I created a fake business and bought it an online reputation | Hacker News Reader

I created a fake business and bought it an online reputation

fusion.net

389 pointsjianshen168 comments

If you're a street musician setting up to play with an open instrument case, it's pretty important to "seed the pot" by throwing a few dollars and some change into the case. Not because you make any money from your own "donation", but because it shows passers-by what the arrangement is, and adds an element of social pressure ("Somebody donated, so I should too!").

I can see this as similar. There's something pathetic about being the first Twitter follower, or the 6th "Like", and something scary, particularly for a local business, about vouching for something you're not really 100% sure on. Writing the 4th 5 star review is a lot easier than the 1st one, and it's easy, and more anonymous, to be the 45th Like.

I don't support these services, and think we need good heuristics to stop them, but I can totally see why even an early influx of a few fake followers and reviews would "seed the pot", and be worth way more to a business's eventual health than $20-40.

Homeless do the opposite of the musician really, throw a coin/bill into their glass and they will take it out and save it somewhere else.

"Nobody has donated, have some compassion" works for some.

They often (usually?) will leave some small change though.

Panhandlers are common targets of theft, which may explain stashing most of what they get in various places less exposed.

Yeah, there are a lot of factors into how much you leave in the pot. Grandparent's post was just one of the pressures to leave more money, though one that most people aren't aware of.

Yep. You don't want to be the guy with $100 in a cup held out in front of you. Remember, it's almost as easy to dip a bill out as it is to dip one in.

I like when they pick out the pennies from my donations and throw them on the ground. I then pick them up, because the pennies are valuable to me.

You have a home to stash the pennies in to save them up. Someone who lives on the street can't carry around pounds of pennies around until they have thousands of them to exchange for larger currency. Pennies might have value long-term, but they are worthless if you can't stash them in bulk.

Disclaimer: Canadian, we abolished pennies because they are worthless and a pain in the ass. ;)

I tried this with a blog I used to run. Back when RSS subsriber count actually mattered, I faked the subscribe button to show that I had 15,984 subscribers instead of 134.

I would get just 1 subscriber every day earlier. After the change, I started getting 3 a day with the same traffic.

When I was working on developing a site, before it had its own domain setup, I used FB like code for another domain just so it would work. It was a popular domain with tens of thousands of likes. I didn't change it right away while I got the site up, one of the early testers was super impressed how many likes my site had though. Despite being connected to a completely different domain. It was social proof that was never actually checked. I updated to my own domain and nobody has ever cared since, sadly.

I used to go to a cafe where they often had a $20 bill in the tip jar near the cash register. I always felt like that bill hadn't been left by a customer...

They'll exchange smaller bills for bigger bills if they need them, or if the tip jar gets too full.

source: worked in an ice cream shop

But you also could do an extra effort to impress a small number of customers and ask for reviews personally. Buy fake reviews is not the only option here.

It's the same thing as posting a "tear off this phone number" page on a bulletin board. You always tear off a tab or two when you first post it, to make it look like people have shown interest.

I run studies. Guilty as charged. The key is to remember to how many you tear, and to always tear the same one off (5th from the right for me), so you know which poster locations are getting actual traffic and merit re-posting later.

That must depend on what you are advertising.

Looking for cheap appartmnts near a university, I would always go for the ones that didn't have lots of numbers taken, as it meant there was more chance I would be first to get it.

which site do u use for your search?

this was back n the days of pen and paper ads on a noticeboard

Excellent observation regarding the street musician (busker) faring better with a little grease visible to potential audience / donors!

I'd like to point out that the music industry has long been a bastion of other techniques as well, some more underhanded than others, depending on one's moral compass. The notion of Payola is still alive and well, and has changed with the new music and marketing avenues. I'm not in a position to name specifics, but there's word of payment for review / features, which seems the most blatant. Personally I can attest to navigating all sorts of sketchy "Promote through us!" type operations that may or may not be doing legit advertising...

Getting a network of Twitter or Soundcloud followers - or appearing to have one - is a pretty standard objective, which I guess why there exist services or short-cuts to those ends.

> If you're a street musician setting up to play with an open instrument case, it's pretty important to "seed the pot"

This concept is discussed at length, and backed up with scientific studies, in the excellent book Influence, by Robert Cialdini, PhD. This particular concept he calls "Social Proof", and it's one of about 8 major categories of influence that he discusses in his book.

Incidentally, dropping the PhD on the end of his name is another one of the forms of influence. While reading the book, I couldn't help but get tons of ideas for a marketing campaign involving social proof. The concept is of course much broader than seeding donation hats. In the book he even gets into the elements of how the Jim Jones tragedy went down, and social proof is at the core of it.

Here is a video of the author outlining the major categories of influence. http://youtu.be/HOypv1AqYu0

http://www.ted.com/talks/derek_sivers_how_to_start_a_movemen...

