51. Behind the Likes: What Small Business Owners Must Know About Social Media Fraud

In this episode of UnHacked, hosts Justin Shelley and Mario Zake are joined by Tim O'Hearn, author and former software engineer, to explore the darker side of social media engagement strategies. Tim shares insights from his personal journey of creating social media bots and discusses the implications of fake activity on platforms like Instagram. The conversation delves into the potential harm of artificial social proof, highlighting the need for small business owners to be discerning about social media metrics. Tim also discusses his book, Framed: A Villain's Perspective on Social Media, and the broader impact of deceptive online practices. Listen to uncover the hidden threats and learn how to navigate the complexities of social media with your business's integrity intact.

Buy Tim's book here: https://amzn.to/3XoxUC1

Episode Transcript

Justin (00:15):
Welcome everybody to episode 51 of Unhacked. Guys,
like I say every week, Unhacked,bit of a misnomer. Here's the
truth. Ninety seven percent ofbreaches can be prevented by
doing the basic cybersecuritymeasures. We talk about that
week after week.
We give you that formula. Butwhat we cannot do, in fact, is
unhack you once you getbreached. So today, we've got a

(00:39):
special guest who he he's gonnago we're gonna dig into a little
bit of the social media side ofthings. What's going on in the
past? What's going on now?
And what as business owners wereally need to know and
understand about social media?Real quick, I'm gonna introduce
myself and then Mario, and thenwe're gonna bring on our guest.
I am Justin Shelley, CEO ofPhoenix IT Advisors. And I help

(01:00):
you make money with technology,and then I help you protect it
from the Russian hackers, thegovernment auditors and
assessors and penaltycollectors, And finally the
attorneys who are going to comehoover up anything that might be
left if you were to getbreached. So that's what I
specialize in.
Work in the great state of Texasand Nevada and Utah. I've spread

(01:21):
all over the place these days.Today, as always, I'm here with
my good friend, Mario. Mario,tell us a little bit about
yourself, who you are, what youdo, and who you do it for.

Mario (01:30):
Mario Zake, CEO of Mastech IT. We're located in New
Jersey. We're at the side ofManhattan. I've been in business
for twenty one years now. Wework with, small to medium sized
businesses, helping them withall the technology needs,
keeping them protected not onlyfrom the Russian hackers, but
from the people in China and allthose other countries that's

(01:54):
Justin doesn't wanna work with.
I don't

Justin (01:57):
like the Russians.

Mario (01:58):
And and like I said before, we specialize in key in
helping business owners sleepbetter at night.

Justin (02:05):
It sounds good, Mario. Always a pleasure to have you
here. Thanks for being here weekafter week.

Mario (02:10):
My pleasure.

Justin (02:11):
And now we are gonna introduce Tim. Tim, say hi.
Thank you for being here today.

Tim (02:18):
Justin, thanks for having me. My name is Tim O'Hearn. I'm
an author who wrote Framed, avillain's perspective on social
media. There is

Justin (02:25):
This book right here, is that the one? That is it. This
is a big book, Tim, and I amgonna read it like I promised
you before. I will confess Ihave not read the entire book as
of this recording. So let metell the audience a little bit
about you.
And then I want you to fill anygaps that I leave. Here's what I
understand, Tim. You are asoftware engineer. I I've got a

(02:48):
pause there. What are yourfavorite languages?
And I mean, with AI, is thisgoing away? Two questions. Go.
What do you got?

Tim (02:57):
Primarily, I code in Python. I've done Python. Think
from, like, yeah, C plus plusJava as well. So the, question
of is it going away? I would sayprobably not.
A lot of what I've done has beenhighly domain specific. So many
companies that I've worked atwouldn't really entrust, AI with
some of the things I've done andthe precision, required is not

(03:19):
really, not really feasiblecurrently. So I think Python's
still a pretty safe bet, but Idon't know if I would give that
same advice to someone who'sjust going into college today.

Justin (03:28):
I mean, that's a great great point. What would you
suggest if somebody was going tocollege right now? I'm gonna
come back to your intro. Don'tworry. But I'm I'm just getting
ready to launch my collegecareer, and I wanna be a
software developer.
What are you gonna tell me?

Tim (03:43):
I would go all in on math and hard sciences. Really?
Basic, like, yeah, I would goall in on math and some
statistics depending on whattype of college you're in and if
you can get classes through the,like, statistics taught by the
math professors, not by thebusiness professors, and then
also a much deeper understandingof things like physics.

Mario (04:06):
Little fun fact for you. When I started college, I was an
information science major, and Idid c plus plus. And, you know,
I was kind of frustrated withthe, you know you know, trying
to, you know, make sure I didn'tmiss, like, any commas or
exclamations in in the codingand stuff like that. And then

(04:28):
for me, it was calculus. Icouldn't get past calc one.
I took that shit three times.And I'm like, forget this. I
ended up because of calculus, Iended up changing majors, and
I'm like, I don't wanna haveanything ever to do with
programming or calc. So Iswitched and never looked back.

Justin (04:48):
I mean, I don't wanna brag Mario, but I got the way
all the way to calc three beforeI just, like, one day had this
existential crisis in the middleof class. I'm like, I have no
goddamn idea what they'retalking about anymore. Like
forget about solving theproblem. I don't even understand
what we're talking about. Andand that was when I I just said,
screw this.
I'm gonna be Bill Gates. I'mgonna drop out of college and
I'm gonna start my own business.This is not advice to you

(05:13):
youngsters, but I like Tim'sadvice much better. Okay, so
back to introductions. Tim, Iunderstand you have created at
your own admission some of themost pesky and effective bots to
ever be unleashed on socialmedia.
I want a little bit moreinformation there. Tell me about

(05:34):
these bots you made and why arethey pesky?

Tim (05:38):
Later in the twenty tens on sites like Instagram, it was
pretty common to make a new postwith a given set of hashtags and
start attracting some commentsand some likes from accounts you
had never seen before. Thisappeals to us greatly as humans
because we enjoy the interactionand we enjoy it when people are
patting us on the back, so tospeak. But really one of the

(06:00):
things that happened during thattime was people like me created
bots that found these newpictures with these new hashtags
and interacted with them. Andthe point was that you could,
algorithmically find these postsand then interact with them in
order to get something back. Somaybe if I follow you, there's a

(06:21):
10% chance that you'll follow meback.
If I do that a couple times perday, who cares? But if I'm able
to do that, you know,industrially with thousands of
actions per day, it can have ameaningful result. Right? The
reciprocation is meaningful, andthat was my entire business. So
the system that I built, at acompany which I code named Shark
Social was entirely based onthis model called follow

(06:45):
unfollow, which attracted,reciprocal actions specifically
on Instagram.

