[00:00] (0.16s)
PayPal does not want you seeing this
[00:02] (2.16s)
video. A few hours ago, PayPal's lawyer
[00:04] (4.32s)
sent me a cease and desist letter and
[00:06] (6.24s)
requested Patreon take down my video
[00:08] (8.40s)
under a copyright infringement claim. I
[00:10] (10.56s)
consider this a direct attack against my
[00:12] (12.56s)
fundamental rights as an independent
[00:14] (14.48s)
journalist. PayPal has specifically
[00:16] (16.48s)
taken issue with the fact that an
[00:18] (18.08s)
anonymous source discovered Honey had
[00:19] (19.84s)
left their source code exposed within
[00:21] (21.76s)
their iOS app, which if you know where
[00:24] (24.00s)
to look, is publicly accessible. This
[00:26] (26.00s)
anonymous source found the code and sent
[00:28] (28.24s)
it to me. I reviewed the code for
[00:30] (30.40s)
security research purposes, which is a
[00:32] (32.48s)
protected act, and I only published
[00:34] (34.72s)
relevant sections of the code that
[00:36] (36.48s)
provide important context to the story,
[00:38] (38.48s)
which I consider within the realm of
[00:40] (40.40s)
public interest. Nevertheless, I have
[00:42] (42.48s)
removed sections of the video containing
[00:44] (44.32s)
said code, not because I believe PayPal
[00:46] (46.40s)
are correct, but because I know they
[00:48] (48.08s)
will have YouTube take this video down.
[00:50] (50.08s)
It was only a small segment of the
[00:51] (51.52s)
video. Therefore, I have made this
[00:53] (53.12s)
decision quickly under duress and to
[00:55] (55.28s)
proceed with this publication as quickly
[00:57] (57.04s)
as possible. If PayPal thought this
[00:58] (58.72s)
would stop me, they were sorely
[01:00] (60.56s)
mistaken. Enjoy the video.
[01:02] (62.56s)
>> I don't trust Honey, I don't know what
[01:04] (64.96s)
it is. I don't trust it. They push so
[01:08] (68.24s)
much advertising. They push so much
[01:10] (70.64s)
advertising and then people are getting
[01:12] (72.32s)
money out of it and then people are are
[01:14] (74.00s)
saving money out of it and that makes
[01:15] (75.44s)
sense, but the other part doesn't make
[01:16] (76.72s)
sense. And then something's going on
[01:18] (78.08s)
here. Who where's all this money coming
[01:20] (80.24s)
from? I have a feeling there's going to
[01:22] (82.64s)
in like a couple years there's going to
[01:23] (83.84s)
be the Great Honey Conspiracy.
[01:27] (87.76s)
PayPal's Honey Money Saving browser
[01:30] (90.16s)
extension is accused of scamming
[01:32] (92.00s)
customers and YouTube creators.
[01:33] (93.76s)
>> The biggest YouTube scam has just been
[01:35] (95.92s)
revealed. The greatest scam in the
[01:38] (98.48s)
history of the creator economy.
[01:40] (100.24s)
>> People have been getting influenced and
[01:41] (101.84s)
scammed, and I've been a part of it.
[01:43] (103.36s)
>> It involves almost every high-profile
[01:45] (105.28s)
creator that I can think of, including
[01:47] (107.28s)
myself. Honey paid a lot of influencers
[01:49] (109.44s)
up front while taking money out of their
[01:51] (111.84s)
back pockets.
[01:52] (112.72s)
>> Wa, that's a little shady, mind you. No
[01:56] (116.48s)
way. That signs you up for their
[01:59] (119.04s)
affiliate link. No way.
[02:01] (121.76s)
>> What? Brother. Brother.
[02:06] (126.24s)
>> I'm the biggest Honey fan in the world.
[02:09] (129.84s)
Lwig. No. They've also been lying to
[02:12] (132.96s)
consumers.
[02:13] (133.68s)
>> They are clearly not giving the best
[02:15] (135.52s)
coupon codes and discounts to users. If
[02:17] (137.92s)
you have downloaded Honey, I do
[02:19] (139.76s)
recommend removing it from your browser.
[02:21] (141.60s)
I do think that this investigation for
[02:23] (143.36s)
Mega Lag will lead to a huge class
[02:25] (145.52s)
action lawsuit against Honey.
[02:26] (146.72s)
>> So, on behalf of Creators Everywhere, I
[02:28] (148.80s)
have filed a class action lawsuit.
[02:30] (150.40s)
>> We're suing PayPal and Honey, Gamers
[02:32] (152.16s)
Nexus is the lead plaintiff in a class
[02:33] (153.84s)
action complaint filed against PayPal
[02:35] (155.68s)
and its property Honey.
[02:39] (159.28s)
Well, it's safe to say PayPal's lawyers
[02:42] (162.64s)
love my last video. But don't worry,
[02:44] (164.80s)
PayPal. You know what they say, all
[02:46] (166.96s)
press is good press, right? Let's see.
[02:50] (170.72s)
Since releasing part one of my
[02:52] (172.08s)
investigation, Honey has lost over 6
[02:54] (174.32s)
million users. Google has changed their
[02:56] (176.16s)
policies for browser extensions,
[02:57] (177.52s)
blocking Honey from claiming undeserved
[02:59] (179.20s)
commissions, and most significantly,
[03:01] (181.04s)
PayPal has been slammed with over 20
[03:03] (183.36s)
class action lawsuits, accusing them of
[03:05] (185.92s)
wiretapping, computer hacking, unfair
[03:08] (188.08s)
competition, consumer fraud, tortious
[03:09] (189.92s)
interference, and unjust enrichment.
[03:12] (192.88s)
Yikes. Maybe not them. And unfortunately
[03:15] (195.12s)
for PayPal, I suspect more lawsuits will
[03:17] (197.28s)
be filed following the release of this
[03:19] (199.20s)
video. Why? Well, you see, besides lying
[03:22] (202.16s)
to consumers and stealing from
[03:23] (203.52s)
influencers, Honey was also causing
[03:25] (205.76s)
serious harm to small businesses. So
[03:27] (207.92s)
much so that I believe brands have
[03:29] (209.84s)
likely suffered millions of dollars in
[03:32] (212.00s)
damages. But that's not all. In this
[03:33] (213.92s)
video, you're also going to learn how
[03:35] (215.60s)
Honey was collecting and sharing your
[03:37] (217.12s)
data and how they, in my opinion,
[03:39] (219.68s)
illegally targeted miners through
[03:41] (221.36s)
influencers such as Mr. Beast. Strong
[03:43] (223.76s)
accusations, I know, but I have the
[03:45] (225.84s)
receipts. I've reviewed internal data,
[03:47] (227.92s)
Honey source code, their pitch deck to
[03:49] (229.52s)
investors, disturbing emails between
[03:51] (231.36s)
Honey and small businesses. I've
[03:52] (232.72s)
reviewed interviews, and much, much
[03:54] (234.88s)
more. You're about to get insights into
[03:57] (237.20s)
one of the most scummy, predatory
[03:59] (239.28s)
business models I have ever seen. You'll
[04:01] (241.68s)
also find out why my investigation has
[04:04] (244.00s)
triggered multiple class action lawsuits
[04:06] (246.32s)
against other corporate giants such as
[04:08] (248.56s)
Microsoft, Capital 1, Cler, Rakutin, and
[04:11] (251.76s)
Retail Me Not. Per usual, the views,
[04:14] (254.08s)
allegations, and conclusions expressed
[04:15] (255.76s)
in this series are my opinions based on
[04:17] (257.76s)
evidence I have gathered, which will be
[04:19] (259.28s)
shared throughout. Ladies and gentlemen,
[04:21] (261.44s)
welcome to part two of the honey trap.
[04:36] (276.32s)
There were many moments throughout this
[04:37] (277.84s)
investigation where I thought I'd
[04:39] (279.28s)
reached the end of the rabbit hole. I
[04:40] (280.72s)
thought I understood everything there
[04:42] (282.32s)
was to know about Honey's dirty business
[04:44] (284.24s)
model. They were claiming the credit for
[04:46] (286.24s)
sales they hadn't earned, poaching the
[04:47] (287.84s)
affiliate commissions from content
[04:49] (289.12s)
creators, and they were intentionally
[04:50] (290.72s)
withholding discounts from users for
[04:52] (292.48s)
their own financial gain, rendering the
[04:54] (294.56s)
entire value behind their core product a
[04:57] (297.12s)
lie. Kind of hard to believe it could
[04:58] (298.96s)
get much worse than that, but
[05:00] (300.32s)
unfortunately, it does. And I realized
[05:02] (302.56s)
this after noticing a strange but
[05:04] (304.56s)
reoccurring anomaly. You see, in most
[05:06] (306.96s)
cases, when Honey popped up at checkout
[05:08] (308.64s)
and I clicked apply coupons, a discrete
[05:11] (311.12s)
tab would open in the top left corner of
[05:13] (313.28s)
my browser. That's Honey injecting a
[05:15] (315.76s)
simulated referral click, attempting to
[05:17] (317.68s)
claim the credit for a sale. But here's
[05:19] (319.60s)
where things got weird. Every once in a
[05:21] (321.76s)
while, I'd come across a store where
[05:23] (323.44s)
after clicking apply coupons, Honey
[05:26] (326.08s)
wouldn't load their affiliate link. So,
[05:28] (328.96s)
why not? At first, I thought it might be
[05:31] (331.36s)
a bug, but quickly ruled that out after
[05:33] (333.60s)
noticing the same anomaly on multiple
[05:35] (335.68s)
websites. Even stranger, when this would
[05:38] (338.16s)
happen, I was generally finding better
[05:40] (340.56s)
deals. So, what gives? Well, for the
[05:43] (343.52s)
longest time, I was under the assumption
[05:45] (345.36s)
that Honey would only pop up on websites
[05:47] (347.76s)
that had agreed to partner with them.
[05:49] (349.92s)
But I couldn't have been more wrong. And
[05:52] (352.08s)
I realized this after stumbling across a
[05:54] (354.08s)
post on Shopify's community forum where
[05:56] (356.24s)
a store owner complained that Honey was
[05:58] (358.40s)
leaking their coupon codes without their
[06:00] (360.32s)
permission. And as I read through the
[06:02] (362.32s)
replies, I could see other stores had
[06:04] (364.64s)
the same issue. Now, this was a pivotal
[06:06] (366.72s)
moment in my investigation because it
[06:08] (368.56s)
turned my entire understanding of
[06:10] (370.08s)
Honey's business model upside down.