"With help from some surprising footage, Derek Sivers explains how movements really get started. (Hint: it takes two.)"

When you look at the book reviews on amazon it really looks like amazon doesn't care about integrity in the review system.

For example if you search for 'learn java', there's a total joke of a book at the top with torrent of fake 5 star reviews.

The first one I read even calls Java, 'the javascript' at one point.

It's an insult to decent authors and to the people who go to the trouble to write reviews that amazon allow this

Edit: I just realised why I found the names of the editor and illustrator familiar!

Ruby C. Perl and Ada R Swift

I stumbled upon a book by James Franco on Amazon.com a long time ago, and it had a 4-star review on the search page and on the product page.

I thought this was bizarre, because I just happened to have read reviews of it on literary and nonliterary sites giving it extremely poor reviews.

I clicked on "See all # customer reviews" on the Amazon page and the average rating on top immediately went down to 2 stars.

I thought this was a glitch, so I tried to access the product/book page a different way. The same thing happened to all of them.

There were always the same amount of reviews, as well.

About 6 months later I looked up the book again, and the "mistake" was gone. But I had taken a screenshot of it the first time, and there was only two more reviews. The 4-star average appeared on every page.

I look for it now, and it is back at 2 stars like it always was. The book in question is Actors Anonymous.

"Facebook says it uses pattern recognition to find and eliminate click farms and accounts that exist solely to like pages. Since March, says Facebook, it’s “notified 200,000 Pages that we’ve protected their accounts from fake likes.” I love the way they put this: “Protected,” as if the companies hadn’t paid for them."

This may be more accurate than the author of this article believes. I've read elsewhere that precisely because FB is using pattern recognition to combat this, the cleverest of the fake accounts spend a certain amount of their time doing random/normal stuff to avoid detection. And this includes liking random businesses. And one of the problems is that if a legitimate business pays FB to promote one of their posts (so it is shown to a bigger percentage of their followers), then the more fake followers they have, the more real money they are wasting to advertise to bots. So if a business is actually using FB to engage with their userbase, then the fake likes are in fact detrimental to them.

Great points

> And one of the problems is that if a legitimate business pays FB to promote one of their posts

The same applies to non-paid content. Facebook will only expose your content to a portion of your users, obviously the more fake users you have the fewer real people you'll reach. Engagement levels have shown to decrease when adding fake users in previous experiments.

I would like to add that if they would punish the businesses, this could of course be abused: if I have to pay $50 to have my competitor have serious spam issues with Facebook, possibly resulting in a penalty, that is some serious damage I could do.

This is a valid SEO tactic as well. You can have crummy back links to your competitor and google will penalize them...

valid dirtbag SEO tactic

Isn't that redundant?

Not just a problem if you spend money promoting to unwanted fake likes - EdgeRank decides who sees your un-promoted content based in part upon your average engagement levels, if half of your fans never engage because they are fake then your stats look worse to Facebook and so your content will be seen by a lower proportion of your real fans, even before you start spending $ promoting stuff.

Presumably, the snowshoe fake likes are somewhat even distributed, so there's no reason to think legitimate business A would be preferentially punished by fake like low engagement versus legitimate business B, right?

Fake likes may be randomly distributed, but real likes are not, so evenly distributed fake likes will be more detrimental to businesses with fewer real likes.

Veritaseum did a nice analysis and summary of this exact situation. https://www.youtube.com/watch?v=oVfHeWTKjag

"Protected" might be more accurate than you think initially. One of the first things that happens when blackhat SEO experts discover a "penalty" mechanism, is they will attempt to invoke it upon their competition.

Google deals with this re:bad links. Once they started penalizing, it became a new tactic to suppress competition.

And if they don't penalize for it, then it's a 0-risk proposition to try buying thousands of links just in case a few of them sneak by and boost your rank. What do you have to lose?

But like you said, Google doesn't know who paid for them, so if they penalize spam links you can use that against competitors. There's no good solution other than "Be 100% perfect at identifying and ignoring fake links."

Googles doesn't really say, and I believe they are caught in a tough spot - unable to create an enforceable rule.

https://support.google.com/webmasters/answer/2648487?hl=en

I actually think one of the things that stops competitors from buying poor quality links, is the fear that it might have the opposite effect. Google simply can't keep up with the spam that is created - so they are forced to take a moderate approach.

We did a Facebook page for an university project, and created a small campaign for it (we were trying to do landing-page style validation).

It got a lot of "likes" which looked a lot like fake accounts trying to look less suspicious, and none of them converted to the landing page, they were just clicking "like".

Facebook is a mess and, to some extent, a huge scam. I spent tens of thousands of dollars advertising on Facebook to build-up an audience for a page for a business we were spinning up. We used very narrow criteria to guide the advertising and, over time, got tons of likes to our page.