Justin (06:50):
Interesting. Okay. And you did this relatively
recently. Right?

Tim (06:57):
Things began in 2017, and the final iteration of the
business was in 2022. So it wasrecent in that I began writing
about it while it was stillgoing on. But the book, of
course, was not published untila few years later.

Justin (07:13):
Yeah. Got it. All right. So let's talk a little bit about
the book. This came out inFebruary, right?
Very hot off the presses. Whatit you know, and like I said, I
have not read the entire 2,516pages yet. Actually, it's what
is it? 400? Listen, I have a400, so if it's more than about

(07:34):
15 pages, I kinda cry.
So give me if I were to feedthis into chat GPT and say
summarize this for me, what whatwould it spit back out?

Tim (07:47):
Succinctly, it's the complications of the Internet
age.

Justin (07:51):
Okay.

Tim (07:52):
And beyond that, if we're thinking about the extra
paragraphs and all the em dashesthat chat GPT might introduce
there, There's a lot in that.It's not necessarily a
narrative, a chronicle of mylife. More so, it's a chronicle
of the Internet, meaning fromearly web two point o, so
MySpace times, all the waythrough TikTok. What was the

(08:15):
same? What was different?
Not just from the technologystandpoint, but also from the
perspective of consumerpsychology. So why were people
doing these weird things thatthey were doing on MySpace in
02/2008? And by 2018, why werethese things not considered
weird at all? How did, you know,life evolve around the Internet?

(08:36):
So What were these weird things?
So the main practice, which Iaddress in one of the opening
chapters, it was called Myspacewhoring. So if you were on
Myspace, and back then we had alot of new terms that were
emerging from, like, you know,emo music and scene like,
different subcultures, I wouldsay, in America, there was this

(08:56):
concept of the attention whore,which we saw in, like, several
movies at that time, but itreally was someone who was
obsessed with getting moreattention. And once the Internet
became more accessible,especially people got broadband
Internet in 02/2006, '2 thousand'7, and so on, we were able to
see these people who desiredthat validation come to exist in

(09:18):
the digital space. And MySpacewas really one of the first
places where that existed. Andit became a practice known as
MySpace whoring as a shorthandfor attention whore.
But as time went on, I thinkkids like me, if you were like
13 years old on Myspace, youdidn't really think of it as an
attention whore. You were justvery much interested in that it

(09:40):
was like a salacious term, like,oh, it's a whore. So like, what
is that? What does that mean?And so it attracted a lot of
people to it.
But what it means for theInternet today is that this was
the first place where we'vepositively identified people
willing to negatively affecttheir user experience and
disrupt the experience of thosearound them just to increase

(10:02):
those digital metrics, meaningfriend count on Myspace. Whereas
today, we don't think twiceabout it. On Myspace, it was
totally new. And I'll tell youthis, if you had 2,000 friends
on Myspace in 02/2008, your newsfeed or whatever semblance of a
news feed, it was basicallyunusable.

Justin (10:21):
Yeah. Yeah. I imagine. Because now the algorithm sort
filter and well, I mean,honestly, they they do what they
they push whatever to the topthey want us to believe. That
might

Mario (10:31):
be the the most.

Justin (10:32):
Yeah. That might be a subject for a different
different episode. I don't know.So okay, Tim. You you did this
up until '22.
You've been writing a book.What's your day job now?

Tim (10:46):
Currently, I'm all in on writing. I spent most of my
twenties. My primary employmentwas in quantitative trading. So
I worked at different,proprietary trading firms,
meaning we didn't have we'relike hedge funds without
customers, so to speak. And it'sa very unique place in that the
innovation is very rapid.
The pace of change is veryrapid. The problems are really

(11:09):
exciting. And frankly, thecompensation is great too. For
me, it's allowed me a lot offreedom, both to take long
periods of time off to pursueother things and to just kind of
consider what I want to do withmy skill set. So going into this
year and really even late lastyear, I started to realize that
the book was taking so long thatI needed to finally split up

(11:31):
these different aspects of mylife.
And the way that it went there,I think everything worked out
well and that I decided to takea break. And that break allowed
me to not only publish Framed,but also work with an editor and
a promotional team who helped meput out the best product despite
the fact that I was anindependent author.

Mario (11:50):
Okay. And so you said you right now, you're you're doing
on a daily basis, you're justwriting. Are you talking about
writing code or writing morebooks? Or

Tim (12:00):
Most of my activities so far this year has been directly
related to Framed itself. Sothat's both, like, writing my
newsletter, writing my blog,doing promotional work for it. I
also do sports journalism aswell. So there's been quite a
few articles there. Like justthis week, I interviewed Allison
Felix, is like the mostdecorated, track and field

(12:22):
athlete in history.
So I've been doing both acombination of like writing
there, supporting my book, andthen of course trying other
efforts. But if there's onething that the journey taught
me, it's that you could spendyour entire life writing the
next one or like writing thefirst So I'm hesitant to say,
oh, there's a new book coming orthere's a new something coming,

(12:44):
just because I can appreciatenow just how insane of an effort
it is and knowing that that youprobably shouldn't tell people
until you're, like, 90% of theway there.

Justin (12:54):
Do you do you have a guess on how many hours you
spent writing this book?

Tim (13:00):
I could say this. As far as content, going into
Thanksgiving, I was well over ahundred thousand words. Well,
well over probably about a15,000. The manuscript that I
gave to my editor in lateJanuary was about a 50,000,

(13:21):
which we then got down to a36,000. I can say that in terms
of hours just betweenThanksgiving and getting the
book out, you know, let's say,into mid March, it was roughly
just to put it simply, I wasputting in forty hours per
weekend for most of that periodwhile working a normal job and

(13:44):
while also training at a prettyhigh level as an athlete.
So to extrapolate that out, wewere definitely talking well
over a one thousand hourproject.

Justin (13:53):
Yeah. That's a lot. My hats off to you. Mario, you got
any questions for Tim before wekind of jump into the meat of
this thing?

