore they are under no real obligation to keep buying ad impressions from you, so I'm not sure 'lost revenue' liability would be particularly fair or enforceable.

But I agree that reimbursement for the cost of a call that results from their inability to deal with your problem through any other channels would be appropriate.

In the Adsense cases, the argument can be made (not sure I fully agree, but I don't use Adsense so I can't really speak specifically about it, but customers can be held accountable for interfering with business operations). In the Adwords cases (patio1, e.g.), Google is the vendor.

It seems to me that the way you deal with having too many users to economically support is to have a support route with a high barrier to entry. That's not to say that the route should be hidden; but, for example, it could cost money, which is refundable in the case where there's a genuine grievance.

Google doesn't have too many users to "economically" support. They are simply reaping higher profits by not scaling support along with revenue.

barrkel, I can speak to some of the thinking on the search result ranking side. I believe the feeling has been that asking people to pay for support creates worse incentives, because it turns something that people shouldn't have to do (contact Google for support) into a source of revenue.

It's a similar situation to pay-for-inclusion in the search index: a search engine should include as many pages/sites in its index as possible, and if someone can pay the search engine to include their site, then it creates a weird incentive for the search engine to not do a great job of crawling the web.

At least on the organic/ranking side, that's at least part of the reason why Google has never done pay-for-inclusion or pay-for support for webmasters.

it turns something that people shouldn't have to do (contact Google for support) into a source of revenue

That's why the model stipulates that [the money] is refundable in the case where there's a genuine grievance

Define genuine grievance?

Fully agree about perverse incentives in search result / ranking, but do not think it applies in nearly the same way when there's a business relationship, such as adwords etc.

Maybe someday your super sophisticated fraud detection software could be made to check and see if someone got slashdotted.

I understand this would involve knowing the content of billions of websites and figuring out which ones link to each other.

Who could possibly be expected to keep track of all that information?

So keep your pants on everyone. The scope of this problem is insurmountable especially since there are only thirty days in the month (or six) before you get paid for your clicks and impressions to discover fraud.

I can't tell if your comment is sarcastic, but I imagine Google, the masters of search, keep track of all that information already. And checking major aggregate news sites is far from "super sophisticated". I wouldn't be surprised if most false-positive frauds were attributed to sudden spikes attributed to slashdot, digg, or HN.

Support should scale with the size of your business.

Sorry, Matt. Yahoo has excellent customer service. Why can't Google do the same? They can, they just won't until they are forced to.

When this many people express frustration with their inability to talk to a person, you have a problem. Your methods aren't working. Sending an email isn't enough; people need feedback to be reassured someone is, in fact, helping them.

"There is a support channel for appeals,"

When I asked them to reinstate my account when I was blocked (a good couple years ago) for, what they told me, click-fraud, they took a couple days and said something in the line of "we looked into it and we were right, but we can't tell you what or how we identified it". In other words, people are punished and they can't even look at the evidence that bases the accusation.

That's really bad feedback.

And the balance was of about US$ 50.

The problem is that the support channel is a black hole. There's no way of knowing where your request is in the queue. Has it been reviewed? dismissed? do you need to be more patient? should you submit it again?

What about starting with giving out the average response time and queue length when submitting a request? That'd be a good first step...

How about using Twitter like Comcast does? They have about 10 reps on Twitter monitoring & resolving 1Ks of customer care issues each day.

Can Google not follow in it's path? Not everyone's blog post is going to get this much attention or attention similar posts have received.

Even if you had a "I need to get this resolved NOW" option that costs $100, I'd be delighted. I've had one false positive and I was never able to get it resolved and I never knew why it occurred. I'm not a huge blogger, I just run a few small unique sites that all adhere properly. I don't even know which site at the time was to blame.

There's a support channel for appeals? Not hardly.

As I mentioned in my anti-Google Checkout article on Slash7 (which you commented on), not only has Google attempted to steal money from me by closing my Checkout account, but my husband lost over $2,000 of Adsense money when his account was arbitrarily closed.

He's the author of Scriptaculous - and the ads were on the Scriptaculous site - a totally innocent open source contributor.

So, support money was effectively stolen from a popular open source project.

Of course he appealed.

And he never heard back.

The commenter above is right: One-off help for those who are able to make a lot of noise, and hurt Google's image, is not enough. It got me my Checkout account back, and it will probably get the OP's Adsense account back.

But my husband never made a public fuss, and so of course all his attempts to reach out to Google were completely ignored.

Until there's a system in place other than "be famous and make noise," nobody should trust Google with their money, in any way.