[06:12] (372.16s)
Until now, I was so focused on the
[06:14] (374.24s)
impact Honey had on consumers and
[06:15] (375.92s)
influencers that I didn't even think to
[06:18] (378.00s)
consider the potential impact to
[06:19] (379.68s)
businesses. Though in my defense, Honey
[06:21] (381.68s)
consistently referred to the 30,000
[06:23] (383.44s)
stores on its platform as participating
[06:26] (386.48s)
stores. So, one would assume that every
[06:28] (388.64s)
store on Honey's platform joined
[06:30] (390.48s)
willingly and with consent. Yet,
[06:32] (392.40s)
clearly, that wasn't the case. So, I
[06:35] (395.12s)
reached out to several business owners
[06:36] (396.96s)
to learn more. Honey takes the codes
[06:41] (401.68s)
that people use on our website, the
[06:44] (404.08s)
discount codes, and makes them public to
[06:46] (406.88s)
everybody who's using Honey. We started
[06:49] (409.12s)
having problems where customers would
[06:51] (411.52s)
get in touch via our support website
[06:54] (414.08s)
saying, "Oh, hey, I just tried to use
[06:56] (416.40s)
this discount, this 60% off coupon, and
[06:58] (418.88s)
it doesn't work." And I'd be like, "We
[07:01] (421.84s)
can look into it." Like, "Dude, that is
[07:04] (424.24s)
that is not your discount. Like, how did
[07:06] (426.24s)
you get that?" And they'd be like, "Oh,
[07:07] (427.76s)
it it just showed up through the Honey
[07:09] (429.60s)
extension." I hadn't heard about Honey
[07:12] (432.32s)
before and I was just hanging out on the
[07:15] (435.44s)
couch with my girlfriend and she made
[07:17] (437.68s)
some comment about oh this app Honey
[07:19] (439.68s)
that basically gives you all these
[07:21] (441.44s)
coupon codes and I think she asked if we
[07:23] (443.76s)
were on it. I said no and then she
[07:26] (446.72s)
checked out our website and said oh you
[07:28] (448.80s)
actually are on it and it's giving this
[07:31] (451.36s)
15 or 20% discount code. And so that's
[07:34] (454.08s)
when I found out about it and I was
[07:35] (455.76s)
pretty I was pretty shocked because I
[07:40] (460.00s)
didn't know how long it had been going
[07:41] (461.76s)
on for. Um, yeah, it cost us thousands
[07:44] (464.80s)
and I haven't even looked into how far
[07:47] (467.92s)
back it went, but I know we just ended
[07:49] (469.92s)
the code right away. Now, if you think
[07:52] (472.48s)
that sounds bad, well, trust me, it gets
[07:55] (475.44s)
a whole lot worse. But first, it's
[07:57] (477.92s)
important we understand the scale of
[07:59] (479.76s)
this issue. As you just heard, these
[08:01] (481.84s)
store owners clearly never consented to
[08:04] (484.16s)
being on Honey. So, they can't possibly
[08:06] (486.24s)
be among the so-called 30,000
[08:08] (488.56s)
participating stores, right? So, why was
[08:11] (491.20s)
Honey lying to consumers about the true
[08:13] (493.28s)
number of stores supported by its
[08:15] (495.36s)
platform? And just how many stores did
[08:17] (497.84s)
Honey add to its platform without
[08:19] (499.92s)
consent? Was it hundreds, thousands,
[08:22] (502.56s)
tens of thousands? Well, that's where
[08:25] (505.44s)
things get interesting. You see, when I
[08:27] (507.52s)
started combing through the business
[08:28] (508.88s)
side of Honey's website, I noticed that
[08:30] (510.88s)
Honey claims to only be partnered with
[08:33] (513.52s)
10,000 brands. That's 20,000 less than
[08:37] (517.52s)
what they advertise to consumers. So, I
[08:39] (519.60s)
figured, okay, I guess 10,000 stores
[08:42] (522.24s)
intentionally partnered with Honey, and
[08:43] (523.76s)
the remaining 20,000 never signed up.
[08:46] (526.48s)
Now, that would already be incredibly
[08:49] (529.28s)
concerning and deeply misleading. But,
[08:51] (531.60s)
as it turns out, the deception is far
[08:54] (534.40s)
worse than that. You see, as I continued
[08:56] (536.56s)
digging, I stumbled across this file
[08:58] (538.16s)
from Honey's extension called supported
[09:00] (540.40s)
domains, which gets loaded onto your
[09:02] (542.40s)
computer when you install Honey. And
[09:04] (544.80s)
this file contains a list of over
[09:08] (548.48s)
180,000 online stores. That's a massive
[09:12] (552.32s)
discrepancy from what's advertised
[09:14] (554.40s)
anywhere on Honey's website. Now, of
[09:16] (556.72s)
course, we could start making
[09:17] (557.84s)
assumptions based on those numbers
[09:19] (559.52s)
alone, but we don't have to because it
[09:22] (562.72s)
just so happens that Honey also keeps
[09:25] (565.12s)
detailed information for each of the
[09:27] (567.20s)
online stores in this list. And it's
[09:29] (569.92s)
relatively easy to access that data. For
[09:32] (572.48s)
example, on Chrome, if we click manage
[09:34] (574.64s)
extensions, turn on developer mode,
[09:36] (576.72s)
click on the service worker for Honey,
[09:38] (578.40s)
and click on the network tab, we can now
[09:40] (580.64s)
monitor what Honey does behind the
[09:42] (582.16s)
scenes a little closer. So, if we say
[09:45] (585.04s)
visitapple.com, we can see that Honey is
[09:47] (587.68s)
now accessing information from their
[09:49] (589.44s)
database related to Apple. Looking at
[09:51] (591.76s)
this one in particular, we can see that
[09:53] (593.68s)
2.9 million Honey users visited Apple's
[09:56] (596.64s)
website in the past 30 days. We can also
[09:59] (599.36s)
see that none of those users saved any
[10:01] (601.12s)
money from coupons because, well, Honey
[10:03] (603.44s)
has no coupon codes for Apple's website.
[10:05] (605.68s)
We can even see internal support and
[10:08] (608.00s)
monetization notes that were left by
[10:10] (610.16s)
Honey's employees. Interesting. Most
[10:13] (613.44s)
importantly, however, we can also see
[10:15] (615.12s)
that Honey has an affiliate link for
[10:17] (617.04s)
Apple's website, confirming that Apple
[10:19] (619.20s)
does have an affiliate partnership with
[10:21] (621.28s)
Honey. But if we look at the same data
[10:23] (623.36s)
for say Abocrombian Fitch, we can see
[10:25] (625.68s)
that Honey doesn't have an affiliate
[10:27] (627.60s)
link. Yet, Honey is still offering
[10:29] (629.36s)
coupon codes for that store. And sure
[10:31] (631.68s)
enough, if we click apply coupons on
[10:33] (633.76s)
Abocrombian Fitch, Honey doesn't load an
[10:36] (636.56s)
affiliate link through that infamous
[10:38] (638.32s)
sneaky tab. Aha, things are starting to
[10:41] (641.76s)
make more sense. So, all I had to do now
[10:44] (644.48s)
was collect this data for every store in
[10:46] (646.64s)
Honey's database. The only problem is
[10:48] (648.80s)
collecting this data for one store is
[10:50] (650.56s)
easy, but doing it for 180,000, that
[10:53] (653.68s)
would literally take years. Unless, of
[10:56] (656.08s)
course, you created a little program
[10:57] (657.84s)
that automated the entire process,
[11:00] (660.16s)
crawling through that massive list,
[11:01] (661.84s)
scraping Honey's data store by store,
[11:04] (664.40s)
and neatly organizing the most relevant
[11:06] (666.48s)
data points into a nice, easy to read
[11:09] (669.28s)
spreadsheet. Well, that's exactly what
[11:11] (671.84s)
we have here. But I can't take any
[11:14] (674.24s)
credit for the spreadsheet because this
[11:15] (675.92s)
was all the work of someone much better
[11:17] (677.92s)
at this stuff than I am. And he goes by
[11:19] (679.92s)
the name of Yelta. He's a developer from
[11:21] (681.76s)
the Netherlands. awesome guy and I
[11:24] (684.08s)
cannot thank him enough because holy
[11:26] (686.56s)
[ __ ] there is a treasure trove of data
[11:29] (689.68s)
in this spreadsheet. We now have all the
[11:32] (692.40s)
details for every store, every coupon
[11:34] (694.48s)
code, every developer note, all curated
[11:36] (696.96s)
into one lovely spreadsheet. It's
[11:39] (699.52s)
glorious. PayPal will not like me having
[11:42] (702.08s)
this data, but too bad. And Yela will be
[11:44] (704.32s)
releasing this data publicly on Twitter
[11:46] (706.16s)
shortly following the release of this
[11:47] (707.92s)
video. Link to his Twitter is below, so
[11:50] (710.00s)
make sure you give him a follow if you
[11:51] (711.36s)
want to catch that drop. All right, so
[11:53] (713.28s)
now that we have the spreadsheet, we can
[11:54] (714.88s)
finally confirm that at the time this
[11:56] (716.96s)
data was collected, Honey supported
[11:59] (719.44s)
35,000 stores with affiliate links and
[12:02] (722.64s)
an additional 146,000 stores without
[12:07] (727.04s)
affiliate links. So yes, there was some
[12:09] (729.36s)
truth to their claim of supporting
[12:10] (730.80s)
30,000 participating stores. They just
[12:13] (733.84s)
conveniently left out the anyweeny fact
[12:16] (736.00s)
that they dragged an additional
[12:18] (738.72s)
146,000 stores onto their platform,
[12:21] (741.44s)
presumably without consent. Now,
[12:24] (744.00s)
consider this. If you're Honey and your
[12:26] (746.24s)
platform supports
[12:28] (748.48s)
180,000 stores, surely you'd be
[12:31] (751.44s)
screaming that from the rooftops, right?
[12:34] (754.32s)
180,000 sounds a whole lot more
[12:36] (756.56s)
impressive than 30,000. Honey knows
[12:39] (759.36s)
that. So, why hide the real number?
[12:41] (761.84s)
Clearly, Honey decided that keeping this
[12:44] (764.08s)
little secret to themselves outweighed
[12:46] (766.00s)
the benefits of advertising it. Now, as
[12:48] (768.40s)
a consumer, you might be thinking, "Ah,
[12:50] (770.64s)
so what if Honey leaks a few coupon
[12:52] (772.24s)
codes? Even if a business never gave
[12:54] (774.00s)
consent, surely that would just drive
[12:55] (775.92s)
more sales for the business, right?"
[12:58] (778.40s)
Well, it's not that simple. Why? Well,
[13:01] (781.52s)
you see, coupon codes are more than just
[13:03] (783.60s)
discounts for consumers. They're also a
[13:05] (785.84s)
vital marketing tool for businesses.
[13:08] (788.00s)
Virtually all coupon codes have a
[13:09] (789.84s)
strategic purpose behind them. For
[13:11] (791.68s)
example, many online stores will offer a
[13:14] (794.00s)
10% discount as an incentive for signing
[13:16] (796.32s)
up to their newsletter. The consumer
[13:18] (798.00s)
gets a 10% discount and the business
[13:20] (800.24s)
gets your email address for marketing.