And then it happened. FB changed the percentage of people you can reach with posts to your own page. In some cases you are lucky to get 1%. Of course, they also added the ability to reach them through advertising. Which is a travesty. I paid plenty to get them to my page in the first place and now FB wants me to pay again and again every single time I want to say something to them. Fuck them.

I know of lots of people who spent a lot more than I ever have on FB who have abandoned the platform for marketing. One guy was doing a million dollars a month with them. Gone. He brought-up exactly the same issue I describe above. So, yeah, again, fuck them. Thankfully there are other ways to reach people on the 'net.

Yelp was the only company that caught us, hiding both of the reviews I bought behind their “not recommended” click wall and not counting them in F.A.K.E.’s rating. It has software that screens out suspicious reviews, not including them in a company’s star-rating. If they see too many of them, they will penalize a business’s page, putting a “consumer alert” on the profile for 90 days warning visitors that they think the business is buying fake reviews. People on Fiverr who were selling reviews would often say “No Yelp” in their descriptions, saying it was hard to make those “stick.”

This is why Yelp almost never disappoints. Kudos to the Yelp team for keeping the quality control strong and providing a great service.

> Kudos to the Yelp team for keeping the quality control strong and providing a great service.

Well that's not something you hear (read) everyday.

I think people looking for reviews generally find Yelp to be useful but business owners often hate them.

I make the same observation. My friends that own restaurants often talk about Yelp messing with their reviews and not showing the good ones. Myself, as someone that usually just uses Yelp to find a place to eat, I find it very useful and have no real complaints with the reviews.

I draw the exact opposite conclusion: Yelp is selling reviews and need to control their product. Twitter and Facebook aren't selling followers or friends, so they fight fakes... but not too hard, because that isn't their product.

Isn't the conventional wisdom that Yelp is a protection racket?

Exactly my thoughts. It's becoming more and more clear that Yelp pretty much extorts businesses to upgrade or they surface negative reviews at the top of the page. Since a typical person will read 2-3 reviews this can be pretty impactful.

I also recently heard it alleged that they submitted fake reviews when someone wouldn't upgrade, but that was the first time I had seen that come up.

And with a sense of humor about it in social media followup: https://twitter.com/Yelp/status/644230950159671296 https://twitter.com/Yelp/status/644244909994676226 :)

well

https://www.youtube.com/playlist?list=PLkVbIsAWN2luXUhlA1bO1...

TLDR: YELP salespeople keep spamming you with phone calls, Louis was getting tired of it, but still polite (I would simply tell them to eat duck), so pointed out his friends business two blocks over pays $5K every month and seeing his books there was zero effect. What does Yelp salesperson do? She emails him PRIVATE INFORMATION OF COMPETING business :o.

> A Lincoln could get me a Facebook review, a Google review, an Amazon review, or, less easily, a Yelp review.

Mostly off topic, but what's "A Lincoln"?

A 5 USD bill.

I was thinking a penny, but that makes so much more sense.

As far as I know, it's not something people say in real life, just a funny turn of phrase; "a Benjamin" is the only common use of a president's name for the denomination.

I'd be willing to bet a Washington that no one in real life uses the term Lincoln.

Friendly correction that Benjamin Franklin was not a president, your point stands though.

Hah! Something I even knew. Thanks for the correction. I guess we need to stop calling cash "dead presidents" [1] now.

[1] http://www.urbandictionary.com/define.php?term=dead+presiden...

Benjamin Franklin was never a President.

American writers and journalists are always doing this. Do they really not know it makes the text unintelligible?

Even googling "A Lincoln" or "Lincoln" doesn't make it clear what they're trying to say, or how it applies.

When I read that I initially thought they meant the car company (Lincoln Motor Company), therefore I thought they were trying to say "if you spend car's worth of money you get <this>" except when you read on it is clear they didn't get enough value for that to make sense.

The short is: If you're a professional writer, blogger, journalist, or whatever and have an international audience, just stop it.

There's a whole seedy world under the surface of Fiverr, and I'm surprised no one seems to have done an in-depth "Inside Fiverr" piece yet. Vectorization services that present as individuals in the US but are actually farmed out to workshops in non-English speaking countries. Voice actors whose work samples consist mainly of dubious get-rich-quick ads. And as this article points out, "SEO" and "reputation building" services as far as the eye can see. (For $5, you can have your yoga "article" featured on this woman's website, complete with up to "three... dofollow links": https://www.fiverr.com/maigent/post-your-content-on-my-yoga-...)

Really, even yoga?

In 2004, two czech graduate created massive campain for fake hypermarket: https://en.wikipedia.org/wiki/Czech_Dream

The thing about that hoax, is that they put together a massive PR and ad campaign, with billboards and other media ads. That's not quite the same as doing everything for a couple hundred bucks.