Anyone who's interested in more back story (and my recommendations for improving the Google experience), these are the two articles I wrote:



If you really mean that they closed your husband's account and walked away with money they owed you, then this seems like exactly what the legal system is made for. (Though I'd expect them to settle rather than actually go to court, unless they think you're lying outright.)

If your customers have to invoke the "legal system" to get around your incompetence, you've failed in every way that matters.

unfortunately, I do not see this as incompetence. Google has made a choice to not put "enough" of their profits in customer service.

That is the action of an incompetent company.

Well, they seem to have come out $2000 ahead on this one. Perhaps they ran the math and realized "the legal system" was intimidating enough that to deter most who would be forced to look to it for help.

To clarify: I was not in the least suggesting that the fact that it might be possible to get your money back from Google by suing them means that Google haven't failed. I was suggesting that it might be a more effective way of getting them to take notice than going through their (clearly rather useless) support channels and blogging about it.

He lives in Austria. Which legal system? And even if it was in the US, you'd be very hard-pressed - as a professional - to justify going through the time and effort required, even in small claims court, for $2k.

For Google, it is a win/win/win situation... until people holler about it.

If you actually expected them to fight it all the way, that might be true. I suspect, though, that "Google found guilty of stealing customer's money" would be an unpalatable enough headline that they'd settle pretty quickly once the threat was made.

But yeah, not being in a country where Google has enough presence to bring them under its jurisdiction could be a problem.

That dude who was on the HuffPost, who sued them for his $700 in missing Adsense money... they fought him in small claims court, even tho their representative was unable to prove a case, they appealed and are dragging it out.

You think they want that kind of precedent known?

Given the hundreds of millions of users, tens of millions of webmasters, and many advertisers and publishers that Google interacts with, it would be difficult to have one-on-one conversations with all of them even if everyone at Google did nothing but try to provide that support.

Not saying it is easy or free to get there, but the same arguments can mostly be said for Microsoft and they seem to be doing pretty OK.

I've had excellent support from them, 24/7, on the phone, including weekends, where I've been assigned one person and one person only to fix some critical issues which had to be dealt with before monday 7am.

I have nothing but praise for the Microsoft support team. However just trying to find non-automated support for my google-services seems impossible.

I do realize Microsoft and google has different business models and don't mean to come off hostile, but Google upping their customer-service standards a little bit wouldn't hurt.

Your last sentence is dead-on. Microsoft ascribes to the traditional enterprise-y business model. The products work okay but often need support, and if you pay enough you get great support (and if you don't pay enough you get passable support). This has made them (and many other companies) a ton of money, but Google is absolutely killing all competition by following a very different path. It's hard to say how much of Google's success is due to their aversion to acting like a regular enterprise company, but I think it's safe to say they won't be drastically changing tack any time soon. The stories like this of people getting screwed are unfortunate - it's collateral damage.

Collateral damage? Well, then it appears Google's experiencing some blow back, and deservedly so. It shouldn't matter what trails your blazing, you can't leave your customers burning in the blaze behind you.

PayPal also grew this way, with collateral damage - until they were slapped with several class action lawsuits, which went through, and (I believe?) they ended up being at least partially classified as a bank. Or threatened with being classified as a bank. Something to that order.

In any event, they really had to clean up their act.

That's going to happen with Google, too, I would expect, and there's no way they are going to be surprised.

If you ask me, it's no different than a chemical manufacturing company deciding that it is more cost-effective to pay the EPA fines than to clean up their act. It's about pure numbers.

That is a point of view I will never condone.

"Given the hundreds of millions of users" - I doubt that anyone who is using Google for search purposes will need to contact customer support.

Tens of millions of webmasters - I would suggest that only a fraction of them would need to call customer support, same with the publishers and advertisers.

Banks also have millions of customers, just when the call center is busy, they are put on hold. I think the best investment that Google cam make right now is not to sacrifice customer service in order to get on with throwing stuff on the wall and see what sticks. Instead, just hire some recent grads to deal with people's issues. How much would that cost, lets say ten grads, 100k a month each, that's 1 million, a drop in the water as far as Google is concerned. The return on this investment on the other hand would be so great as people can then continue loving Google rather than suggest that Google is being evil.

"I doubt that anyone who is using Google for search purposes will need to contact customer support."

You'd be surprised. There's still some people who think that when they type in a query, people at Google manually decide which results to show. For every result on the page. For every query. Hundreds of millions of times a day.

A lot of those searchers do a search for "John Smith" and then want to talk about why their school friend John Smith wasn't #1 instead of that other John Smith.