[13:22] (802.32s)
It's a simple tit fortat exchange. But
[13:24] (804.80s)
if Honey snatches up that coupon and
[13:26] (806.72s)
makes it available to everyone, well,
[13:28] (808.72s)
now the business is losing 10% while
[13:30] (810.96s)
gaining nothing in return. The entire
[13:33] (813.44s)
purpose behind that coupon code gets
[13:34] (814.96s)
destroyed, stripping away any incentive
[13:37] (817.36s)
for the business to continue offering
[13:39] (819.12s)
it. But it's not just newsletter
[13:41] (821.04s)
coupons. Stores will also offer
[13:42] (822.64s)
discounts to reward and maintain
[13:44] (824.48s)
loyalty, like offering exclusive VIP
[13:46] (826.88s)
codes to repeat high-v valueue
[13:48] (828.64s)
customers. Many businesses will even
[13:50] (830.64s)
offer specialty discounts to military
[13:53] (833.12s)
veterans or first responders just to
[13:55] (835.52s)
show their appreciation. But also, some
[13:57] (837.76s)
businesses have private discount codes
[14:00] (840.40s)
like 50 or 80% off that are strictly
[14:03] (843.20s)
intended for use by employees or friends
[14:05] (845.60s)
and family. So, when Honey comes along,
[14:07] (847.92s)
finds all of these codes, and hands them
[14:09] (849.76s)
out like candy to its millions of users,
[14:12] (852.32s)
not only does it cause a negative
[14:14] (854.32s)
strategic impact, but it also causes
[14:16] (856.72s)
significant financial harm. In fact,
[14:19] (859.12s)
this has become such a significant
[14:20] (860.88s)
problem for online stores that there are
[14:22] (862.72s)
now multiple paid services to help
[14:25] (865.28s)
e-commerce stores block extensions like
[14:27] (867.84s)
Honey. Even well-known retailers are
[14:30] (870.16s)
paying to prevent these issues. Now, you
[14:32] (872.40s)
might be wondering, how on earth does
[14:33] (873.92s)
Honey get access to private coupon codes
[14:35] (875.76s)
like employee discounts if they're not
[14:37] (877.92s)
available to the public? Well, I came
[14:40] (880.08s)
across this marketing podcast where they
[14:42] (882.24s)
discussed the issue. What happens is if
[14:44] (884.48s)
I'm if I have Honey, for example, and
[14:47] (887.92s)
I'm one of your best customers at your
[14:49] (889.76s)
e-commerce store and you decide you're
[14:51] (891.92s)
going to do a promotion for your VIP
[14:53] (893.68s)
customers, maybe it's a flash sale, and
[14:56] (896.24s)
you send me a special code. I go to your
[14:58] (898.80s)
website, I legitimately have that code
[15:00] (900.72s)
from you. I type it into the promo code
[15:02] (902.96s)
field at checkout. Because I have Honey
[15:05] (905.28s)
in my browser, it is able to scrape that
[15:07] (907.76s)
code then and give it to everybody who
[15:10] (910.80s)
uses it. We've seen codes like military
[15:13] (913.12s)
hero 30 clearly intended for a veteran.
[15:17] (917.28s)
And then in another case,
[15:19] (919.28s)
>> Wow.
[15:19] (919.76s)
>> Even worse, in another case, my CEO
[15:22] (922.56s)
found a men's apparel brand where they
[15:25] (925.44s)
had a $75 off coupon code, but they
[15:28] (928.16s)
didn't set a minimum order value. And so
[15:31] (931.28s)
people were able to get unlimited
[15:33] (933.60s)
merchandise for free as long as they
[15:35] (935.68s)
kept their order value at $75 or under.
[15:38] (938.64s)
>> Okay, that's insane. But I have to
[15:40] (940.56s)
admit, at first I was a little skeptical
[15:42] (942.72s)
of the claim that Honey was snatching
[15:44] (944.40s)
codes from users without their consent
[15:46] (946.32s)
because whenever I entered in a coupon
[15:48] (948.00s)
code Honey didn't recognize, the
[15:49] (949.60s)
extension would always ask for my
[15:51] (951.20s)
permission before taking it. Or so I
[15:53] (953.36s)
thought. But as it turns out, the moment
[15:56] (956.00s)
you type in a coupon code, Honey
[15:58] (958.00s)
immediately sends that code directly to
[16:00] (960.32s)
their servers and then they ask for your
[16:03] (963.28s)
consent. How do I know? Well, you see, a
[16:05] (965.68s)
developer who would like to remain
[16:06] (966.96s)
anonymous noticed that Honey leaked
[16:08] (968.88s)
their source code inside of their iOS
[16:11] (971.36s)
app. Whoops. So, this developer
[16:14] (974.08s)
extracted the code and sent it to me.
[16:18] (978.08s)
We can actually further verify this with
[16:19] (979.84s)
a live example. Watch what Honey does in
[16:21] (981.84s)
the background as I enter in this
[16:23] (983.52s)
coupon. Honey realizes that it doesn't
[16:25] (985.76s)
have this code yet and asks for my
[16:27] (987.76s)
consent to share it. But look at this.
[16:29] (989.60s)
Before I've clicked anything, Honey has
[16:31] (991.76s)
already sent that coupon code to their
[16:33] (993.68s)
server, including the details of how
[16:35] (995.84s)
much money that coupon saved me, all
[16:37] (997.84s)
without my consent. Now, Honey might
[16:40] (1000.24s)
argue that if you click don't share,
[16:42] (1002.16s)
they never actually use those coupons.
[16:44] (1004.48s)
And maybe they don't. Maybe those codes
[16:46] (1006.96s)
just sit in their database doing
[16:48] (1008.64s)
nothing. We can't say for certain.
[16:50] (1010.56s)
However, if you read Honey's privacy
[16:52] (1012.64s)
policy, it explicitly states that they
[16:54] (1014.96s)
collect data such as coupons, promo
[16:57] (1017.12s)
codes, and deals you found. So, while
[16:59] (1019.44s)
they ask for a user's consent, their
[17:01] (1021.52s)
privacy policy clearly allows them to
[17:03] (1023.52s)
take that data without it. So, why does
[17:06] (1026.40s)
this matter? Well, it matters because
[17:08] (1028.40s)
when private codes get leaked,
[17:09] (1029.92s)
businesses don't just lose money,
[17:11] (1031.76s)
they're often left in the dark, unable
[17:13] (1033.76s)
to trace how those codes were exposed in
[17:16] (1036.48s)
the first place. I've personally found
[17:18] (1038.40s)
several coupon codes through Honey that
[17:20] (1040.24s)
were clearly intended for private use.
[17:22] (1042.72s)
This one coupon I found worked like a
[17:25] (1045.12s)
$35 gift voucher, meaning I can make
[17:27] (1047.60s)
unlimited free orders, provided my cart
[17:30] (1050.16s)
value was $35 or under. When I alerted
[17:33] (1053.44s)
the business, they appeared to have no
[17:35] (1055.52s)
idea how the code had been leaked. Now,
[17:38] (1058.72s)
believe it or not, this nightmare for
[17:40] (1060.32s)
online stores only gets worse. So, we
[17:44] (1064.72s)
usually use discount codes as a way to
[17:46] (1066.72s)
track when influencers or partners are
[17:50] (1070.32s)
sending traffic to us so that we can
[17:52] (1072.64s)
identify the sale was attributed to that
[17:55] (1075.04s)
influencer and pay them a commission.
[17:57] (1077.20s)
So, I think we saw one code started to
[17:59] (1079.12s)
get used like way more and we were like,
[18:01] (1081.36s)
"Wow, we're paying this influencer so
[18:03] (1083.12s)
much. They're doing so well." And I was
[18:05] (1085.28s)
like a little bit suspicious. A lot of
[18:07] (1087.60s)
times discount codes um are the only way
[18:09] (1089.84s)
for you to understand if a YouTuber that
[18:12] (1092.56s)
you love and would love to partner with
[18:14] (1094.48s)
is actually driving incremental value to
[18:16] (1096.96s)
your business. Those codes would
[18:18] (1098.48s)
immediately get picked up by Honey and
[18:19] (1099.92s)
then they'd be used hundreds of
[18:21] (1101.44s)
thousands of times and you'd be like,
[18:22] (1102.96s)
"All right, like not only does that
[18:24] (1104.32s)
negatively impact our business, it was
[18:25] (1105.76s)
for that audience, but um we have no
[18:28] (1108.64s)
idea which podcaster we should continue
[18:30] (1110.88s)
to support." Now, if you're someone who
[18:33] (1113.04s)
listens to podcasts, you probably know
[18:35] (1115.12s)
exactly what Chip is talking about here.
[18:37] (1117.04s)
When brands advertise on podcasts,
[18:38] (1118.88s)
affiliate links aren't always practical.
[18:41] (1121.20s)
Listeners could be running, driving, or
[18:43] (1123.20s)
doing the laundry without easy access to
[18:45] (1125.52s)
a computer. So, instead, brands will
[18:48] (1128.08s)
provide podcasters with a unique coupon
[18:50] (1130.48s)
code. And each time the listener uses
[18:52] (1132.24s)
that coupon code at checkout, the
[18:54] (1134.24s)
podcaster earns a commission just like
[18:56] (1136.80s)
an affiliate link. So, these are not
[18:58] (1138.80s)
your typical coupon codes. behind them.
[19:01] (1141.12s)
You have an influencer that's trying to
[19:02] (1142.56s)
earn a living. You have a business
[19:04] (1144.32s)
paying to promote their product. And
[19:05] (1145.84s)
there's usually a contract binding the
[19:07] (1147.84s)
two together. Regardless, Honey will
[19:10] (1150.00s)
step in, leak the codes to users,
[19:11] (1151.92s)
disrupting that business relationship.
[19:14] (1154.08s)
The result, not only do brands end up
[19:16] (1156.48s)
losing money from the leaked discounts,
[19:18] (1158.32s)
on top of that, they're now forced to
[19:20] (1160.24s)
pay undeserved commissions to the
[19:22] (1162.32s)
influencers whose codes got leaked. It's
[19:24] (1164.96s)
easy to imagine how quickly something
[19:26] (1166.64s)
like this could spiral out of control.
[19:28] (1168.72s)
In fact, one business owner on Twitter
[19:30] (1170.96s)
reported losing $100,000 from this exact
[19:34] (1174.96s)
issue, and he only realized it happening
[19:37] (1177.44s)
several months later. It's no laughing
[19:39] (1179.92s)
matter. And if you think this is a huge
[19:41] (1181.60s)
win for influencers because they get
[19:43] (1183.28s)
paid a bunch of undeserved commissions,
[19:45] (1185.60s)
well, think again.
[19:47] (1187.12s)
>> I mean, we basically pulled all out of
[19:48] (1188.88s)
the podcast realm because there is
[19:50] (1190.72s)
really no way to figure out how to how
[19:53] (1193.76s)
to attribute any successes or failures.
[19:56] (1196.56s)
We basically can't do that anymore
[19:59] (1199.20s)
because as soon as a code becomes used
[20:02] (1202.24s)
by one person, it's suddenly
[20:05] (1205.12s)
used by everybody. This is a nightmare
[20:08] (1208.56s)
situation for everyone involved. But not
[20:11] (1211.36s)
all businesses are affected equally.