I remember seeing that when researching The Yes Men.

https://en.wikipedia.org/wiki/The_Yes_Men

A lot of this problem seems to come from Facebook/Twitter showing:

"You have 3 followers"

Perhaps it would be better if they showed publicly facing ranges:

  - < 1,000 followers
  - < 2,500 followers
  - < 5,000 followers
  - < 10,000 followers

Then above 10,000 (or whatever) it shows the actual number.

I see these fake like paid services losing a ton of business if social platforms took this approach.

Then you'd just spend $20 extra and buy 10k+ followers :)

Facebook et al would be able to spot 10,000+ like purchases a lot easier than a few hundred.

Then you buy them in batches of 2k/week.

In all seriousness, I've actually purchased 10k Twitter followers for $10. Twitter did not care one bit.

That was a good write-up. Thing is, this works. Even (or especially) for startups. Only if to kickstart the hype a bit. Real followers and likes will follow, but let's be honest here: Who doesn't automatically trust a business that has 19K Twitter followers and a few good reviews out there?

From the article: And these days, which is more real: a fake business with a real website or a brick and mortar business with no online presence?

> Real followers and likes will follow, but let's be honest here: Who doesn't automatically trust a business that has 19K Twitter followers and a few good reviews out there?

For now. As more companies cheat the system, people will be less likely to trust them based on things like followers and reviews.

Currently I don't trust the reviews of lots of physical businesses in my area. In my experience places with few reviews tend to be more reliable than those with many. My theory is that customers of physical businesses are more likely to leave reviews of negative experiences and that businesses respond with fake reviews. This is because there is a high 'activation energy' to making an online review of a place that you normally walk into and that only really angry customer bother. I tend to trust e.g. App store reviews more because the barrier to entry to leaving a review is lower so real positive reviews seem to happen more.

I find this sad.

Reminds me of reddit, and their fake users to start it off.

The thing is that the internet is so biased towards incumbents, you need to cheat a little to get the initial pull.

To me there's a big difference between real activity under multiple "fake" names, where content is still being submitted, and artificially raising the profile of one's company by just buying fake reviews (or followers or likes, which are basically micro-reviews of approval). Just like there's a difference between paying to have Facebook promote your page to real users, who can then decide whether or not they would like your page or status on their own, from just paying directly for the likes.

As an example, reddit kickstarted the hype quite well, and see, now they're worth 500 million ;)

I remember reading a research paper on how liking a post for initial two or three time could greatly increase the probability of the post actually taking off. I can't find the exact paper now though, so if anyone can remember reading a similar one, can you please link it?

Same kinda thing with reddit/HN- getting one or two upvotes within the first couple minutes of posting is essental to a post getting anywhere. It keeps the post afloat long enough for people to notice it, instead of getting lost with everything else.

I think it's the reverse that's more insidious. Who doesn't automatically distrust a business with just 24 likes?

Creates a situation where you almost have to buy some fakes to get to perception neutrality where real likes will follow.

Why would I trust a company's carefully curated Facebook and Twitter marketing pages any more than I would the company's advertisements that tell me how awesome they are?

This site brought up at lease two separate overlay popups while attempting to read it. I don't usually complain about this, but that's totally obnoxious and meant I didn't finish reading the article.

Is there a good method to block this aggressively user-unfriendly bullshit?

Telerik Fiddler2. I use the autoresponder to block patterns.

uBlock?

Firefox: https://addons.mozilla.org/en-us/firefox/addon/ublock/

Chrome: https://chrome.google.com/webstore/detail/ublock-origin/cjpa...

Safari: https://www.ublock.org/newdl/safari/0.9.5.2

Alternatively: Firefox has a built-in "Reader View" button in the address bar that extracts just the text and formats it for easy reading.

uBlock didn't fix it for me. Also, because the second overlay was huge, I couldn't close it down so I closed the tab instead...

Maybe it's time to NoScript?

I get the feeling having NoScript can be more annoying that not having it so I have been procrastinating installing it...

Speaking as someone who has NoScript installed... you are sadly correct.

I use "medium mode" in uBlock[1], which causes only 3rd-party scripts/frames to be blocked, not 1st-party ones. I could read the article fine, I had no idea there were issues with overlays with it, and it otherwise rendered fine.

Not blocking 1st-party scripts remove a lot of the tediousness of blocking scripts by default.

[1] https://github.com/gorhill/uBlock/wiki/Blocking-mode:-medium...

There's also ublock matrix which is noscript but easier to work with.

Thanks for the suggestion, it looks very interesting. And, obviously, thanks for all the work on uBlock Origin, interneting got much less annoying after installing it in every machine I get my hands on. Thanks!

Why? Just don't read articles from such annoying sites.

Right click, Inspect Element, Delete. most of those seem to be about 20 divs deep, but usually have a parent div labelled "modal_something", so move up the chain and delete that parent.

It is annoying to have to do that, so I only do it if I truly care about reading the page. But once it is in your browser, it is in your control. Use that control.

I do it a lot for Stylebot but only a really small minority of websites deserve that waste of time. It's a good tip, one I tell everyone I know about, most sites I read on a daily basis are highly tweaked by the Inspect Element tool plus Stylebot.

I read it in "reader view" in FF, that done the trick.

I always use Firefox's Reader View when reading text-based articles. Especially on mobile. It makes the pages all the same, distraction-free, and fast to scroll.

In Chrome the 'Clearly' extension works similarly. Not on mobile sadly.

Adblock Plus + Element Hiding Helper for Adblock Plus

In Safari, simply switch to the Reader view, even when the popups are up. It gave me the text in a clean format without any distractions.

Firefox and Safari has a build in reader-view. In either browser they show up in the URL bar, on safar it's a few horizontal lines, in Firefox it's a little open book.

It'll strip all menus, ads and other cruft leaving you just the text, headlines and pictures nicely formatted.

I use BehindTheOverlay on Chrome.

But who, in this day and age, would trust or hire a new online business that had a page with 0-100 likes? The apparent presumption on seeing under 1k likes or under a few thousand twitter followers is the business is defective. Or forgotten, or no good. The incorrect wisdom of the crowd is that a legit business has a certain number of likes, followers etc, or it looks suspicious

For a new local business who is never going to get a HN or /. effect, or get some "viral buzz" from the cool kids, buying a few thousand can be a way to get to perceived neutrality. Enough likes or followers that they're no longer clicked away from for being "suspiciously" unloved.

Fake reviews on the other hand are much more directly fraudulent. Mind I'm surprised anyone at all still trusts online reviews, apart from negative ones!

As someone who does not use Facebook or Twitter, I don't give a hoot about how many "likes" or "followers" a business has, any more than I believe in any other marketing campaign. These measurements measure nothing meaningful to me as a potential customer, as this article clearly demonstrates. I've always found it baffling that people even remotely believe online "reviews" written by strangers.

I'll turn it around: Who is stupid enough to think, "Wow, this business's carefully curated Facebook marketing page has 10,000 likes.. That must mean they are truly reputable!"

Personally I agree completely. As someone in tech I find it very difficult to ascribe meaning to a meaningless number. Especially when people "like" things they hate for comedy or irony purposes, or clicked accidentally a few years ago. Reading online reviews is a case of trying to glean a little truth from something everyone should know is being heavily gamed. But they don't.

Amongst the non-tech population there is trust taken from these numbers. People believe those reviews. People seem to believe that a lot of likes is indicative of trust. People used to believe that clicking blue links was good, and so the web used to be a cesspool of adsense sites providing no real information.

Sad fact is it works. Both with the algorithms - more likes gets more random likes from nowhere over time, and with the visitor perceptions. probably only the non-technical ones.

If it's on the web it must be true, right? /s

I'm very aware of this problem. I set up Sitetruth.com to try to combat it. Here's what Sitetruth has on Launchrock.[1] Is there an street address on the web site? No. Anything we can tie to a list of US businesses? No. Better Business Bureau link? No. SEC filings? No. Rating: Do Not Enter. SiteTruth takes the position, "when in doubt, rate it down".

We'd use D&B data if it wasn't so expensive; this is just the demo version. There's a whole industry out there tracking business info, and the search industry isn't plugged into it. (Why? Google gets about a third of its revenue from AdSense ads on other sites. If Google really cracked down on clickbait sites, Google's own revenue would drop. We once calculated it would drop 14-17%.)

Back in 2011, I wrote "Social is bad for search, and search is bad for social".[2] That's when Google had started using social signals for search ranking, and the black-hat SEO industry had discovered how easy it was to manipulate Google via social. Old-style link farming, setting up lots of fake web sites to create links, is expensive, with servers to run and sites to fill. With social spamming, the social network hosts the spam for you, for free! You can still buy "bulk likes" for Facebook.[3][4] The price has gone up a little, indicating that Facebook is having some success in getting rid of fake accounts, but not much.

[1] http://www.sitetruth.com/rating/freakinawesomekaraoke.launch... [2] http://www.sitetruth.com/doc/socialisbadforsearch09.pdf [3] http://www.buylikesandfollowers.net/buy-facebook-likes-cheap... [4] https://boostlikes.com/

I really like this idea, but it seems fairly punishing for new / all remote / self-funded businesses. If you're running a company out of your own pocket (no SEC filings), aren't anywhere near big or old enough to have a BBB entry, and are remote (what street address should someone use? their apartment?) I'd like to know what signals you'd recommend bootstrappers use to engender trust for a service like yours, because I don't think "person is known on HN, posts on medium / Twitter about their service" is necessarily amenable to automated analysis.

"it seems fairly punishing for new / all remote / self-funded businesses."

There's no right to run an anonymous business. It's illegal in the EU, and a criminal offense California if you accept credit cards. The customer needs to be able to find you and, if necessary, haul you into court to get a refund. You can register a D/B/A name and get a post office box to create a legal identity. That's fine. Hiding behind a web site with "private registration" is a scumbag flag.

Read yesterday's article about how Venmo helps scammers by making anonymous payments easy.

I definitely agree with that part. Does your service check WHOIS data, then?

No, WHOIS data is of such low quality that it's useless. SSL cert data is of better quality, for OV (Organization Validated) and EV (Extended validation) certs. Contact addresses found in text on the web site are also useful. Those we match to commercial business directory information.

Considering that many smaller companies use e.g. CloudFlare's SSL, I'm not sure that would help them, but good to know.

We ignore Cloudflare's many SSL certs. We have a short blacklist of MITM-as-a-service content delivery networks. Here's my paper on that.[1]

The list is short. Here it is:

    cloudflare.com – a front-end network for sites, controlling 36,280 domains.
    incapsula.com – a front-end network for sites
    sonymusic.com – operates sites for their range of artists. 
    Janrainengage.com – customer tracking service
    edgecastcdn.net – Verizon caching system
    fiducia.de – security service for banks
    vin65.com – wine seller with many sites for various wine brands.
    practiceweb.co.uk – a hosting service for accountants

Sites which use those services are not blacklisted by Sitetruth, but the ownership data in their SSL certs is ignored as meaningless. The CA/Browser Forum is looking into ways to express this better in SSL certs. A cert with fifty unrelated businesses is just silly, and it's a transitional thing until everybody gets TLS-capable OSs and browsers so shared IP doesn't mean shared cert. (Windows XP/IE 6 being the problem).

[1] http://john-nagle.github.io/certscan/whoamitalkingto04.pdf

> Better Business Bureau link?

Is lacking a BBB link supposed to be a bad thing or a good thing?

http://abcnews.go.com/Blotter/business-bureau-best-ratings-m...

BBB links usually provide a way to get a street address, from the BBB site, for the business. The BBB pages also link back to the site they're describing, and we check that. (Yes, there are fake BBB links.)

SiteTruth can be fooled, but you usually have to commit a crime to do it. We cross-check with PhishTank to down-rate phishing sites, which helps put a lid on that problem.

I have been on Fiverr for few weeks and ordered two articles for $5 from some guy claiming from the US. The result was hilarious; a kindergarten kid can use much better grammar.

Also found some gigs encouraging piracy, found a guy who claims to give 3 ebooks ( you name it) for $5 ?!

Another one who can give you any 3 Udemy courses download for $5 - https://uk.fiverr.com/udemy_guru/give-you-any-4-u-demy-cours...

I think Fiverr should hire a team to monitor the activity and authenticity of its sellers.

> I think Fiverr should hire a team to monitor the activity and authenticity of its sellers.

That would cause most of their revenue to evaporate.

They also don't have a "report this post" button anywhere to allow people to flag spam or illegal activity. That would at least be something they could do...

Could't agree more. That's the least they can do.

Great point. If they won't actively police their sellers at least they could let visitors volunteer to do so through such a feature.

Yes, and if the police won't actively police every square foot where there are people, they could at least let citizens do so through concealed carry of a weapon.

And yet, somehow, there's a large group of people that is always for people meddling in other people's business by telling on them (including calling the police when they see a citizen legally open-carrying!).

It must be because laissez faire is not easy to spell.

I don't really see how this is relevant however reporting suspicious Fiverr jobs is totally different giving citizens the right to employ lethal force. The stakes are slightly different.

How would that benefit them at all?

At some point they will do it - when they "grow up" and they can afford to crack down. Tumblr cracked down on porn once they became big. Yelp didn't care about fake reviews, but now they do as they've become big. And so on.

>"Tumblr cracked down on porn once they became big."

Define "cracked down"? Because from my perspective, there is plenty of it on Tumblr.

You really shouldn't expect anything better for an article that pays the writer just $4 (after Fiverr takes its cut).

Good writing isn't something everyone can do. It takes me at least an hour to write around 500 words. A full 2k word post can take 6+ hours (since longer posts tend to be more complicated).

If I was charging Fiverr rates, I would end up making $4/hour.

I am not complaining about the research that went into the article, but the content at least must be comprehensible.

My observations so far - the top gigs are the ones who are doing what they are good at. A few examples - https://uk.fiverr.com/ukulelebandito/write-perform-and-recor...

https://uk.fiverr.com/reverbdominatio/professional-voice-for...

I think Fiverr is for someone who can help others with their hobby, but I highly doubt if it can ever be fully grown into a professional freelancer network!?

If I was to use Fiverr, I would use it as a source of cheap leads. Offer a low cost service (even at a loss), then upsell more expensive services.

A personal observation in the Chrome store.

I created an extension which I published in the Chrome store. I did notice a curious statistics with another similar purpose extension, which is also quite popular -- both are over 1 million users:

The ratings-per-user ratio for the other extension is almost 5 times that of my extension.[1]

I can't figure any other convincing explanation for this other than the other extension is buying ratings. If there is an alternative explanation I would like to hear it, but I can't think of one -- five times is way beyond one would expect IMO.

[1] number of ratings / number of users.

Perhaps the other extension is prompting the user for a rating like you see so often in Android/iOS apps. This tactic genuinely works. If not, then yes, it's plausible that they're buying ratings.

Good point, I didn't think of this... I looked and it does indeed invite users to rate it in at the top of its options page. So as far as I am concerned this explains the difference.

I could probably quantify how much this works by adding the same sort of invitation to rate but this would go counter to the spirit of the project, so I will pass on the experimentation.

I actually don't know whether to be annoyed/dismayed or impressed. There's something really quite accomplished about this bizarre, entirely abstract, virtual enterprise.

This could pass as art ;)

"My sudden surge in popularity raised no flags at Twitter headquarters as far as I could tell, even though many of the followers had egg avatars who hadn’t tweeted in years."

I'm not sure that Twitter could use how often you tweet as a metric. I haven't tweeted anything in years either but I still use the service to see what my favourite people are saying. I'm sure there are plenty of lurkers like me.

How can we be sure that this article isn't a deep-cover "fake review" for Yelp? It tells a convincing story and manages to throw in Yelp's huge music selection - I mean, its effectiveness at removing fake reviews.

I'd have read that but I didn't get more than two lines in before an overlay appeared in front of the content. So I gave up.

I made it to the second overlay halfway through.

Right-click, inspect, then delete a node or two.

Speaking of which, it would be nice if chrome or firefox had a developer keyboard shortcut to delete the node the mouse is currently hovering on without having to inspect or open the dev tools. That'd make dealing with intrusive content on the web much easier/faster.

This is particularly bad in the moving industry. Not only are movers notorious for bribing customers to post good reviews but there are also several well ranking "Moving Specific" review sites that are completely fake. The industry also happens to have some of the most horrific scams...

The less frequently someone needs a given service, the more confusing (in the best case) and scammy (in the normal to worse case), the sales process is.

Look at weddings, funerals, residential remodels, residential HVAC, and the like as other examples. Some of those make used car sales look pleasant and transparent...

I've noticed this phenomenom too. I think mattresses is another good example. Is there a name for this?

Good point. Anything with low repeat buying is going to run into these problems. That's why verified reviews are so critical in these verticals.

In France, an agency created a fake company selling kebap sandwich on Mars. They had 105.000 views on their video, 46k followers, 20k facebook fans...and orders.

What's scary isn't the fact people buy fake reviews/followers/etc. but the fact people actually believe in these vanity metrics and want to order afterwards.

To be fair, almost the entire advertising industry is built around trying to effectively lie to people.

True that :( I've worked in the industry during 3 years before figuring that out! The problem is mostly for platforms to find a way of identifying which reviews are real or fake, depending on IP, rate of reviews, recency of the user, etc.

Reviews aren't actually a bad source of information if you are ordering on line. If you can't check it out directly what else can you go on that is better? I tend to look at the negative reviews which probably have not been paid for and see if they seem that bad or not.

Anecdotal, but I've been aware of this fake review industry for probably about 6 years and I am amazed that almost everyone I talk to does not believe the scale and the scope. They think there might be one or two fake reviews on an Amazon product when in reality it can be closer to 15 out of 18. Or in some cases every single one.

People who are seriously doing this aren't paying $5 a review. They are paying $100-$500. At this point I just assume almost all reviews are fake. Especially on Amazon and to a slightly lesser extent Yelp.

"He claimed he’d created 5 million Twitter accounts...."

Anyone still wondering if Twitter is overvalued at ~$19 billion, wonder no more.

Paul Graham said that if there is alot of potential energy from a broken system there must be a way to monetize it. Really, it seems like someone can make a startup to verify the company and reviews.

In fact, why dont we all open source this thing.

refering to this: https://www.youtube.com/watch?v=q7K0vRUKXKc

That sounds a lot like Dun & Bradstreet[0], no?

[0] https://en.wikipedia.org/wiki/Dun_%26_Bradstreet

Sounds like Dun & Bradstreet, yes. But, I'm not sure if they can verify the identity of an amazon/ebay seller [0]. [0] https://www.reddit.com/r/books/comments/3l1epw/how_to_scam_a...

I've thought this would be a good new feature for Stripe. After payment, allow user to leave a review. Then you know the person bought the product.

That would def be a good feature for stripe.

A karaoke truck kind of sounds like a good idea. Shame it's fake.

I think it even hit the front page of reddit. I remember seeing something about a karaoke truck recently.

So maybe somebody should make it real.

Maybe that was part of the marketing this author bought, just not explicitly listed in the article. If you can buy all those other things, why not reddit upvotes too?

CBS Canada, did this about a year ago, very interesting watch: https://www.youtube.com/watch?v=3y0pYUdfGiw

Yes, writing fake reviews is illegal. False advertising is a misdemeanor crime

Does this mean those Amazon affiliate sites are illegal too? A lot of these sites review products as if they have used the them, but there is no way they could've paid real money for the dozens of products they review?

Also, how is this different from celebrities endorsing products and appearing in ads? How do we know they bothered to use the product even once, other than the time they were shooting for the ad?

My guess is that most celebrity endorsements happen after at least some examination of the product. After all, celebrities have brands to protect too. And I imagine their endorsements are carefully worded; it's only false advertising when somebody says something that isn't true.

Of course, as you demonstrate here, there's at least some public understanding that celebrity endorsements aren't actually reviews, so a good lawyer would argue that since you don't expect celebrity endorsements to be factually true, they couldn't possibly be false advertising.

There are probably plenty of fakes, but if you're successful, brands will send you their products for you to review.

Some people are given free product in return for a good review.

In England you must make it very clear if content is your regular content of if it's sponsored content. This is hard for YouTubers to comply with because a big pre-roll splash probably isn't enough.

http://www.bbc.co.uk/news/uk-30203816

http://www.theguardian.com/media/2014/nov/26/vloggers-must-t...

https://www.asa.org.uk/News-resources/Media-Centre/2014/Maki...

You don't have to pay for it, btw. There are sites you can post to in order to attract people that will just like/follow your stuff for free. It doesn't take much searching to find them. You won't get the same numbers, but if you're on a budget, it works.

FB's own advertising is as cheap than fiverr. $5 can get you a thousand or two likes if you geo locate the campaign to certain parts of the world. Some parts of the world seem to like everything!

Seems to be all legit accounts, so presumably far safer from any FB fake detection algorithms. Just don't expect them to buy your service!

"He claimed he’d created 5 million Twitter accounts using scripts, and had completed 10,000 Fiverr orders for Twitter followers. He’d been banned from Fiverr three times before but says he doesn’t know why."

Duh...

This puts most corporate marketing departments to shame.

Why?

I mean, if it wasn't all fake traffic... if he did this "for real", it would put them to shame... heh

As an entrepreneur, I became so disappointed with this article about 60% of the way through, that I had to stop reading. When the author started getting phone calls on his burner phone from people wanting to hire his van, he should have manned up, stuck with this damn business, and written instead about bootstrapping. With stats like the ones he reported, he should have stuck with Freakin' Awesome Karaoke Express even if it meant moonlighting as a KJ and eventually backing into a seed round he never anticipated.

Seriously. Those facebook reviews sound authentic because those are real user stories just waiting to really happen. The real user queries prove it.

I'm so disappointed at those unfielded phone calls that I can't even bring myself to read any more. A business was born three months premature, and instead of nurturing it in neonatal intensive care, the author just let this beautiful creature die. :*(

There it is, folks. The first "Pro-Life" argument to emerge from the "corporate personhood" metaphor.

All downhill from here. Grab your popcorn.

Out of curiosity, why do you assume the author is male?

I'm not going to update my comment because it would look like I made the premature birth analogy due to the author's gender, which I literally had incorrectly assumed was male (so my comment and choice of analogy was absolutely not due to the author's gender).

Sorry about the mistake/assumption. You asked why I made the mistake - no particular reason. (Though anecdotally I've heard males talk about these kinds of actions, while women business owners I know don't, in fact are principled about not buying fake reviews, likes, or twitter followers. The first paragraph also says "You might have seen a review on Yelp that said it’s perfect for a girl’s night out" which I read as a male stating it. Anyway I shouldn't have made the assumption.)

Out of curiosity, why do you assume that because the author said 'he' that he projects that gender onto the author, as opposed to not actually knowing and choosing his own gender, as is common, or choosing a random gender, as is common?

Are you one of those people who insists that everyone say "he/she" every single time they don't know? If so, then why don't you also demand that the person say "she/he" 50% of the time? With no more than a specified number of imbalances per time interval and instance interval? And no more than a specified change in number of imbalances per same?

I only bring this up b/c you seem to grant yourself the title of auditor, so out of curiosity, I'd like to see if you can stand up to auditing yourself, or if you simply expect more out of others than you're able to give. My hypothesis is that you have no grounding or ability to generate introspective thought, just a little provocative scheme to grant yourself power opportunistically.

you're coming down on my side, but I think this 3-paragraph tirade is a bit over the top... the comment you're replying to was 11 words, and also correct in that, yes, I had assumed the author was male.

Maybe it's not his life's ambition to own and operate a truck that drives around to cater to screeching drunk people?

Turn that frown upside down buddy, have I got some good news for you!

Having read that article, you now know about the demand for karaoke vans and can go serve that market.

It's not faking, it's just doing things that don't scale /s

Reply on news.ycombinator.com