Mario (14:01):
No, no, let's do it. Let's do it.

Justin (14:03):
Alright. So Tim, this is a cybersecurity podcast. And
again, our goal is to protectthe business owners, Because
here's the bottom line. We'rebusy, right? I mean, you're
talking about all these hoursyou spent writing a book.
No matter what we're doing, itbecomes almost obsessive, right?

(14:24):
It's just like day and night, Ieat, breathe, sleep this stuff.
I'm up at 06:00 working,Sometimes I'm up at 04:00
working. Hell, sometimes I'm upat two in the morning because I
can't sleep working and I workfrom home so I can do that.
Right?
There's no shutoff. I worked allweekend long. I took a break to
go out in the yard for a fewminutes to to just get some
exercise and some sun and dosome yard work. But otherwise, I

(14:44):
am hyper focused on growing mybusiness and delivering a
product. I live in the world ofcybersecurity and if I'm being
completely honest, completelytransparent, I don't always
think about everything that Ishould and as far as protecting
my business from these threats.

(15:04):
Being again honest, the reason Ido this podcast is to combat
that. My natural instinct is tofocus on one thing, I get
distracted too easy. So this isthe problem that I believe most
business owners face. We don'tknow what's going on outside of
the world where the empire we'retrying to build. Now we're

(15:26):
bringing in something that Idon't know that's ever really
crossed my mind as I'm makingall these confessions is the
social media.
I mean, does cross my mind. Ithink about it a lot actually,
but I don't really dive into it.So let's dive into how is you
know, the the stuff that you'vedone and the stuff that you see
going on out there with socialmedia, how could that impact a

(15:46):
business owner and what do weneed to be aware of? If we just
start at the very high level andthen we'll kind of break this
down into details.

Tim (15:52):
Sure. We can get the most technical aspect out of the way
first and decide how deep wewant to go. But I think what's
most relevant to listeners of acybersecurity podcast is that
there are many, vectors ofattack, in the cybersecurity
space. And one that was actuallycentral to building my business
and improving my product was APIhacking. So you've probably seen

(16:17):
Corey Ball's book called HackingAPIs.
And it's not hacking in the samebreaking and entering, like
Trojan horse virus type space.What it means is you're
accessing an API, so anapplication pro programming
interface that might be exposedinternally. So this is not
necessarily a small businessowner problem, but anyone who is

(16:42):
in the software space couldleave something exposed that
then leaves them vulnerable toan attack, to a hacker, so to
speak. And this is relevant towhat I did when I was at Shark
Social in that the entirebusiness was based around
reverse engineering codefunctionality that was provided
by Instagram and then buildingthe business around that. So

(17:05):
that's one thing where, yeah, itwas hacking.
We weren't, necessarily, like,harming Instagram directly in
the same way that a hacker mighthurt your clients, but it's
still a relevant practice.

Mario (17:19):
Now in this to kind of put things together for some of
our our audience that may notknow what an API is, like, it
it's it's just Tim, correct meif I'm wrong. It give software
gives the ability forprogrammers to use this code to
integrate in, you know, incomingand outgoing from their program.

(17:39):
Now to relate it to a previouspodcast that we had, you know,
Justin, like, with with yourcustomer that wanted to have
this third party companyintegrated into their medical
software, if if that medicalsoftware had an API that was
vulnerable for somebody to hackinto it, they could essentially

(18:00):
hack into a third party company,which could, you know,
essentially then, you know,affect the medical software and
the medical practice and theirdata. So it could be, you know,
a big chain of attacks thatcould end up leading to
something similar. Correct?

Tim (18:17):
Yeah. I would say, generally speaking, the hack is
not necessarily as destructiveor as, pen deeply penetrating as
we would think a typicalcybercrime attack. But it could
be something as simple as, okay.Well, the API isn't protected
against a denial of serviceattack. So maybe you have

(18:37):
Cloudflare set up in front ofyour your website or whatever,
but you forgot about Cloudflareor some load balancer for this
route you left open.
So you're still vulnerable. Eventhough the loss at the end of
the day might just be AWScharges, it's still disruptive.
And the way I think a listenercan understand this is that it's
not this type of, like, blackgloved superhuman, like, hacker

(19:03):
who knows all the special codes.It can be thought of as you take
your phone out, you're onInstagram. You tap a button,
which we know is synonymous withlike.
So you like a picture. So itmight say, Tim likes Justin's
picture. The translation layerthere, it doesn't translate it
to French. It's actually stillin basically plain English, just

(19:25):
with a little bit moresyntactical structure there. So
when that makes it toInstagram's private API, it is
as simple as API slash v twoslash Tim slash relationship
slash Justin picture one slashlike, more more or less.
So essentially, it's juststructured in a way that looks

(19:48):
totally similar to a directoryto a directory structure on your
own computer or in the structureof a URL. And that's what these
web APIs are. They're just URLs.

Justin (19:59):
So if you know the if you know what what that URL is
looking for, first of all,you've gotta have the actual
URL. But after the URL is wherethey put all the information.
Right? And if you know whatthey're looking for, that's how
you can get in there and andmanipulate things.

Tim (20:14):
Yeah. That's true. So it could be a query parameter or it
could be built into the routeitself, and that kind of gives
way to the practice of sniffing,which is probably outside the
scope of what we wanna talkabout here. But Yeah.
Right. It's it's kind of up tothe the attacker to decide how
they're going to uncover it,they're gonna build upon code
that's already been released, orif they're gonna take their best
guess. Like, for example, Iwould guess if there's something

(20:37):
called like, there's probablyalso something called comment
because those are the featuresthat I'm used to having on
Instagram. So sometimesintuition guides us in ways
where typical sniffingtechniques, maybe don't work the
same way.

Justin (20:53):
So, you know, when when we talked before, you and I
chatted for a few minutes beforethe recording, you talked about
intentions, you talked about howwe can identify fake activity,
and and I wanna get into that,but at first I wanna ask, why
does it matter? Why why do weneed to worry about fake
activity? And again, in in thebusiness world, is there damage?

(21:14):
Because what you did before isyou helped businesses gain
followers and I'm putting thatin quotes, likes, right?
Whatever it's social currency,It's trust.
It's credibility. Right? That'sthat's what they're buying from
you. Is that is that fair?

Tim (21:30):
That is fair.

Justin (21:31):
Okay. So what's the danger in that?

Tim (21:35):
If you're able to buy this social currency, it's akin to
entering a cheat code. So whereeverybody shows up, at the start
of the day and they think, okay.Here's what I expect from
someone with 10,000 followers.Here's what I expect to with
someone from a hundred thousandfollowers. Regardless of how
long you've been on the platformfor, you likely have some model

(21:57):
for understanding these, youknow, certain levels of of of
importance.
The problem is that if itbecomes so easy to buy and to
fake, then also, anyauthenticity, including of real
accounts, suddenly loses itsmeaning as well. So regardless
of the fact that to get there,possibly extremely disruptive

(22:18):
things have to happen to getthese accounts to like one
account or to follow oneaccount. But also that, at the
end of the day, if it goesunchecked, then users en masse
will lose confidence in themetrics to begin with.

Justin (22:35):
Okay. So first, you're gonna trust things that you
shouldn't be trusting, and thenwe're just gonna say this whole
system's screwed. Right? Kind ofin a nutshell.

Tim (22:43):
It it quickly gets to this very, very cynical view of
what's going on, what is theInternet actually. I would give
the I would make the point thatthe business I ran was not
purely fake. It was actuallylogging into customer accounts.
So although I controlled fakeaccounts to do certain things

(23:03):
mainly along the lines oftesting, I never had access to a
massive botnet of 10 to ahundred thousand accounts. That
was kind of in the previous era.
But the danger in somebody beingable to control those just to
give you a harmless like is thatthey could also participate in a
massive propaganda campaign. Andthat's what we started to see

(23:27):
when people started talkingabout it in you know,
interference in elections whereyou're like, yeah. If you have
even a thousand accounts and nowyou're using AI to write
comments and to astroturfdiscussions, that's really all
you need to sway people in oneway or another. Forget a hundred
thousand. But the biggestplayers, right in that spot in

(23:47):
2017 when I started gettinginvolved, The biggest players
had hundreds of thousands ofaccounts, and they could just
create them no problem.
For us, it was a little bitdifferent, and we never got into
that the darkest the darkestpart there. Gotcha.

Justin (24:02):
Mario Puckus, how many social media likes do you have?

Mario (24:06):
Likes? I

Justin (24:07):
don't know. I

Mario (24:08):
think I could count them on my hand. One hand. Yeah. I
mean, honestly, there's timeswhere I have to go on to, like,
my personal account to go, like,my my business account one, and
then sometimes we'll postsomething and get, like, a a
hundred likes. You know, itreally depends on on on what it

(24:29):
is and, you know, the topic.
Some things, you know, do betterthan others. You know, anybody
in the social media world willwill Yeah.

Justin (24:41):
I think I cut you off. You're gonna ask a question.
Let's go back to that.

Mario (24:44):
Yeah. I I forgot it. No. So my my question was, and if I
could remember real quick, isthis so this this happens on a
daily basis, this, you know,just mass liking and and stuff
like that? And is it on everyplatform?
Is it on Instagram, Facebook,LinkedIn?

Tim (25:09):
On some platforms, I would like to think that it's been
cleaned up. And my point wouldbe, if it hadn't been cleaned up
on Instagram, there's a shotthat I would still be doing it.
You know? So, like, onInstagram, it's not necessarily
something where we've seen, itproliferate. It really stopped
around 2022, which is around thesame time that I stepped away

(25:30):
from the business.
But on you know, the thing isabout the other platforms is
that on Reddit now, we see a lotof this different type of
astroturfing and a lot ofdifferent type of interference
that is directly probablydrawing to this type of, you
know, bot bot behavior. So it'sbecoming more profitable, but
it's also becoming much, muchmore, restricted, by the

(25:53):
platforms themselves because theplatforms actually gain very
little, from people abusing theplatforms that way.

Justin (26:01):
Well, you're talking about manipulating people
through this I mean, basicallywhat what you did. I've got a
real world example of that. AndMario, you you introduced it,
the the third party integrationthat we've talked about a couple
times now. In in trying to vetthis company, I'll try to keep
the details out of it. I did goto their social media and I saw

(26:25):
a lot of social currencyattached to their social media
account.
And and for a second, I almostlet my guard down. I'm like, oh,
everybody trust these guys. It'sprobably fine. And and I I'm
like I said, I I really had tocatch myself because that
doesn't mean anything. AndRight.
You know, we're we're Tim, we'retalking about how do we what

(26:47):
what does this mean for abusiness owner? This is
something that I think we allneed to really watch for. We put
a lot of confidence. Well, I'llspeak for myself. I put a lot of
confidence in how many followersan account has or how many
comments they get when theypublish something.
I spend a fair amount of time onLinkedIn and I'm stupid kinds of
envious of people who can throwout shit content. I'm not even

(27:08):
gonna lie, it's it's shit. Andthen everybody's liking it and
glazing them as the kids thesedays would say, and throwing
stupid comments at it. Andthere's that envy, there's also,
there's just a built in trustlike, well, they must be smart,
they must be connected, theymust be good people, they must

(27:28):
be a good company or whatever.So what what can we do to combat
this?

Tim (27:36):
We have to change everything that we've taught
ourselves about social media andabout these metrics.
Unfortunately, in the earlydays, these were very, very
useful metrics and they werehard to fake. On YouTube, you
know, on YouTube, it's a littlebit easier. The subscribers are
a little bit harder to fake.Video views are a little bit

(27:56):
harder to fake.
And I find that for any YouTubevideo, you know, I've evaluated
tons of accounts as I've madepodcast appearances as well. I
can pretty easily get an ideafor, like, what people actually
think of the podcast and whetherthe podcast is actually what
they say it is. On other sites,it's becoming increasingly

(28:17):
difficult because now there'sSEO, right? We have this thing
where we've had, for decades,people have been selling us
services of how to look moreimportant without necessarily
changing your content. So it'sjust this big conundrum of like,
you know, faking it and thenmaking it and ultimately just
like fooling the algorithms atthe end of the day.

(28:38):
I think the point you bring upor the question you ask is one
that millions of people areasking them, especially people
who are trying to make moremoney or generate leads, from
social media, and that it's justsimply very, very hard now, to
differentiate, like, versus,like, kind of real, but boosted.

Justin (29:00):
Yeah.

Mario (29:00):
Can I just go on a little slight rant for a second? I used
to love when YouTube not onlyhad the like, but it also had
the dislike button. And you cansee how many people disliked
that that video. Because if I'mlooking for something that I'm,
like, you know, googling and Iwanna watch a video of it, and

(29:21):
then I would go somewhere andhad, like, three likes, but 637
dislikes. I pretty I pretty muchknew that don't watch this
video.
It's a waste of your time. Andthey did away with it. They they
still can give you the abilityto dislike a video, but nobody
knows who the fuck dislikes thevideo. And they did away with

(29:43):
the number, and I hated that. K.
That that was just my littlerant, you know, because I don't
know why they would get rid ofthem because

Tim (29:51):
I do know what?

Mario (29:52):
Very helpful.

Tim (29:53):
Yeah. I so I and it's funny because I had another podcast
earlier today, and we got on thesame topic of, like, these
indicators of of interest. And,in my book, I do address
specifically why they eliminatedthe display of downvotes or or
thumbs downs on YouTube. And thethought is that a lot of it

(30:14):
comes down to, like, at the verysurface level, like
cyberbullying. So you never wantto be seen as liable of, like,
oh, this kid was trying to do xy z, and they were bullied by
getting a low rating, whichmakes them hate themselves, I
guess.
But the more commercial reasonis something you're going to

(30:36):
hate to hear, which is that noadvertiser wants their ad to be
played on a video that has a badratio of up to down. And so the
big thing is that like, youknow, YouTube or Google was
like, oh, okay. We're gonna stopdisplaying it. But the real
reason is probably to pander toadvertisers who don't want to

(30:58):
who start bickering over, oh,this wasn't a good video
because, you know, people arejust attracted to it now and
it's just like this brigadingthing. It's like, okay.
People are just downvoting it asmuch as possible. So that's
probably the answer. But I doaddress it in the book, and I
cite a couple of really goodsources on it.

Justin (31:17):
Follow the money. I had a friend who would always say
that. Follow the money. Itdoesn't matter if it's business,
politics, religion. Follow themoney.

Mario (31:24):
Can you come up with, like, some sort of program where
we can paste the the URL and itshows us how many dislikes or
something like that? Because,you know, the I I think that's a
million dollar idea if if youknow? And it's on the recording
here, so, you know, I came upwith it first.

Tim (31:43):
Yeah. So I hear Tim. No. I I think it's a good idea. And
the point I make is that I'vedesigned persuasive technology
systems, which includerecommendation engines.
And to get a recommendation, youhave to know what the user wants
and what they don't want. Whatbetter way than to track their
ups and their downs? So thetruth is that Google still uses

(32:05):
it behind the scenes to bettertune the recommendation
algorithms because it's what wecall an explicit feedback. It's
very, very clear what you mean.Whereas, like, just because you
watch a video all the waythrough, maybe it's because you
found it, like, absurd.
Right? Maybe it's that you justthought it was a joke. But an up
or down is actually much better,and that's why, you know, Reddit

(32:28):
probably has one of the bestrecommendation systems we've
ever seen. As far as how tosubvert when, like, platforms or
platforms collude to strip thesethings away. There have been
some efforts over the years toessentially provide an overlay,
either for explicit up downbinary feedback, but also for

(32:51):
comments as well.
So how weird is it that when youwant to comment on an article
today, you are then gonna haveto search and find it on Reddit
and then hope that somebodyreposted it to Reddit and then
engage in Reddit comments there,but you're bound by the rules of
whatever community it wasreposted in. What if there was
just an overlay in your browserthat regardless of what site you

(33:12):
were on, you could leavecomments and you could see other
people's comments? That'ssomething that's been approached
a lot of different times overthe years. And eventually, I
think it ends up coming back tolike violating some type of,
like, intellectual property orsome type of privacy thing. But
people have tried it.
And I think fundamentally, Iwould love it. But then the

(33:34):
question, of course, becomes whois going to moderate the
comments? And that's like thisyou talk about a million dollar
idea, that's the billion dollarquestion is who's gonna moderate
it once you do figure it out?

Justin (33:49):
What was the question?

Tim (33:50):
AI. AI

Justin (33:53):
is gonna be doing all of this. Speaking of computers who
have no heart and no soul, Tim,what you were doing was shady at
best. Right? Well, I mean, youeven say a villain's
perspective. What talk to meabout your your value system.
Was this what I mean, how didyou sleep at night doing this?
Did it did it bother you at all?Were you cool with it? Do you

(34:15):
regret it?

Tim (34:18):
I felt very cool in 2017. I was, you know, essentially 23
years old, working a wonderfulday job, living in a high rise
apartment in Chicago, and thenmaking all this income on top of
it, specifically working in aspace that made me more cool.
Right? If people went and I waslike, hey, I'm doing this on

(34:38):
Instagram, it was so much morecool than saying, hey, I'm a
software engineer at thisproprietary trading firm. Have
you ever heard of?
And the the answer was no. Noneof these people had heard of the
trading firm, but everyone knewwhat it meant to have more
followers on Instagram. So atfirst, there's this rush of,
like, of popularity, and youfeel the social benefits of it
in ways that are honestly quitestrange. As time went on,

(35:01):
though, I became aware of justhow far customers were going to
buy services like this and howmuch they were just putting on
the hope that they could gaintheir followers because, you
know, we would say, okay. You'regonna get this many followers
per month.
You're gonna get this muchexposure. But for them, they
thought that was the same asbuilding a successful business

(35:23):
with I wish like, I'm sure bothof you know that's not how you
build a successful business.But, of course, if you spent
thousands of dollars on it andyou still didn't have a
successful business, you wouldprobably feel like shit. And a
lot of times, the customerswould come back to us with these
ridiculous stories of of this,that, and the other thing. We're
like, hey.
If we could suddenly start asuccessful business in any niche

(35:45):
we chose, do you think we wouldbe getting people more
followers? No. We would have ourown private equity company.
Right? Like, that's that's howit would work.
So as time went on, I started tobe more aware of the fact that
people weren't really buyinglike, they really weren't
putting their best foot forward.A lot of people were spending
money they didn't have. We wereseeing cards that get declined
because they had literally zerobalances on debit cards. And

(36:09):
then for me myself, I was like,jeez. Like, this isn't what I
remember about the Internet.
Like, this isn't what I lovedwhen I started logging in in
02/2006 and 02/2007. So overtime, I think part of it is just
getting older. But the otherpart is realizing that you are
operating in violation of termsof use, and you are perhaps
accelerating some of theseunhealthy antisocial behaviors

(36:30):
in your customer base, but justencouraging them on the platform
in general by continuing tosuggest and continuing to put up
marketing material, saying morefollowers means more social
proof equals equals good. So so,yeah, ultimately, I slept at
night just fine. But once westarted to hear once we started
to become aware of actual legalproceedings against, some of our

(36:52):
competitors And then once webecame aware of just massive
shutdowns, like actualsettlements, like the settle
that's much scarier than ceaseand desist is settlement, then
it actually affected my sleep.
And in some parts of the book, Ikind of talk about this. We were
just in this conundrum where itwas really hard to run the
business. The customers weren'thappy. We were feeling like we

(37:16):
like we didn't even know why wewere doing it anymore. And then
also, you have some customerscoming to us with, like, direct
threats saying, like, I knowwhere you live and things like
And that was and that was prettymuch what and he did know where
one of us lived.
It was and that was, like, in2021 and 2022, we finally split

(37:37):
it up with saying, like, youknow, I don't think we're even
providing the best serviceanymore, and it's not worth
just, like, you know, pipping acouple hundred or a couple
thousand dollars a month fromthese people. So that ultimately
led to the shutdown, but it wasover a very long period of time
that, you know, my feelings onit evolved.

Mario (37:54):
I was curious. What what were what was the customers
essentially bitching aboutbecause they knew what they were
purchasing?

Tim (38:02):
That's complicated, because we did not explicitly tell them
what they were purchasing. So wewere selling them a organic
growth service. We did notadvertise that it was actually
software growing their accounts,but we also did not say that it
was a human growing theiraccounts. YouTube would probably

(38:25):
intuit that if you were onlypaying $35 a month and you're
getting a thousand followers,it's probably more on the side
of automation.

Mario (38:34):
Yeah.

Tim (38:35):
For a lot of people on you know, in our customer base, they
had no idea. So sometimes theywould they would be confused.
Sometimes people would sign upand then provide their login
details as a part of the sign upprocess and then say, why did
you log in to my account? Sosometimes there were just these
natural just like, people justdidn't understand what was going

(38:57):
on. And also, Mario, like, theexpectations were just
outrageous.
We were pretty straightforwardin saying, hey, on the absolute
low end, you're gonna get 300followers based on what we know.
On the highest end, we hadpeople getting 2,500 in the best
days. But everyone has a reasonto be unhappy or there was
always blowback for things thatdidn't, you know, go their way.

Mario (39:21):
And then when you shut down shut down everything, did
those all those followers justdisappear?

Tim (39:28):
No. So the followers actually were organic. So from
our perspective, it was genuinethey were genuine people saying,
oh, this account, that lookspretty interesting. I'll follow
them back. Engineers from Metateamed up with researchers from
University of California SanDiego and actually did a
research study about this in2019.

(39:50):
And they found basically thesame thing that we found, which
is that even if your account ispretty crappy, you can still get
a reciprocation rate of about8%. So if you could only put out
about a hundred follows per day,you still get eight back. You
multiply that by 30, that's like240 followers a month. Most

(40:11):
people who are just starting outand haven't reached the 10 k
mark are totally happy withthat, especially for 35, even
for $50 a month. So after theservice stopped, the followers
tend tended to stick.
They tended to stick becausethey weren't fake. They did
represent real people. But, ofcourse, over time, they there's
some decay there.

Mario (40:33):
I'll just randomly, like, click on my number on TikTok,
and I see all these randompeople just following me.
They're all hoping that I atleast I follow them back. That's
pretty much why I'm I'm gettingthese random followers even
though I barely post them in.

Tim (40:51):
Generally speaking, they would only be doing two things.
One would be they're trying toget more followers, so hoping
that you follow follow themback. Or they're serving another
purpose, and they need to seemmore real by interacting with
accounts that are notmainstream. So of course, anyone
can go and follow like acelebrity. Like, no offense,

(41:12):
you're not an a list celebrity.
But the benefit of findingsomeone like you is that you
still are considered a subjectmatter expert. So if I was to go
in forensically and say, okay,What is likely a fake account?
It's probably someone who'sfollowing Drake rather than
someone who's following Mario.And that's kind of how they can

(41:32):
construct these things. It'sactually pretty advanced of of
the the lengths the bots can goto to to make their histories
and their, you know, theirchains of engagement seem more
believable.

Mario (41:45):
Interesting. That's actually really

Justin (41:48):
good Yeah. So you wrote this book as an ego boost.
Right? You just wanted to showeverybody how cool Tim O'Harran
is?

Tim (42:05):
Justin, I think for anybody to write such a long book, ego
has to play a part in it. Andwhether it's Fair.

Justin (42:12):
Yes.

Tim (42:13):
And, you know, and whether, you know, whether it's upholding
the ego on a daily basis or it'slike reading a chapter that you
wrote and that you just knowisn't good, you need to have
some level of self belief. Andat some point, like, that self
belief can become delusional oreven, like, maniacal in terms
of, like, how firmly you mustbelieve when you don't have a

(42:37):
traditional, like, publishinghouse saying, oh, this is
wonderful. Or, like, you don'thave a supportive team around
you. I had a couple friends, andwhen I sent them essays or even
when I sent my mom essays,everyone said the same thing.
They were like, you will hateyourself forever if you do not
publish this book.
And that was how I motivated myyou know, and that was how I

(42:57):
motivated myself. And,hopefully, people can see the
the evidence there. But I think,you know, ego and importance
plays a part. But I'll tell youthis. If there were you know, I
knew personally about 10 to 12other people who were
participating in the space whohad the same roughly technical
background that I had, and mostof these people made more money
than me.

(43:18):
If one of these people decidedto write a book four years ago,
I probably wouldn't have writtenthis. As I went through and as I
tried to interview these peopleand, you know, reestablish
contact with them, I knew thatif I didn't capture these
things, nobody was ever going tocapture them. So I feel myself
as doing, you know, a service tothe Internet and the Internet I

(43:41):
remember of preserving thehistory, in a way that's, know,
fundamentally unbiased.

Justin (43:48):
Going back to your you know, I I mentioned values and,
you know, how do you sleep atnight and, you know, I'm poking
fun too. Right? But when when wefirst spoke, you said something
about this book being like anact of penance. Tell me a little
bit about that.

Tim (44:03):
I think it's accurate to say that framed is a way that I
am asking for forgiveness inperhaps not the most direct way.
We have a lot of memoirs intoday's world or even going
back, you know, to you know,decades with sports stars and
other celebrities who arelooking for a second chance. I'm
not looking for a second chancebecause I'm not a public person.

(44:25):
So at the end of the day, nobodyknows what the difference is,
and I could have just slippedinto the shadows, and nobody
would have known either way. Butfor me personally, feeling like
I was a part of a generationthat was completely railroaded
by the development of addictivetechnology and of Web two point

(44:45):
zero and of having nobody on ourside, nobody looking out for the
kids and the teenagers whenthere were it was the Wild West.
I felt like I could do somethingdecent to help the next
generation and to say, this ishow things could be different,
and here are some of the thingswe could do to protect those who
are under, you know, 13 yearsold, for example. So I feel like

(45:06):
it's it's a way of saying, yeah,I was the villain. I mainly made
money because it was cool tobreak the rules and because it
was cool to make money. Butnever did I do a single thing
that was purely villainous in itbeing destructive or it being
directly harmful. And I thinkthat is an important thing to
bring up.

Justin (45:26):
Fair. Absolutely. Yep. % agree with that.

Mario (45:30):
So, Phil, let me ask you a question. You know, for for
business owners, what whatwould, like when they're trying
to to establish themselves on onsocial media, what what does
Meta actually, like, care aboutright now? What what do they
wanna see, you know, people, youknow, from owners point of view

(45:51):
to, like, put on there that'sthat's gonna actually grow?

Tim (45:56):
Unfortunately, the main thing that they want is ad
spend. It's so much harder, togo viral or to even get
increased, exposure today, evencompared to five years ago and
especially compared to ten ortwelve years ago. What we
remember of this like organic,oh, hey, I make a post and all

(46:16):
my friends see it. Today, that'salmost, it's ludicrous because
of the way that everything isgated and the way that the feed
is constructed. It's very, verymuch built in a way to pressure
you if you have something toshare somewhat urgently to spend
money on ads.
I think the second thing thatthey're probably looking for
that a business owner can takeaway is that there are

(46:39):
increasingly strict standardsfor content. What the content
contains, what it can and can'tsay. And this is it goes well
beyond free speech, and it goeswell beyond decency. It's that
each platform has this, like,moving, you know, moving
spectrum of of enforcement andcensorship. And I think it's

(47:00):
important that people understandthat before they take a
particularly controversialstance.
I'm not a conspiracy theorist.I'm not saying left, right,
middle, who can or can't post.But I think it's worth
mentioning that before someonetries to establish themselves in
a particularly controversialspace, to also understand that
you have no control over yourvoice when you're on one of

(47:22):
these platforms. You own nothingyou create on one of these
platforms, and one day you couldwake up to it all being gone.

Mario (47:30):
Interesting. And, you know, one thing I definitely
noticed sometimes whenespecially with, like, Google
and, like, Bing, you know, adsand stuff like that, is there's
a lot of bots out there thatthat seems like it's just, you
know, like, we have other thingsand, you know, to measure the
visitors or whatever. And itseems like to me, it seems like

(47:53):
it's a waste of money, you know,putting all this money into it
because theirs is gonna takeyour money and you're not gonna
get what you're you're actuallyputting into it. What's your
thoughts on that?

Tim (48:06):
I know I said I don't get into conspiracy theories, but,
also, the opening chapter of mybook does describe framed as a
collection of conspiratorialdiatribes. And one of the
earliest diatribes that Icaptured in my book was directly
related to bots and their effecton the efficacy of ads,
specifically on Facebook. Backin 2015, I was in a class, and

(48:29):
the class was actually reallyahead of its time about
monetization of, oh, you have astartup, how do you make money?
And one thing I said at the timewas, you know, there are rumors
on the Internet that, like, 20%of all traffic that's piped
through to ads on Facebookbefore we called it Meta, 20% of
this traffic is actually bots.And as a consumer or as like a

(48:52):
business owner, somebody runningads, you really have no way of
knowing whether that's true ornot.
You just have to accept thateverybody's doing this and these
are the numbers and this is whatthey tell you you're going to
get or not. But at the end ofthe day, if there is a bot
watching your ad, there's a botwatching your ad, and they are
never going to convert. So it'sa huge, huge problem. I would I

(49:15):
I would say the the one thing wecan have faith in is that nobody
would be spending the amount ofmoney like these, like, 6 figure
monthly ad accounts and stuff.They would never be spending
that if they weren't stillgetting conversions.
I can say for me yeah. I can sayfor me with my book, I
experimented on every singleplatform except TikTok. But I

(49:36):
was on Pinterest. I did tabulaads. I did Google ads.
I even advertised on Meta, whichis probably a terrible idea. But
the point is that even for lowquality traffic sources, I sold
books. And so that tells me thatthere might be a conspiracy
there and maybe it's x percentmore expensive because some of
this traffic is just totallybogus. But at least somebody

(49:59):
bought my book, and it wasn'tone of the bots.

Justin (50:02):
Yeah. Do you think there's a a correlation between
your ad spend and your organicsearch results?

Tim (50:09):
This is a controversial one as well. Okay. There are rumors
that certain platforms willprioritize certain things based
on your ad spend in manydifferent ways. So for example,
on Amazon, one of my bestorganic days ever was a day that

(50:29):
I absolutely, blasted my spendon Taboola. And that day, I
sold, like, 16 books on Amazon,which for how early it was, that
was amazing.
But then you find out thatAmazon actually does have this
hidden algorithm where if theAmazon page is getting more
traffic, then it actually willshow higher in search results

(50:51):
even if the traffic is comingfrom an external source. So the
same point, I think, is valid onFacebook, especially where
people are saying, nobody seesanything unless I pay for ads.
Well, what if you're payingenough, but then you step away
for a little while? Might youhave some residual boost in an

(51:12):
invisible way? I would say,unfortunately, that is probably
the case where things are set upsimilar to a dating app where
you're brought so close to maybesomething that you want or
something that appeals to you.
And then immediately you'refaced with that wall of, okay,
well, now you have to pay. Andit would not be difficult to
engineer this into Facebook. Itwould be harder than to do it on

(51:35):
a dating app. But I've I'vefound that myself where I'm
like, it doesn't make sense incollege. I could post random
stuff and get, you know, ahundred plus likes.
But then posting my book, like,I got, like, five likes every
time I did it. And these arepeople who, like, really, really
know me. So, yeah, I I think,like, I think Facebook is
probably one of the craziestenvironments right now for

(51:56):
running ads where I don't seereally any way to be successful
there organically.

Justin (52:00):
Really? Interesting. Well, I'm not sure, Tim, if I'm
coming away from this encouragedor depressed. But it's
definitely interesting dialogue.You know, I'm just just kind of
dabbling in the world of socialmedia ads.
I haven't done them for quite awhile getting back into it. And

(52:22):
I'm not gonna lie. I was I waslooking at Facebook, so now I'm
scratching my head. Anyways,that that said, let's let's go
ahead and wrap this up. And Ijust wanna we'll go around the
room here figuratively speaking,of course, because we're all in
separate rooms.
And let's just just keytakeaways, you know, Mario and
then Tim, if this was allsomebody was gonna listen to,

(52:44):
what would be the most importantthing for them to hear from
today's episode? Mario, what areyour thoughts on that?

Mario (52:50):
Bet no matter what you would do, you're kinda fucked
unless you spend enough money onthere. And, you know, there
there's gonna be a a good amountthat kind of goes to bots. You
know? But, you know, at the endof the day, it's you're you're
you're praying that it's goingto automatically bring up some

(53:12):
organic, you know you know,algorithm that it's gonna help
you either directly orindirectly. And

Justin (53:21):
if

Mario (53:21):
you don't spend any money, you're not getting it.

Justin (53:25):
Fair enough. Tim, what are your thoughts?

Tim (53:28):
In my book, I describe social media as a tragicomedy.
Meaning, it's kind of funny, butit's also a very, very terrible
place. And I think what makes itworse today is that we have
legions of fake pundits. And,hopefully, people come away from
this chat not thinking that I'mone of them. But you have

(53:49):
hundreds, if not thousands ofpeople on the Internet, on
social media, proclaimingexpertise for things that they
know absolutely nothing about.
So what I would leave listenerswith is, hopefully, you read my
book and you realize that I amone of the people who does know
what he's speaking about. Butfor anyone who maybe doesn't

(54:10):
wanna read the full 400 pages,my advice is to actually spend
less time engaging with thesepeople and to hinge less of your
hopes and your business goals onsocial media because it's
becoming more and more viciouswith each day. And I think my
book can get people a little bitcloser to realizing that truth.

Justin (54:29):
I love that. I was in a meeting this morning and they
were talking about what what'sworking in the world of
marketing and what's not. And,mean, honestly, we're we're kind
of coming back full full circlepost COVID to live events, human
interaction, you know, with allthe AI, with all the bot, with
all this horse shit, thetechnology, you know, it's
supposed to make our livesbetter and it's just Jesus, it's

(54:49):
complicating things to no end. Imean, maybe it's time for us to
just go grab a beer and talkbusiness that way. That's kind
of my takeaway from this.
Tim, really, really appreciateyour time here today. Tell me
what with this book, what's kindof your vision? What's your
goal? What are you all aboutthese days as you're kind of

(55:11):
transitioning out of thatbusiness? Now you're writing.
What is it you're trying toaccomplish?

Tim (55:17):
In writing the book, I was really interested of you know,
in preserving the history of theearly Internet and even the
things that I say were onlycaptured by serious weirdos or
other types of ethnographers. Aswe see local journalism lose
funding and lose interest Mhmm.And basically lose viewership, I

(55:40):
think every single person who'sbeen on the Internet for more
than a year can do their partinto helping preserve their most
fond memories, what theyremember, whether that was last
year, whether that was in02/2005 or sometime in between.
I think we have some duty hereto preserve what's been talked
about. And then at some point,of course, in the future to
debate it and to reach greaterunderstanding.

(56:01):
So my book hopefully is a bigstepping stone in that. And
people can come away with thesame things that I enjoyed when
I first logged in, which wasthat the Internet was this very
diverse place where you can logon looking for one thing and
come away, you know, engaged andengrossed by something totally
different. I I have a real hopethat the Internet can still be
that type of destination formany people.

Justin (56:24):
Okay. Perfect. And I I think if I remember correctly
from our first conversation, doyou have some sort of a target
of how many of these books youhope to sell?

Tim (56:35):
Yeah. So the goal for this yeah. Sure. I I do. This goal is
probably about a thousand booksfor this year.
I'm very happy. I yeah. I'm veryhappy. I'm right about halfway
there. It's over

Justin (56:47):
Nice. Four

Tim (56:48):
yes. It's over 400 at this point, and that's kind of what's
motivating me to work on theaudiobook. Of course, it's not
paying my my rent here in NewYork City, but it is something
that I'm offering well, wellbelow the cost of other of of
like books just because I wannaget the word out. So I think a
thousand is a good startingpoint. And once the audiobook's

(57:10):
out, we'll we'll see what'snext.

Justin (57:12):
Oh, I I am an audiobook fan. I I promise to read this
book. I'm going to read it, butit would sure make my life
easier if I was listening to it.I'm holding this up. If
anybody's watching Framed,villain's perspective.
I'm trying to move my finger outof the way. Mean, framed a
villain's perspective on socialmedia. Tim, again, thank you for

(57:33):
being here. Brilliant stuff.Myself.
I I like to believe that I spenda lot of time in a room full of
people smarter than myself.You've helped me accomplish that
today. So thank you for that.Mario, always a pleasure
chatting with you. There again.
What you've done with yourbusiness is phenomenal. Mario,

(57:55):
Tim, Mario just a year ago wonthe largest competition in the
IT industry. In fact, the prizewas a car. What kind of a car
did you end up with?

Mario (58:05):
Well, Land Rover Defender. Wow.

Justin (58:08):
Like a serious competition. Mario won that. So,
I love hanging around withpeople who are brilliant in
their areas of expertise. Sothank you both for being here.
And with that, guys, we're gonnago ahead and say goodbye and
wrap up.
We'll see you all next week.Take care.

Mario (58:23):
Thank Justin, thank you, guys.

Justin (58:26):
Thanks, Tim. We'll see you guys.