[20:13] (1213.20s)
Honey's actions are especially
[20:15] (1215.20s)
devastating for small businesses that
[20:17] (1217.20s)
lack the resources of larger brands. The
[20:19] (1219.52s)
big retailers can push back. They have
[20:21] (1221.52s)
dedicated marketing teams, advanced
[20:23] (1223.44s)
analytic tools, and the legal muscle to
[20:25] (1225.60s)
respond. Smaller businesses, on the
[20:27] (1227.52s)
other hand, don't have that luxury. So,
[20:29] (1229.68s)
when a multi-billion dollar company like
[20:31] (1231.76s)
PayPal piles on more obstacles, it's not
[20:34] (1234.24s)
just frustrating, it's downright
[20:36] (1236.56s)
predatory. In fact, Maiden Cookware CEO
[20:39] (1239.04s)
Chip perfectly summed up the devastating
[20:40] (1240.96s)
impact Honey's actions can have on a
[20:43] (1243.28s)
small business.
[20:44] (1244.40s)
>> When you're in the early stages of
[20:46] (1246.00s)
building a business, you're in the stage
[20:47] (1247.76s)
of will this company work? is their
[20:50] (1250.08s)
product market fit. How do I scale this
[20:52] (1252.16s)
thing? Can the economics work? Right?
[20:54] (1254.08s)
And so all the things we talked about
[20:55] (1255.68s)
these discount codes and the kind of
[20:57] (1257.28s)
removal of attribution and killing
[20:58] (1258.72s)
margin like those really affect the
[21:00] (1260.56s)
early stages of a company. Um from
[21:02] (1262.80s)
everything from where do I put the next
[21:04] (1264.08s)
dollar in to make sure I can survive
[21:05] (1265.76s)
down to um you know you're taking 10 15
[21:08] (1268.64s)
20% of my sales without me wanting it.
[21:11] (1271.20s)
And we have digital marketing costs, we
[21:13] (1273.52s)
have shipping costs, we have all these
[21:15] (1275.44s)
stacks of costs that not only are
[21:17] (1277.76s)
getting more inflated over time, like
[21:18] (1278.96s)
FedEx has been raising their rates over
[21:20] (1280.64s)
time. Um, warehousing costs have gone
[21:22] (1282.40s)
up, cogs have gone up, as you know,
[21:23] (1283.84s)
through inflation for the last four or
[21:25] (1285.36s)
five years, right? Like all the cost
[21:27] (1287.28s)
structures have been going up and to
[21:28] (1288.80s)
have someone just stealing and
[21:30] (1290.72s)
compressing the actual revenue you want
[21:32] (1292.80s)
like puts small businesses in a real
[21:34] (1294.80s)
squeeze and it's just really
[21:36] (1296.32s)
disappointing. And like I it doesn't
[21:38] (1298.00s)
even work because at the end of the day
[21:39] (1299.12s)
like you need to make the numbers work.
[21:40] (1300.64s)
And if Honey is going to steal 10% of
[21:42] (1302.48s)
your revenue all the time, you're going
[21:44] (1304.24s)
to have to raise prices at the end of
[21:45] (1305.52s)
the day to make up for that, right? And
[21:47] (1307.12s)
so it's not only not effective, but it
[21:50] (1310.48s)
makes it worse for the consumer, it
[21:52] (1312.32s)
compresses for the small business owner
[21:53] (1313.76s)
any margin. And it's it's just like a
[21:56] (1316.00s)
really awful business model. Like we've
[21:57] (1317.28s)
been so mad at them for so long.
[22:00] (1320.08s)
>> Really sad. And Chip's point about the
[22:01] (1321.76s)
impact to consumers is especially
[22:03] (1323.52s)
important. If honey's eroding profit
[22:05] (1325.12s)
margins from retailers and they're
[22:06] (1326.48s)
forced to increase prices, well, that
[22:08] (1328.72s)
ultimately affects consumers. Now, as
[22:10] (1330.88s)
you'd expect, these businesses don't
[22:12] (1332.80s)
just let these issues slide. Naturally,
[22:14] (1334.88s)
they reach out to Honey to complain,
[22:16] (1336.72s)
asking to be removed from the platform,
[22:19] (1339.04s)
and that unfortunately is where the
[22:21] (1341.20s)
story takes a much darker turn. You see,
[22:24] (1344.32s)
Honey is well aware of the inconvenience
[22:26] (1346.72s)
it creates for these online stores. And
[22:28] (1348.80s)
so when they inevitably reach out to
[22:30] (1350.56s)
Honey asking to be removed, does Honey
[22:33] (1353.04s)
oblige? Of course not. Instead, Honey
[22:36] (1356.48s)
goes, "How about this? We'll give you
[22:39] (1359.12s)
full control over which coupons go live
[22:41] (1361.20s)
on our platform, but only if you partner
[22:44] (1364.00s)
with us. I kid you not. In fact, Chip
[22:47] (1367.52s)
was kind enough to share his email
[22:48] (1368.96s)
exchanges with Honey from 2020. And you
[22:51] (1371.76s)
need to see it because holy [ __ ] it's
[22:54] (1374.88s)
bad." Now, it is a long email chain, so
[22:57] (1377.20s)
I'm not going to read the entire thing,
[22:58] (1378.72s)
but you're welcome to pause to read the
[23:00] (1380.80s)
bits I skim over. Hi, I am the CEO of
[23:03] (1383.44s)
Med and Cookware. Please remove us from
[23:04] (1384.96s)
your app. You've scraped a private
[23:06] (1386.32s)
friends and family code from our
[23:07] (1387.52s)
checkout and put it on the platform for
[23:09] (1389.04s)
others to use. We've lost a bunch of
[23:10] (1390.88s)
revenue. 4 days later, Honey responds
[23:13] (1393.12s)
with, "Hi, Chip. Thanks for reaching
[23:15] (1395.12s)
out. Honey supports over 40,000 stores
[23:17] (1397.12s)
online, and we always prioritize
[23:18] (1398.88s)
protecting the Honey experience for
[23:20] (1400.32s)
users by supporting available stores and
[23:22] (1402.56s)
displaying available codes. We can
[23:24] (1404.72s)
absolutely remove the code in question.
[23:26] (1406.40s)
In order to protect the Honey experience
[23:27] (1407.84s)
for our users, we typically do not
[23:30] (1410.00s)
remove codes unless we have a working
[23:32] (1412.48s)
relationship.
[23:34] (1414.00s)
I mean, right off the bat, they're
[23:35] (1415.68s)
basically saying, "We won't compromise
[23:37] (1417.44s)
the experience for our users unless you
[23:39] (1419.84s)
pay us." Classy. We'd love to discuss
[23:42] (1422.56s)
how we can work more closely and partner
[23:44] (1424.64s)
with your brand. Chip replies the same
[23:46] (1426.64s)
day. We don't offer affiliate deals to
[23:48] (1428.40s)
coupon sites. We'd like to be removed
[23:50] (1430.00s)
from your site an extension completely.
[23:52] (1432.24s)
With the sales pitch not going to plan,
[23:54] (1434.08s)
Honey brings in their global
[23:55] (1435.44s)
partnerships manager, Kelly Roodec, who
[23:57] (1437.52s)
explains to Chip that his leak code was
[23:59] (1439.76s)
added through Honey's user generated
[24:01] (1441.76s)
coupon functionality. No surprises
[24:04] (1444.08s)
there. She completely ignores his
[24:06] (1446.00s)
request to be removed from the platform,
[24:07] (1447.60s)
only offering to remove the leaked
[24:09] (1449.36s)
employee discount. Chip replies the same
[24:11] (1451.52s)
day, "A friends and family code was
[24:13] (1453.84s)
listed on your site. That was an
[24:15] (1455.28s)
internal code meant for a small group of
[24:17] (1457.04s)
people that was never published
[24:18] (1458.48s)
anywhere. A whole site can't be removed
[24:20] (1460.48s)
from your app. Honey only responds a
[24:22] (1462.48s)
week later, confirming they removed the
[24:24] (1464.08s)
employee discount, but again completely
[24:26] (1466.40s)
ignores Chip's request to be removed
[24:28] (1468.40s)
from Honey. Chip responds immediately
[24:30] (1470.48s)
pushing for an answer, but they ignore
[24:32] (1472.72s)
his email completely. A month later and
[24:35] (1475.20s)
Honey leaks another private discount
[24:37] (1477.76s)
code. Hi, please remove from your app.
[24:40] (1480.88s)
That is a private code behind a private
[24:42] (1482.80s)
employee perk login. This should never
[24:45] (1485.20s)
be public. I don't understand how you
[24:47] (1487.36s)
feel taking a private employee perk and
[24:49] (1489.44s)
making that public to the world is
[24:51] (1491.28s)
helping our business. Honey responds
[24:53] (1493.44s)
confirming they removed the code, but
[24:55] (1495.20s)
instead of taking responsibility, they
[24:57] (1497.52s)
try shifting the blame onto Chip,
[24:59] (1499.44s)
suggesting his employees might be
[25:01] (1501.44s)
responsible for sharing the code with
[25:03] (1503.04s)
Honey. But given what we know about
[25:05] (1505.44s)
their system, I'm willing to bet Honey
[25:07] (1507.52s)
were responsible for the leak, not
[25:09] (1509.36s)
Chip's employees. Honey then suggests
[25:12] (1512.00s)
that Chip creates single-use codes for
[25:14] (1514.32s)
each and every employee, essentially
[25:16] (1516.48s)
making it his responsibility to fix a
[25:18] (1518.88s)
problem they created in the first place.
[25:21] (1521.60s)
I would be fuming if this were my
[25:23] (1523.76s)
business. 2 months later and another
[25:26] (1526.08s)
private code gets leaked and once again,
[25:28] (1528.40s)
Chip asks to be completely removed from
[25:30] (1530.24s)
the platform and this time Honey finally
[25:32] (1532.80s)
addresses the question. We proudly host
[25:34] (1534.96s)
a consistent shopping experience for all
[25:36] (1536.56s)
Honey shoppers who rely on our shopping
[25:38] (1538.64s)
tools. Therefore, we cannot disable
[25:40] (1540.72s)
Honey for individual stores and never
[25:43] (1543.12s)
have. Never have, huh? Well, that's
[25:45] (1545.92s)
interesting because it just so happens
[25:47] (1547.52s)
that Andrew from Truget Texture Supply
[25:49] (1549.84s)
also shared his email exchange with
[25:51] (1551.68s)
Honey. And would you believe it, he was
[25:54] (1554.00s)
dealing with the exact same Honey
[25:56] (1556.00s)
employee as Chip. And while Honey
[25:58] (1558.00s)
attempted the same sales pitch with
[25:59] (1559.76s)
Andrew, they did eventually remove his
[26:02] (1562.24s)
store from the platform. So, yes, it is
[26:04] (1564.88s)
possible. And yes, they have done it
[26:07] (1567.28s)
before. Once again, we've caught Honey
[26:09] (1569.60s)
in yet another calculated and deliberate
[26:12] (1572.56s)
light. And make no mistake, it was
[26:14] (1574.56s)
deliberate because this employee, Kelly,
[26:17] (1577.12s)
confirmed the removal of Andrew store
[26:19] (1579.12s)
just one month before assuring Chip that
[26:21] (1581.44s)
Honey had never removed a store before.