That's doesn't negate your point; I'm just saying that orders of magnitude more people want to contact Google than there are Google employees, and many of those people want to have prolonged discussions with Google. So the challenge is to find scalable ways to communicate with those people.

I have heard answers of that nature coming from Google before. I think they are avoiding the question. Searchers? Fine. They are using a free, take it or leave it service. Almost anyone would understand that they will be shuffled in to some sort of 'scalable communication method.' Webmasters in your index? You have no commercial relationship with them. They may rely on Google. But there is still no real commercial relationship.

Adwords/Adsense customers/vendors? That is a different story. As I have mentioned several times on this topic, I have encountered shocking responses from Google regarding large ($100k+) accounts.

Google does not have that many customers & vendors.

Why not just charge $80/hour for support, and refund it for when people call in to fix false positives in the fraud detector? Seems like a win/win - make money off scammers trying to cheat you, and give people an easy way to re-enable their account.

I cannot stress how important this is! Matt, if you can give one thing to the AdSense team, let this be it. $100 a call, $80 an hour, whatever it may be. If I'm out $1000, I'll pay $100 to try and get it back (especially if I stand to lose $100 within a week of waiting). Snail customer support when it concerns accounts that hold money is not good.

Google makes billions even in a slow economy so it should be able to afford customer service for at least its paying customers.

There are tens of millions of Indians out there eager for such a job. This is a better opportunity for them than "outsourcing SEO services" where they mess up the Google index by low quality and high quantity "link building".

I couldn't agree more. Much smaller companies with much smaller capital have options to contact for support issues in a person to person basis. With Google Adwords, it is virtually non-existant. Considering the fact that you have to either really screw up or get a lot of attention to yourself for Google to get to you, without that you are out of luck.

Automated response is not a solution to all problems.

Here's my thinking:

1. Google knows how to scale computers, and perhaps engineering teams, but not people in general.

2. "You make a hundred times what my bank does off me" - Exactly, and hiring pesky call center employees, who would have to be good on top of it, would ruin Google's profit/stock price.

We're all desperate, depending on practically the only ad game in town, so we bark but have no bite. A higher-end (aka non-cpc|cpm) revenue model is the only way out, I'm afraid.

I don't agree that costs throughout the system need to rise. If people want (or need) a higher level of support, they should be able to pay for it, preferably on a pay as you go model with refunds for genuine grievances.

there are other ad networks. uh, look at my other comments.

Personally, I think you are right that 'cost per click' and 'cost per impression' are both easy to game. The auction model, I think, might mean the beginning of a better way. Essentially, advertisers bid on what they are willing to pay per day. (now, personally, I think project wonderful's implementation is slightly flawed, but it is moving in the correct direction.)

"Beware the Leopard" is only written on disused lavatory stalls.

Hi Matt,

No offense, but if it takes posting your gripes on HN to get customer support then something is very much broken.

The simple fact that a multiple billion dollar company should depend on outside channels for its support cues is very sad imho.

Maybe, instead of passing on messages like these pass the general message that Google tech support as perceived from the outside absolutely sucks.

I'm not exactly a newbie on the adsense/adwords front, have spent lots of money on adwords before giving up because of inactivity on googles part dealing with click fraud, still using adsense because of a lack of alternatives.

We're talking about a website that normally gets 100 page views per day. Are you saying that Google should hire thousands of full-time employees to service customers who each only account for ten or fifteen bucks worth of business a year?

If one percent of them call once in a year, why not?

I can call a toll-free number about the 1 dollar pack of gum I bought. I don't get stellar interaction but it is still more than what we are talking about.

If you can not make it work for sites that have 100 page views per day then you should not accept them. Once you accept them you have to support them.

You can get just as bad with a $100,000 account.

Flip side: HN apparently expects that Apple should have enough people on staff to provide timely, professional support to developers whose apps might never sell as many as a hundred copies total.

What about the countless other people who can't cause a big PR fuss and therefore don't attract your attention? I'm not saying they're all innocent of ToS breaches, but neither are they all guilty, and Matt doesn't scale to look at each of these situations.

The Google search results are constantly being updated with new data and new algos because Google knows that to keep the SERPs relevant, they have to keep questioning their approach and improving on it. Why isn't the same healthy self-criticism of the AdSense termination process being portrayed in public? Every week we hear the same exact story, and it's now a regular feature in the headlines.

I guess you would make more webmasters happy besides yourself by making Google set up a team that deals with such issues where webmasters don't have to raise hell first to get noticed.

Maybe Google should hire the 10k "permanent temps" again it spit out a few months ago during the "financial crisis" panic.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact