Cut The Tie Podcast with Ryan Miller
Ryan Miller has lived on the front lines of sales. Cold calls. Door knocking. Endless outreach. The grind most people never see but everyone feels.
Then the game changed.
In this episode of Cut The Tie, Ryan explains why modern outreach is broken, how over-automation has destroyed trust, and why AI only works when it amplifies human judgment instead of replacing it. From building an AI-powered outreach platform to leaving the U.S. and relocating to Brazil, Ryan shares what it really means to scale without losing credibility.
This is a conversation about cutting ties with spammy volume, outdated sales playbooks, and the illusion that more messages equal better results.
About Ryan Miller
Ryan Miller is the founder of Zenith Project, an AI-driven outreach platform designed to help businesses create real conversations instead of automated noise. With a background in hands-on sales and deep experience in automation, Ryan focuses on building human-in-the-loop systems that balance efficiency with authenticity.
Now living as an expat in Brazil, Ryan is focused on helping founders and agencies modernize outreach without sacrificing trust, relevance, or relationships.
In this episode, Thomas and Ryan discuss:
- Why mass outreach and copy-paste automation stopped working
- How spam destroyed trust in modern sales
- The difference between AI assistance and AI replacement
- Why personalization without intent feels fake
- Breaking sales conversations into smaller, evolving objectives
- The danger of removing humans entirely from outreach
- Transitioning from a service business to scalable software
- Using AI to create clarity instead of noise
Key Takeaways
- Volume no longer creates leverage
Relevance beats reach every time. - AI without judgment erodes trust
Automation must support thinking, not replace it. - Spam signals indifference
Buyers instantly know when a message wasn’t actually read. - Human oversight is non-negotiable
The best systems still require people in the loop. - Scaling requires cutting old playbooks
Yesterday’s sales tactics don’t survive today’s market.
Connect with Ryan Miller
🌐 Website: https://zenithproject.co
💼 LinkedIn: https://www.linkedin.com/in/ryanmillerhq
📧 Email: ryan@zenithproject.co
Connect with Thomas Helfrich
🌐 Website: https://www.cutthetie.com
💼 LinkedIn: https://www.linkedin.com/in/thomashelfrich
📧 Email: t@instantlyrelevant.com
🚀 Instantly Relevant: https://www.instantlyrelevant.com
Serious about LinkedIn Lead Generation? Stop Guessing what to do on LinkedIn and ignite revenue from relevance with Instantly Relevant Lead System
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_01 (00:00):
Welcome to the Cut the Tie Podcast.
Hello, I'm your host, ThomasHelfrick, and I'm on this
mission to help you own your ownsuccess by cutting the ties to
whatever it is holding you backin life.
Um I'm I'm really passionate asan entrepreneur myself that
people do this because it's it'svery disappointing as people
kind of get to where they wantedto go in life and they realize
that, oops, that wasn't what Ireally wanted.
(00:21):
And there's lots of stuff totell them back from it.
So as we meet entrepreneurs likewe are today with Ryan Miller.
Ryan, how are you?
I'm doing wonderful.
Thanks for having me.
You know, as we meet guys likeyou, we're gonna learn about
your journey and who you are.
So let's start with that.
Who are you and what is it youdo?
SPEAKER_00 (00:37):
Thanks, Thomas.
Uh, my name is Ryan Miller.
I my background's in sales, andI about two years ago realized
that I think I could automate uhthe worst part of sales with AI.
So I started an AI company.
Um, we have a startup, uh, we dooutreach, and it's all AI
driven.
Um, that also allowed me toleave my home in Florida and
(01:01):
move to Florianopolis, Brazil.
And so I live in Brazil rightnow.
I'm an expat.
And uh we are making hugeadvancements in in how outreach
can be automated and howefficiency can be driven and um
really changing the game of theeffectiveness of outreach.
So that's that's me in anutshell.
SPEAKER_01 (01:21):
I mean, that it I think it's cool you're doing the
uh, you know, the live somewhereelse, different cost basis,
different experience in life.
I think that's beautiful.
Uh you know, lead generation isyou know, we're in this, we're
in the game to some degree withit, with consulting and some
execution.
It's it's competitive, but it'sa weird way to say competitive
because there are differentvariants.
There's people who do like$30 amonth fully automated, and it I
(01:43):
don't know what that is and ifit works.
And then there's consultanciesthat charge tens of thousands.
Uh where do you guys fall in theniche of this?
SPEAKER_00 (01:52):
That's a great question.
You know, uh, one thing I'velearned over the past two years
of doing outreach, um, and I wasdoing outreach far before it was
automated.
You know, I did outreach, I didcold calling, I didn't door
knocking.
Uh, if I had a dollar for everytime a door has been slammed in
my face, you know, uh, I'd havea lot of dollars.
And um, I've come to realizethat everyone these days, or a
(02:16):
lot of those cheaper options,are using kind of the old IBM
playbook.
Um, it was IBM, it wasSalesforce.
It's the playbook of if we sendout enough messages and get it
in front of enough people, we'regoing to find those people that
have enough pain that they willwork with the first person that
you know comes through theirdoor.