[26:24] (1584.40s)
Honey clearly has the capability to
[26:26] (1586.48s)
remove stores. They just choose to
[26:28] (1588.16s)
enforce that decision selectively. Keep
[26:30] (1590.24s)
in mind, Honey knows exactly how many of
[26:32] (1592.80s)
their users are visiting a given store
[26:35] (1595.04s)
and how much they are spending. So, I'd
[26:37] (1597.84s)
speculate that this selective
[26:39] (1599.28s)
enforcement comes entirely down to which
[26:41] (1601.60s)
stores Honey believes will make them the
[26:43] (1603.92s)
most money. Now, for the record, I
[26:46] (1606.00s)
didn't even get through that entire
[26:47] (1607.44s)
email chain between Chip and Honey. It
[26:49] (1609.52s)
goes on and on. And this behavior is not
[26:52] (1612.72s)
an isolated incident. It's a pattern.
[26:55] (1615.76s)
When merchants discover their codes have
[26:58] (1618.00s)
leaked and then they want to clamp down
[27:00] (1620.72s)
on that, they go to these coupon
[27:02] (1622.88s)
extensions and plea and request to have
[27:05] (1625.60s)
their codes removed. And nine times out
[27:07] (1627.92s)
of 10, the response is join our
[27:10] (1630.32s)
affiliate program and you'll have more
[27:12] (1632.00s)
granular control over that. And so I
[27:14] (1634.40s)
think merchants tend to feel a little
[27:16] (1636.00s)
bit as though they're being blackmailed
[27:19] (1639.04s)
or extorted, you know, because you join
[27:20] (1640.80s)
the affiliate program and then you have
[27:22] (1642.24s)
to pay that coupon extension company
[27:24] (1644.72s)
every time somebody uses a code
[27:26] (1646.96s)
>> in and you're doing that in order to
[27:28] (1648.80s)
have them not share your code.
[27:31] (1651.04s)
>> I mean, this has got to be one of the
[27:33] (1653.60s)
most [ __ ] up business models I have
[27:35] (1655.68s)
ever seen. And let's be real about
[27:37] (1657.36s)
what's going on here. In my opinion,
[27:39] (1659.52s)
it's economic extortion. Just think
[27:41] (1661.60s)
about it. Imagine one day deciding to
[27:43] (1663.52s)
start your own business. You've saved up
[27:45] (1665.36s)
all this money and invested it all into
[27:47] (1667.36s)
this one big idea. You're taking on a
[27:49] (1669.52s)
huge risk. And as you navigate the daily
[27:51] (1671.68s)
struggles of running a business, a bad
[27:53] (1673.44s)
actor sneaks into your store without you
[27:55] (1675.60s)
ever noticing. They quietly start
[27:57] (1677.28s)
collecting coupons from your customers
[27:58] (1678.88s)
and start handing them out at your
[28:00] (1680.48s)
checkout counter. Then one day, you
[28:02] (1682.56s)
notice a customer with a voucher that
[28:04] (1684.48s)
was clearly never intended for them, and
[28:06] (1686.64s)
you have absolutely no idea how they got
[28:08] (1688.88s)
it. But when you eventually catch the
[28:10] (1690.72s)
culprit and demand that they leave your
[28:12] (1692.32s)
store, they refuse and there's nothing
[28:14] (1694.48s)
you can do about it. Knowing this, the
[28:16] (1696.40s)
bad actor leverages the harm they
[28:18] (1698.16s)
manufactured by strongarmming you into
[28:20] (1700.16s)
signing a partnership deal where the
[28:21] (1701.84s)
only way to stop the damage they created
[28:24] (1704.16s)
is by paying them. Honey's business
[28:26] (1706.08s)
model starts to make a lot more sense
[28:28] (1708.24s)
now, doesn't it? It makes sense why they
[28:30] (1710.56s)
dragged 146,000 stores onto their
[28:33] (1713.52s)
platform and kept quiet about it.
[28:35] (1715.52s)
because by doing so, they wedge
[28:37] (1717.36s)
themselves between those stores and
[28:38] (1718.96s)
their customers, disrupting,
[28:40] (1720.40s)
interfering, and ultimately taking
[28:42] (1722.64s)
control over their coupon strategies.
[28:44] (1724.64s)
And the only way for those stores to
[28:46] (1726.24s)
regain some of that control is by paying
[28:48] (1728.64s)
Honey. I mean, why else would any of
[28:51] (1731.12s)
these stores partner with and pay Honey
[28:53] (1733.76s)
if they're already on the platform for
[28:55] (1735.68s)
free? The answer is simple. The stores
[28:58] (1738.08s)
don't pay for inclusion, they pay for
[29:00] (1740.48s)
exclusion, for damage control. It makes
[29:03] (1743.28s)
sense. Now, to be clear, I'm not
[29:05] (1745.12s)
suggesting that all of Honey's 35,000
[29:07] (1747.20s)
partnered stores were coerced into
[29:09] (1749.04s)
joining their platform. I'm sure many
[29:10] (1750.80s)
signed up willingly. After all, Honey
[29:12] (1752.88s)
does offer a cashback program, which
[29:14] (1754.56s)
some stores genuinely like to offer
[29:16] (1756.32s)
users. And just look at the glowing case
[29:18] (1758.72s)
studies on Honey's website. Some of the
[29:20] (1760.72s)
numbers they throw around make
[29:22] (1762.16s)
partnering with Honey look like an
[29:23] (1763.84s)
absolute no-brainer. Increased average
[29:25] (1765.92s)
order value, reduced card abandonment.
[29:27] (1767.92s)
That's music to the ears of any
[29:29] (1769.52s)
e-commerce store owner. But of course,
[29:31] (1771.20s)
if the benefits were truly as good as
[29:33] (1773.20s)
they say, you'd expect that brands who
[29:35] (1775.12s)
partnered with Honey never look back,
[29:37] (1777.52s)
right? Well, here's a little statistic
[29:39] (1779.44s)
Honey doesn't want brands to see. And
[29:41] (1781.60s)
it's from their own data. This right
[29:43] (1783.36s)
here is a list of stores that at some
[29:45] (1785.44s)
point tried a partnership with Honey,
[29:47] (1787.36s)
then decided, "Yeah, no thanks." and
[29:50] (1790.24s)
ended the partnership. There's 15,000 of
[29:53] (1793.92s)
them. That's 15,000 brands that at some
[29:57] (1797.60s)
point were partnered with Honey but no
[30:00] (1800.08s)
longer are. And this is based on the
[30:02] (1802.00s)
limited data I have access to. The
[30:04] (1804.16s)
actual number is potentially a lot
[30:06] (1806.72s)
higher. I think that says a lot more
[30:08] (1808.64s)
about the value Honey can bring to
[30:10] (1810.08s)
businesses than any of their case
[30:12] (1812.16s)
studies.
[30:14] (1814.16s)
Given everything we've uncovered about
[30:16] (1816.24s)
Honey so far, it really makes you wonder
[30:18] (1818.88s)
why on earth would a Fortune 500 company
[30:21] (1821.04s)
like PayPal sink $4 billion into buying
[30:24] (1824.56s)
this company. Sure, Honey was making
[30:26] (1826.32s)
some sweet affiliate revenue. But still,
[30:28] (1828.80s)
that's a lot of money. To put that price
[30:30] (1830.96s)
tag into perspective, the platform
[30:32] (1832.72s)
you're watching this on, YouTube, was
[30:34] (1834.32s)
acquired for just $1.6 billion. Combine
[30:37] (1837.52s)
that with the acquisitions of Instagram
[30:39] (1839.28s)
and Twitch, and it still equates to less
[30:41] (1841.36s)
than what PayPal paid for Honey, a
[30:44] (1844.16s)
coupon browser extension. Well, Honey
[30:46] (1846.88s)
brought more to the table for PayPal
[30:48] (1848.48s)
than its ability to churn out mountains
[30:50] (1850.32s)
of cash. They also had 17 million users
[30:53] (1853.68s)
worth of data. And as it turns out, lots
[30:56] (1856.48s)
of it. Remember in the last episode how
[30:58] (1858.48s)
I said, "If a product's free, it's
[31:00] (1860.24s)
likely you're the product." Well, we're
[31:02] (1862.32s)
at that part of the story now. You see,
[31:04] (1864.32s)
while Honey was handing out bottom of
[31:05] (1865.84s)
the barrel coupon codes, they were also
[31:07] (1867.84s)
tracking your shopping habits and
[31:09] (1869.36s)
collecting a ton of data along the way.
[31:11] (1871.92s)
In fact, Amazon of all companies began
[31:14] (1874.40s)
warning users that Honey was a security
[31:16] (1876.48s)
risk, stating, "Honey tracks your
[31:18] (1878.16s)
private shopping behavior, collects data
[31:19] (1879.84s)
like your order history and items saved,
[31:21] (1881.44s)
and can read and change any of your data
[31:23] (1883.28s)
on any website you visit." They even
[31:25] (1885.28s)
recommended uninstalling the extension.
[31:27] (1887.28s)
That was somewhat surprising to me given
[31:29] (1889.28s)
how much Honey played into the whole you
[31:31] (1891.12s)
can trust us with your data narrative.
[31:33] (1893.20s)
actually free, not like selling all your
[31:35] (1895.60s)
personal data free.
[31:36] (1896.72s)
>> Does Honey sell user data?
[31:39] (1899.92s)
>> No. True.
[31:42] (1902.16s)
>> This prompted me to investigate Honey's
[31:44] (1904.32s)
data collection practices. And what I
[31:46] (1906.24s)
found was, unsurprisingly, at this
[31:48] (1908.48s)
point, pretty darn concerning.
[31:50] (1910.56s)
Hilariously, the first thing I noticed
[31:52] (1912.24s)
when auditing Honey's website was that
[31:54] (1914.08s)
they explicitly assured users that, hey,
[31:56] (1916.48s)
what we don't do is sell or share your
[31:58] (1918.56s)
data. But the moment you click the link
[32:00] (1920.32s)
to their privacy policy, the first thing
[32:02] (1922.24s)
you see is how we share your data. Uh,
[32:05] (1925.68s)
excuse me, red flag number one. Now,
[32:08] (1928.88s)
this privacy policy was packed with
[32:10] (1930.56s)
vague language when it came to what data
[32:12] (1932.48s)
they were collecting and how they were
[32:14] (1934.08s)
sharing it. So, I dug deeper and came
[32:16] (1936.32s)
across this incredible investigation by
[32:18] (1938.72s)
a German nonprofit called Data Request,
[32:21] (1941.84s)
who advocate for data privacy online.
[32:24] (1944.00s)
and they dug into Honey's data
[32:25] (1945.52s)
collection practices by submitting
[32:27] (1947.12s)
what's known as a GDPR right of access
[32:30] (1950.24s)
request. For those unfamiliar, under
[32:32] (1952.08s)
European law, anyone living in Europe,
[32:34] (1954.08s)
can ask a company, no matter where it's
[32:35] (1955.76s)
located, for a copy of all the personal
[32:37] (1957.92s)
data that company has collected on them.
[32:40] (1960.16s)
So, two of their members, Benny and
[32:41] (1961.84s)
Malta, both Honey users, each submitted
[32:44] (1964.24s)
a GDPR request. Their data revealed that
[32:46] (1966.88s)
Honey was systematically tracking their
[32:48] (1968.88s)
browsing activity across every website
[32:51] (1971.52s)
it considered to be an online store.
[32:54] (1974.16s)
They also collected timestamps, unique
[32:56] (1976.16s)
user IDs, device IDs, operating system
[32:58] (1978.64s)
info, geoloccation details, and the full
[33:01] (1981.20s)
URL of the page visited. Now, data
[33:04] (1984.08s)
requests showed that from these logs
[33:05] (1985.92s)
alone, Honey could infer some incredibly
[33:08] (1988.48s)
personal insights about a user. Here are
[33:11] (1991.28s)
just a few examples from Benny's data
[33:13] (1993.20s)
that was collected in 2020. Honey could
[33:16] (1996.16s)
see that on February 13th at 2:57 p.m.
[33:19] (1999.12s)
Benny viewed an Iix guide on how to swap
[33:21] (2001.28s)
the DVD lens on a Nintendo Wii. They
[33:23] (2003.84s)
could see that Benny checked an
[33:24] (2004.96s)
AliExpress order 13 times, his order ID
[33:27] (2007.84s)
fully visible, including the fact he
[33:29] (2009.68s)
opened a dispute for the order. They
[33:31] (2011.76s)
could see he had a Microsoft family plan
[33:33] (2013.60s)
and that he added a new family member to
[33:35] (2015.76s)
his Office 365 account. Honey also knew
[33:38] (2018.64s)
that Benny looked for an Airbnb in
[33:40] (2020.24s)
Berlin for two adults from the 4th to
[33:42] (2022.56s)
the 5th of March. And apparently Benny
[33:44] (2024.96s)
had issues with his iPhone because he
[33:46] (2026.80s)
viewed an Apple support page on how to
[33:48] (2028.96s)
reset his passcode. On March 23rd at
[33:51] (2031.44s)
around 5:00 p.m., Benny watched a
[33:53] (2033.36s)
documentary called Scanning the Pyramids
[33:55] (2035.60s)
on Curiosity Stream, a service he
[33:57] (2037.76s)
subscribed to just an hour earlier
[33:59] (2039.44s)
through YouTuber Tom Scott's affiliate
[34:01] (2041.68s)
link. And apparently Benny's a gamer
[34:03] (2043.52s)
because he redeemed a game on Steam with
[34:05] (2045.84s)
the serial code 5HGP6.
[34:09] (2049.36s)
You get the point. All that information
[34:12] (2052.40s)
could be inferred from just 27 page
[34:14] (2054.80s)
views. But in total, Honey collected
[34:17] (2057.04s)
over 2 1/2,000 pages of web activity
[34:20] (2060.24s)
between just February and May of 2020.
[34:23] (2063.20s)
Meaning what I just read to you
[34:24] (2064.72s)
represents only 1% of the total data
[34:27] (2067.36s)
Honey collected within a 3month window.
[34:30] (2070.32s)
That is insane. Now, while Benny
[34:32] (2072.32s)
willingly registered an account with
[34:33] (2073.68s)
Honey, Malta, on the other hand, didn't.
[34:36] (2076.24s)
Yet, Honey collected the same data from
[34:38] (2078.24s)
him as well, suggesting whether you
[34:40] (2080.32s)
formally register an account with Honey
[34:42] (2082.00s)
or simply installed the extension. Honey
[34:44] (2084.24s)
was harvesting the same data regardless.
[34:47] (2087.04s)
Huge shout out to Data Request for their
[34:48] (2088.64s)
work. I'll link their full investigation
[34:50] (2090.08s)
below. They also have a great tool where
[34:52] (2092.64s)
eligible consumers can easily request a
[34:54] (2094.88s)
copy of their data from companies like
[34:56] (2096.56s)
Honey and also request to have that data
[34:59] (2099.12s)
deleted. definitely check that out and
[35:01] (2101.04s)
consider supporting their nonprofit. So,
[35:03] (2103.20s)
as you can see, it starts to make a lot
[35:05] (2105.12s)
more sense why a company like PayPal
[35:07] (2107.04s)
would fork out so much money for this
[35:09] (2109.04s)
platform. Honey was sitting on a
[35:10] (2110.96s)
treasure trove of data that provided
[35:13] (2113.12s)
detailed insights into its users
[35:14] (2114.88s)
shopping habits. Honey could see what
[35:16] (2116.96s)
users were buying, how they shopped,
[35:18] (2118.96s)
what they searched for, the decisions
[35:20] (2120.56s)
they made before checkout, and even the
[35:22] (2122.32s)
services they considered but didn't
[35:24] (2124.16s)
purchase. That information is invaluable
[35:26] (2126.96s)
to a payments platform like PayPal. But
[35:29] (2129.36s)
don't just take my word for it. Here's
[35:31] (2131.04s)
Honey's former senior manager of
[35:32] (2132.40s)
partnerships, Daniel Pilington,
[35:34] (2134.24s)
discussing the benefits of Honey Stata
[35:36] (2136.24s)
for their partnered merchants. You know,
[35:38] (2138.24s)
one of the metrics that we assess very
[35:39] (2139.92s)
frequently for merchants that work with
[35:41] (2141.28s)
Honey is uh, you know, cross shopping.
[35:43] (2143.28s)
So, you know, how many other stores is a
[35:45] (2145.76s)
user visiting as well as the store in
[35:47] (2147.60s)
question? You know, are they going to
[35:48] (2148.88s)
six stores? Are they going to seven
[35:50] (2150.00s)
stores? Are they going to 10 stores uh
[35:51] (2151.92s)
before they make a decision on where to
[35:53] (2153.36s)
purchase?
[35:54] (2154.00s)
>> And here's another former senior
[35:55] (2155.36s)
partnerships manager. So, like one of
[35:57] (2157.04s)
the things we have insight into at Honey
[35:58] (2158.48s)
because we're a browser extension and we
[36:00] (2160.16s)
can see our shoppers and and how they
[36:02] (2162.40s)
shop is that shoppers love to
[36:04] (2164.32s)
cross-sight comparison shop. You know,
[36:06] (2166.32s)
we're we're really kind of following the
[36:08] (2168.56s)
shopper where they go and we're with
[36:10] (2170.56s)
them every step of the way. We have all
[36:12] (2172.80s)
sorts of tools to make those consumers
[36:16] (2176.08s)
stickier to those direct to consumer
[36:17] (2177.76s)
brands.
[36:18] (2178.40s)
>> And here's a former PayPal executive
[36:20] (2180.32s)
discussing how beneficial the data is
[36:22] (2182.48s)
for their goals. We've got a full suite
[36:25] (2185.12s)
of solutions across our consumer
[36:26] (2186.96s)
platforms that help drive new customers
[36:30] (2190.08s)
that help drive loyalty, that help drive
[36:32] (2192.00s)
conversion, um, and ultimately sales
[36:34] (2194.96s)
lift and and we have a lot of analytics
[36:36] (2196.96s)
and a lot of data, uh, a lot of shopper
[36:39] (2199.20s)
data obviously, and that's something
[36:40] (2200.56s)
that we can unlock on behalf of our
[36:42] (2202.88s)
partners.
[36:43] (2203.68s)
>> You've got to hand it to Honey. They've
[36:45] (2205.52s)
truly mastered the art of selling
[36:47] (2207.36s)
[ __ ] narratives to users. They
[36:49] (2209.68s)
promised privacy, insisting, "Hey, we
[36:52] (2212.16s)
never sell or share your data." But to
[36:54] (2214.24s)
their merchants, they literally bragged
[36:56] (2216.48s)
about how much of your data they had and
[36:58] (2218.64s)
how it could be leveraged to provide
[37:00] (2220.40s)
them with valuable insights into your
[37:02] (2222.16s)
shopping habits. It's genuinely
[37:04] (2224.96s)
astounding. Now, for clarity, those
[37:06] (2226.88s)
interviews were all recorded after
[37:08] (2228.72s)
PayPal's acquisition of Honey. So, is
[37:11] (2231.36s)
this simply a case of a corporate giant
[37:13] (2233.28s)
turning a once well-intentioned startup
[37:15] (2235.76s)
into a data harvesting machine? Nope. In
[37:18] (2238.88s)
fact, data collection was part of
[37:20] (2240.80s)
Honey's game plan well before they were
[37:22] (2242.96s)
acquired by PayPal. This right here is
[37:24] (2244.88s)
Honey's pitch deck to investors from
[37:26] (2246.80s)
back in 2015. On the business model
[37:29] (2249.36s)
slide, Honey clearly labels personalized
[37:31] (2251.92s)
offers to consumers based on cross-sight
[37:34] (2254.16s)
comparison shopping data as a core
[37:36] (2256.16s)
component of their business strategy.
[37:37] (2257.92s)
And on the very next slide, proudly
[37:40] (2260.00s)
entitled our unfair advantage, it
[37:42] (2262.48s)
explicitly states, "Honey's unique data
[37:44] (2264.80s)
allows us to predict what each user is
[37:46] (2266.88s)
about to buy, when they intend to
[37:48] (2268.64s)
purchase, and how much they are willing
[37:50] (2270.88s)
to pay." They even itemized the data
[37:52] (2272.96s)
collected as user behavioral data,
[37:55] (2275.68s)
stores visited, products viewed, and of
[37:58] (2278.24s)
course, your purchase history. So, not
[38:01] (2281.20s)
only was data collection top of mind
[38:03] (2283.12s)
from the very beginning, but they were
[38:05] (2285.04s)
already actively collecting it and
[38:06] (2286.96s)
pitching that data to investors as a
[38:09] (2289.28s)
valuable asset. For years, Honey framed
[38:12] (2292.00s)
data collection as minimal and purely in
[38:14] (2294.40s)
service of saving users money. Well, it
[38:17] (2297.44s)
appears they were full of [ __ ] And to
[38:20] (2300.00s)
be clear, I'm not suggesting that Honey
[38:21] (2301.92s)
was selling your data to third parties
[38:23] (2303.52s)
for cash. Well, at least I found no
[38:26] (2306.16s)
evidence of that. But were they quietly
[38:28] (2308.32s)
monetizing your data behind the scenes,
[38:30] (2310.16s)
sharing insights with partnered
[38:31] (2311.92s)
merchants while conveniently leaving
[38:34] (2314.00s)
that detail out of their privacy policy?
[38:36] (2316.24s)
Well, it sure looks that way to me. One
[38:39] (2319.12s)
thing is for certain, however, and
[38:40] (2320.88s)
that's that Honey sold your data to
[38:42] (2322.80s)
PayPal, one of the largest online
[38:44] (2324.72s)
payment platforms in the world. PayPal
[38:47] (2327.04s)
didn't spend billions of dollars for a
[38:49] (2329.20s)
simple coupon extension. They were
[38:51] (2331.04s)
buying a window into your life as a
[38:53] (2333.12s)
consumer. And here's why that should
[38:55] (2335.12s)
worry you. PayPal launches an ad
[38:57] (2337.28s)
network. PayPal has official that it
[38:59] (2339.20s)
will be launching an ad network that
[39:00] (2340.88s)
will sell ads, leveraging, oh no.
[39:03] (2343.44s)
Leveraging the data it collects on the
[39:05] (2345.36s)
purchase history and spending of its 400
[39:08] (2348.32s)
million users. This change would also
[39:10] (2350.64s)
likely affect users of PayPal
[39:12] (2352.40s)
subsidiaries like Venmo and Honey. And I
[39:15] (2355.20s)
believe there's more as well, but I'm
[39:16] (2356.80s)
not surprised Honey's in there.
[39:19] (2359.28s)
>> This is like terrible. Couldn't have
[39:21] (2361.68s)
said it better myself. Now, to make this
[39:24] (2364.56s)
whole data collection issue a 100 times
[39:26] (2366.72s)
worse, something else caught my eye
[39:28] (2368.72s)
while reading through Honey's privacy
[39:30] (2370.40s)
policy. Right at the very bottom, Honey
[39:32] (2372.64s)
states that quote, "We created Honey for
[39:34] (2374.88s)
the exclusive use of adults 18 and
[39:37] (2377.20s)
older, and we don't knowingly collect or
[39:39] (2379.76s)
solicit personal information from
[39:41] (2381.76s)
children." Yeah. You see, now there's
[39:44] (2384.16s)
just one problem with that. I have a
[39:46] (2386.88s)
challenge for all of you. Go to every
[39:48] (2388.72s)
computer in your house, your mom's, your
[39:50] (2390.24s)
dad's, your sister, your brother's
[39:51] (2391.60s)
computer, and install Honey. That's
[39:53] (2393.36s)
right. Not only did Honey and PayPal
[39:55] (2395.60s)
knowingly collect data from miners, but
[39:57] (2397.44s)
they intentionally targeted them in
[39:59] (2399.28s)
their advertising. That's not just
[40:00] (2400.72s)
unethical, it's potentially illegal in
[40:03] (2403.28s)
numerous jurisdictions. Many countries
[40:05] (2405.76s)
have very strict regulations when it
[40:07] (2407.84s)
comes to the collection of personal data
[40:09] (2409.60s)
from minor and for good reason.
[40:11] (2411.52s)
Therefore, if you intend on having
[40:13] (2413.20s)
miners use your product, many
[40:14] (2414.64s)
jurisdictions require that you obtain
[40:16] (2416.64s)
parental consent first. But this adds
[40:19] (2419.12s)
all kinds of technical and legal
[40:20] (2420.96s)
complexities, which is why many
[40:22] (2422.56s)
platforms simply limit the use of their
[40:24] (2424.72s)
service to adults only in their privacy
[40:27] (2427.28s)
policy, which was Honey's approach. But
[40:29] (2429.36s)
if Honey were genuinely trying to avoid
[40:31] (2431.20s)
having kids on their platform, they
[40:33] (2433.04s)
probably shouldn't have sponsored one of
[40:34] (2434.80s)
the most popular influencers in the
[40:36] (2436.64s)
world whose key demographic of followers
[40:39] (2439.04s)
is, well, THIS.
[40:43] (2443.28s)
But Honey did sponsor Mr. Beast and
[40:45] (2445.52s)
those videos garnered a whopping 3
[40:48] (2448.32s)
billion views, making him Honey's number
[40:50] (2450.72s)
one sponsor. In fact, according to my
[40:52] (2452.80s)
data, Mr. Beast sponsorships represent
[40:55] (2455.04s)
over onethird of Honey's totaled sponsor
[40:57] (2457.28s)
views on YouTube. But it wasn't just Mr.
[41:00] (2460.08s)
Beast. Honey sponsored many channels
[41:02] (2462.08s)
whose content was, in my opinion,
[41:04] (2464.40s)
clearly targeted at miners. They
[41:06] (2466.40s)
sponsored Minecraft channels, Roblox
[41:08] (2468.16s)
channels, cartoon channels. I mean, just
[41:10] (2470.72s)
watch this video and tell me it's not
[41:13] (2473.12s)
targeted at children.
[41:22] (2482.24s)
>> This video is sponsored by Honey.
[41:24] (2484.00s)
>> If Honey wanted to target miners so
[41:25] (2485.68s)
badly, they should have just sponsored
[41:27] (2487.28s)
one. Oh, wait a minute. That's exactly
[41:29] (2489.60s)
what they did. This, ladies and
[41:31] (2491.28s)
gentlemen, is a Desiree Machado, who at
[41:33] (2493.68s)
the time was just 14 years old.
[41:36] (2496.48s)
>> Guys, I'm only 14. Hilariously, one of
[41:40] (2500.16s)
her videos that Honey sponsored is
[41:42] (2502.24s)
appropriately titled Back to School.
[41:45] (2505.20s)
Honey even ran her sponsored segment as
[41:47] (2507.92s)
a paid advert on YouTube. Now, speaking
[41:51] (2511.04s)
of paid adverts, let's quickly circle
[41:52] (2512.96s)
back to that Mr. Beast ad because of all
[41:55] (2515.68s)
the evidence, this has got to be the
[41:57] (2517.60s)
single worst example of Honey
[41:59] (2519.44s)
intentionally targeting miners. Let me
[42:01] (2521.60s)
play it again. I have a challenge for
[42:03] (2523.52s)
all of you. Go to every computer in your
[42:05] (2525.60s)
house, your mom's, your dad's, your
[42:07] (2527.04s)
sister, your brother's computer, and
[42:08] (2528.32s)
install Honey.
[42:09] (2529.60s)
>> They are literally encouraging kids to
[42:12] (2532.16s)
install a browser extension that tracks
[42:14] (2534.48s)
and collects data onto every computer in
[42:17] (2537.20s)
their household, including those
[42:18] (2538.88s)
belonging to their siblings. That is
[42:21] (2541.28s)
[ __ ] insane. And the emojis, I mean,
[42:24] (2544.48s)
come on. And look, this is not a good
[42:27] (2547.20s)
look for Mr. Beast either. Influencers
[42:29] (2549.52s)
have a duty of care to ensure that any
[42:31] (2551.60s)
advertising they engage in is compliant
[42:33] (2553.60s)
with relevant advertising laws,
[42:35] (2555.36s)
especially when your audience skews
[42:36] (2556.96s)
young. Of course, I did reach out to Mr.
[42:38] (2558.88s)
Beast and his team for comment, but
[42:40] (2560.56s)
never heard back from them. Now, in case
[42:42] (2562.56s)
you're still not convinced that this was
[42:44] (2564.16s)
intentional, because I don't know,
[42:46] (2566.00s)
perhaps Honey had a rogue marketing team
[42:47] (2567.84s)
that was poorly aligned with upper
[42:49] (2569.60s)
management. Well, here's a few words
[42:51] (2571.92s)
from Honey's former president, Joanne
[42:53] (2573.68s)
Bradford, talking about the benefits of
[42:55] (2575.92s)
working with Mr. Beast. We launched it
[42:58] (2578.48s)
with gamers playing YouTube on their
[43:01] (2581.52s)
desktop and a little gamer called Mr.
[43:04] (2584.80s)
Beast. So, we were his first advertiser
[43:07] (2587.68s)
and we did a deal with him. His
[43:10] (2590.08s)
collective ads, I think, have been seen.
[43:12] (2592.40s)
You know, I'm going to say billions,
[43:14] (2594.88s)
three plus close to four billion times.
[43:18] (2598.48s)
Every kid in America knows what honey
[43:21] (2601.20s)
is. Every kid in America was telling
[43:23] (2603.76s)
their moms and their dads they needed to
[43:26] (2606.32s)
download, you know, Honey in order to
[43:28] (2608.24s)
save money.
[43:30] (2610.00s)
>> They weren't just telling their moms and
[43:31] (2611.52s)
dads to install Honey, they were
[43:33] (2613.60s)
installing it themselves. My 5-year-old
[43:36] (2616.08s)
brother installed Honey on his computer.
[43:38] (2618.48s)
Wow, kids start saving money earlier and
[43:40] (2620.88s)
earlier these days. And here's Honey's
[43:43] (2623.36s)
co-founder, Ryan Hudson, discussing how
[43:45] (2625.44s)
Honey leveraged the credibility that
[43:47] (2627.20s)
influencers have with their audience.
[43:49] (2629.36s)
Um, we learned how to work with
[43:51] (2631.52s)
influencers in particular on YouTube to
[43:54] (2634.88s)
lean into the credibility that they have
[43:56] (2636.64s)
with their audiences, the ability to
[43:58] (2638.24s)
speak the language of their audience.
[44:00] (2640.32s)
And it's been incredible for us. And so
[44:02] (2642.72s)
we've done if you if you watch YouTube
[44:05] (2645.84s)
or you have kids that watch YouTube, um,
[44:08] (2648.48s)
you they have seen they have seen if you
[44:11] (2651.84s)
watch Mr. Beast, um,
[44:13] (2653.52s)
>> it's a lot of honey.
[44:14] (2654.40s)
>> A lot of honey.
[44:16] (2656.16s)
>> This was no accident. Honey's leadership
[44:18] (2658.72s)
knew what they were doing and they knew
[44:20] (2660.56s)
it worked well. That Mr. Beast ad alone
[44:23] (2663.04s)
has 118 million paid ad views, the
[44:26] (2666.08s)
highest of any Honey ad on YouTube,
[44:28] (2668.24s)
which strongly suggests it was also
[44:30] (2670.16s)
their best performer. Honey likely
[44:31] (2671.92s)
invested millions into this ad alone.
[44:34] (2674.72s)
Honestly, it's genuinely shocking
[44:36] (2676.88s)
they've gotten away with it for so long.
[44:40] (2680.16s)
Now, given everything we've uncovered so
[44:42] (2682.56s)
far, you might be wondering how on earth
[44:44] (2684.72s)
there could be more to this
[44:45] (2685.76s)
investigation. Well, while Honey was the
[44:48] (2688.32s)
first extension of its kind,
[44:50] (2690.00s)
unfortunately, they certainly weren't
[44:51] (2691.92s)
the last.
[44:57] (2697.20s)
There are now hundreds of coupon and
[44:59] (2699.60s)
cashback extensions. And while I haven't
[45:02] (2702.00s)
investigated all of them to the same
[45:03] (2703.84s)
extent that I have with Honey, many of
[45:06] (2706.00s)
them engage in similar behavior,
[45:08] (2708.32s)
especially when it comes to poaching
[45:10] (2710.24s)
affiliate commissions. And that's why
[45:11] (2711.92s)
there are now multiple class action
[45:13] (2713.68s)
lawsuits targeting these other
[45:15] (2715.52s)
companies. Influencers haven't just been
[45:17] (2717.36s)
losing money to one sleazy salesman at
[45:19] (2719.44s)
checkout. They've been losing money to
[45:21] (2721.44s)
an army of them. In my opinion, these
[45:23] (2723.68s)
extensions are nothing but leeches. They
[45:26] (2726.16s)
are parasites in the world of
[45:28] (2728.16s)
e-commerce. In fact, I've encountered so
[45:30] (2730.80s)
much shady behavior with these
[45:32] (2732.16s)
extensions that I felt it was necessary
[45:34] (2734.24s)
to create my own browser extension to
[45:36] (2736.32s)
alert me anytime an affiliate cookie is
[45:38] (2738.56s)
loaded onto my browser. It's called
[45:40] (2740.48s)
Cookie Guard, and it's especially
[45:42] (2742.64s)
helpful because some of these extensions
[45:44] (2744.80s)
will load their affiliate link in ways
[45:46] (2746.80s)
that are completely invisible to the
[45:48] (2748.88s)
user. For example, if you click to copy
[45:51] (2751.12s)
a coupon on Honey, it looks like a
[45:52] (2752.80s)
harmless feature because you don't see
[45:54] (2754.72s)
that sneaky tab opening in the corner.
[45:56] (2756.80s)
But with Cookie Guard installed, you can
[45:58] (2758.64s)
see that behind the scenes, they are
[46:00] (2760.32s)
quietly stuffing their cookie through
[46:02] (2762.00s)
what's known as a hidden iframe. You can
[46:04] (2764.80s)
Google it. Even worse, I found that the
[46:07] (2767.04s)
extension Karma Now would stuff their
[46:09] (2769.12s)
cookie the moment you landed on a
[46:11] (2771.12s)
checkout page, completely automated. No
[46:14] (2774.00s)
interaction with their extension
[46:15] (2775.60s)
required. They also had a system that
[46:17] (2777.76s)
could detect when you used a
[46:19] (2779.44s)
competitor's extension like Honey, and
[46:21] (2781.44s)
they would wait for that extension's
[46:22] (2782.88s)
affiliate link to load, then immediately
[46:25] (2785.12s)
override it with their own. Again, no
[46:27] (2787.84s)
interaction with their app required.
[46:30] (2790.48s)
It's insane. And this sort of malicious
[46:32] (2792.96s)
behavior is not exclusive to browser
[46:35] (2795.12s)
extensions. Even Microsoft couldn't
[46:37] (2797.12s)
resist the temptation of easy affiliate
[46:39] (2799.36s)
money and decided to offer coupons
[46:41] (2801.44s)
directly from the Edge browser. Watch
[46:43] (2803.68s)
what happens when I visit my NordVPN
[46:45] (2805.84s)
affiliate link. Once I've clicked on one
[46:47] (2807.36s)
of their coupons, Microsoft immediately
[46:49] (2809.84s)
replaced my affiliate cookie with their
[46:51] (2811.60s)
own, poaching the sale. And when I tried
[46:54] (2814.00s)
to reload my affiliate link to reinstate
[46:56] (2816.24s)
my cookie, it comes back momentarily,
[46:58] (2818.56s)
but Microsoft quickly and automatically
[47:01] (2821.28s)
replaces it again with their own. I
[47:03] (2823.92s)
mean, damn, Microsoft, people hate your
[47:06] (2826.32s)
browser enough as it is. And how about
[47:08] (2828.88s)
Opera's browser? They recently stopped
[47:10] (2830.80s)
this, but for some time if you simply
[47:12] (2832.88s)
entered the URL of a store that they
[47:14] (2834.80s)
were partnered with and clicked their
[47:16] (2836.40s)
autocomplete suggestion, Opera
[47:18] (2838.56s)
considered that fair game to inject
[47:20] (2840.56s)
their affiliate link and claim the sale.
[47:23] (2843.28s)
Affiliate marketing has truly become a
[47:26] (2846.32s)
war of the cookies. I could go on and on
[47:29] (2849.12s)
with these examples. Now, this all begs
[47:32] (2852.40s)
the question, how the [ __ ] is all of
[47:34] (2854.96s)
this being allowed? Are there no rules
[47:37] (2857.20s)
in place to prevent this behavior? Well,
[47:40] (2860.16s)
as it turns out, yes, there are in fact
[47:42] (2862.72s)
plenty of rules in place that should
[47:44] (2864.48s)
have prevented much of what I have
[47:46] (2866.32s)
uncovered in this investigation. So, why
[47:50] (2870.24s)
hasn't it? Well, that's where greed and
[47:53] (2873.12s)
incentives comes into the equation. You
[47:56] (2876.08s)
see, so far we have only discussed two
[47:58] (2878.48s)
key players in affiliate marketing,
[48:00] (2880.48s)
merchants and affiliates. Merchants want
[48:02] (2882.64s)
to sell their products and services and
[48:04] (2884.56s)
affiliates want to promote those
[48:06] (2886.08s)
products and services for a commission.
[48:08] (2888.40s)
Pretty straightforward. But there's
[48:10] (2890.24s)
another crucial player that we haven't
[48:12] (2892.24s)
yet discussed and that is the affiliate
[48:15] (2895.12s)
networks. Affiliate networks are like
[48:17] (2897.36s)
dating apps for the industry. They allow
[48:19] (2899.36s)
merchants and affiliates to easily
[48:21] (2901.36s)
discover one another, connect and work
[48:23] (2903.92s)
together. These networks also take care
[48:26] (2906.16s)
of all the technical stuff like the
[48:28] (2908.16s)
affiliate tracking links, reporting and
[48:30] (2910.40s)
commission payments. But most
[48:31] (2911.84s)
importantly, these networks also set the
[48:34] (2914.24s)
rules. They are the gatekeepers. Some of
[48:36] (2916.96s)
the largest networks in this industry
[48:38] (2918.56s)
are Awin, Commission Junction, Impact,
[48:41] (2921.20s)
and Recruitin Advertising. Now, not only
[48:44] (2924.16s)
do these networks have their own strict
[48:45] (2925.92s)
policies, at least on paper. But on top
[48:48] (2928.40s)
of that, over a decade ago, most of
[48:50] (2930.56s)
these networks collectively agreed to
[48:52] (2932.48s)
enforce two codes of conduct. one for
[48:55] (2935.36s)
downloadable softwares like extensions
[48:57] (2937.52s)
and the other for all affiliates that
[48:59] (2939.52s)
promote coupon codes. And the rules set
[49:01] (2941.92s)
out in these codes prohibit many of the
[49:04] (2944.40s)
behaviors uncovered in this video. So
[49:07] (2947.20s)
why aren't these networks enforcing
[49:09] (2949.36s)
their own bloody rules? Well, for every
[49:12] (2952.24s)
sale made in affiliate marketing, these
[49:14] (2954.24s)
networks also get a slice of the pie.
[49:16] (2956.80s)
They generally charge between 20 and 30%
[49:19] (2959.76s)
of each and every commission paid. There
[49:21] (2961.92s)
are some variations, of course. Not all
[49:23] (2963.68s)
networks have the same rates and pricing
[49:25] (2965.44s)
structures, but the core principle is
[49:27] (2967.76s)
almost always the same. The more money
[49:29] (2969.76s)
that flows through these extensions, the
[49:31] (2971.76s)
more the networks get paid. So, when
[49:34] (2974.16s)
these networks enforce rules resulting
[49:36] (2976.16s)
in a loss of income for the extensions,
[49:38] (2978.56s)
they too take a hit. Enforcing their own
[49:40] (2980.88s)
rules basically means shooting
[49:42] (2982.96s)
themselves in the foot. That kind of
[49:44] (2984.56s)
incentive structure doesn't just fail to
[49:46] (2986.72s)
prevent fraud, it practically invites
[49:49] (2989.36s)
it. Part three is going to be huge. And
[49:52] (2992.16s)
that was supposed to be the next video
[49:54] (2994.24s)
in the series. But that's not the video
[49:56] (2996.88s)
you're going to see next. In the next
[49:58] (2998.88s)
video, I'm going to be exposing a secret
[50:01] (3001.28s)
system buried deep within Honey's code
[50:03] (3003.60s)
that was never meant to see the light of
[50:05] (3005.84s)
day. The system was engineered
[50:07] (3007.76s)
specifically to bypass one of the most
[50:10] (3010.40s)
important rules in affiliate marketing.
[50:12] (3012.64s)
A rule intended to protect influencers
[50:14] (3014.96s)
from the very commission theft
[50:16] (3016.96s)
demonstrated in my last video. And that
[50:19] (3019.28s)
rule is known in the industry as
[50:21] (3021.68s)
standown. Standown essentially means
[50:23] (3023.92s)
that extensions must detect if a user
[50:26] (3026.48s)
has already clicked on someone else's
[50:28] (3028.40s)
affiliate link. And if they have, the
[50:30] (3030.56s)
extension must deactivate itself and not
[50:33] (3033.20s)
interfere. Yet during my testing, Honey
[50:36] (3036.00s)
virtually never stood down. Suddenly,
[50:38] (3038.56s)
however, Honey began complying much more
[50:41] (3041.12s)
frequently. And this change in
[50:42] (3042.96s)
compliance
[50:44] (3044.80s)
didn't smell right to me. So, I started
[50:47] (3047.92s)
digging and that's one of the reasons
[50:50] (3050.24s)
why this video was delayed for so long.
[50:53] (3053.12s)
Because, as it turns out, Honey wasn't
[50:55] (3055.36s)
just ignoring standown rules. The system
[50:57] (3057.52s)
they created was designed explicitly to
[51:00] (3060.32s)
bypass it while hiding that behavior
[51:02] (3062.72s)
from compliance testers. And I have all
[51:05] (3065.04s)
the evidence to prove it. In fact, what
[51:07] (3067.52s)
I found was so shocking that I consulted
[51:09] (3069.68s)
with a respected security researcher who
[51:11] (3071.60s)
was able to independently verify and
[51:13] (3073.92s)
validate my discovery. The behavior I'm
[51:16] (3076.08s)
describing, in my opinion, likely
[51:18] (3078.40s)
constitutes criminal behavior. And in a
[51:20] (3080.96s)
few days, everyone will learn about the
[51:22] (3082.96s)
greatest heist in affiliate history.
[51:27] (3087.44s)
They're attempting to stand down as
[51:30] (3090.72s)
little as possible while avoiding
[51:32] (3092.88s)
getting caught. Uh those objectives are
[51:34] (3094.96s)
intention, of course. The more you don't
[51:36] (3096.72s)
stand down, the more you're likely to
[51:38] (3098.48s)
get caught for not standing down. So,
[51:40] (3100.32s)
they're trying to figure out in what
[51:41] (3101.68s)
circumstances can they avoid standing
[51:43] (3103.68s)
down. uh and not face a material risk of
[51:46] (3106.24s)
being caught.