(02:38):
And that isn't a good strategy.
I mean, it worked for them, um,but these days everyone is using
that strategy.
Everyone is spamming everyone.
And I don't know about you,Thomas, but I get about a
hundred messages per day in myLinkedIn with people shoving a
pitch down my throat.
And they all are saying the samethings, and I ignore all of
(03:01):
them.
And so Zenith Project is tryingto cut, break that mold, and I'm
sure you guys are doingsomething similar with your
outreach service.
Um we reach out based on theirposts.
We try to personalizeeverything, everything.
There's zero templating atZenith Project.
Everything is uh AI driven.
So we use AI to help personalizeit.
(03:24):
Um, but then everything is humanreviewed because if you don't
have that human element and andkeep a human layer in between
your messaging and the personreceiving that messaging, you're
gonna go right into the the thespam bucket along with everyone
else reaching out with that IBMstrategy.
Do you do you see somethingsimilar in your experience?
SPEAKER_01 (03:46):
I I see a lot of mass emailing.
I actually see a lot of over AIautomation going on too.
Uh we used in the tech.
Uh I I I turned on the AIautomation one day to see how
the thing replied, and I forgotabout it for like a week.
And it came in and like, man,that thing made me sound like an
expert in employer branding.
I'm like, holy cow, I what isthis?
And I even just emailed the guy,I'm like, my AI took off here.
(04:08):
Here's my calendar, like if youwant to chat.
And he's like, ah, okay.
I was like, man, I have like thethe technology doesn't sometimes
give me the boundary to say,hey, limit your replies to two
or three, but push them to thisone thing, right?
Like, like answer the call toaction and say, hey, listen, why
don't we just have aconversation?
Like, just keep pushing peopleto my link.
(04:28):
Like it, the some of the textout there.
So I'm seeing this across theboard.
Um, I'm actually okay being AIautomated out in the outreach
when I see it, as long as thekind of message sequencing is
fairly like clear.
Like, listen, I'm just reachingout to people.
This is what I do.
If you want to find time, great.
Otherwise, uh, here's somecontent I'll post.
I'm cool with that message allday long.
Yeah, all right, I'll jump in.
If I have an interest, I'll jumpwith you.
(04:50):
But but it's when it's like toopersonalized, or like like no
one cares that much on a firstit's like that doesn't work
either.
SPEAKER_00 (04:57):
So that's true.
Um, we we've had to tone downsometimes the AI.
Some of the things we see a lotis the AI is very enthusiastic,
uh, and it will pull in all ofthe context that it knows about.
SPEAKER_01 (05:10):
Um, and one of the things I get a pub before my
speech recently you lost a childand have cancer, and and how we
tell you how that relates toyour post on nanoscience.
SPEAKER_00 (05:18):
Like, I know.
Well, and that that's the memeon LinkedIn.
Whereas like, oh yeah, my myparents both died in a car
crash.
This is what it taught me aboutsales.
SPEAKER_01 (05:29):
Like, that's that's not the kind of back up like
four years, we were we were onOpenAI's beta before it was chat
GPD, and we were turned thisthing on for automated
commenting.
And and I looked at one of mycomments one day and it said
that my comment literally was mydad was also an NFL star and
died in a plane crash.
I'm like, my dad's alive and wasan optometrist.
(05:51):
Um, it was like, holy cow, liketurn that thing off.
But the funny thing is, I stillsee some kind of version of that
in this current AI where it justtry to like enthusiastically
relate.
Uh so talk about your strategy alittle bit of how you're the
reason I bring that up, not forjust for humor context, but for
what you guys do, uh is youknow, you have you talking about
(06:12):
your journey in sales a littlebit, but tell me how you're
solving that a bit because thatthat's that's the nuance, I
think, where it happens, whereit works, or it doesn't.
SPEAKER_00 (06:20):
Yeah, and I'm happy to uh uh pull back the curtain a
little bit.
Uh everyone listening, feel freeto steal this idea.
Um, so what we realized is thatAI is very good at accomplishing
small goals.
You know, it's you've probablyall heard, you know, uh it's
like a you know, a new intern.
You can give it small tasks andit completes those rather well.
(06:42):
And so the the concept is tryingto break a sales conversation up
into smaller tasks where theobjection, objective is actually
evolving throughout theconversation.
So at the start of a salesconversation, a cold sales
conversation on LinkedIn, you'regonna have a very different
objective for the AI messagegeneration than at the end.
(07:05):
Because if you try and have thesame objective throughout the
conversation, it's just going tobe falling back on the pitch,
basically.
It's gonna say, I know this ishow I can get convinced people
to set a meeting.
I'm gonna send this messaging.
That's not what you want to do.
You wanna start by getting themin involved in the conversation,
(07:26):
asking curiosity-drivenquestions, building rapport, and
then pivot that to asking abouttheir challenges and try and
guide them to the subject thatyou want to talk about.
So we have built a structurethat actually feeds these
evolving objectives to the AI.
And we have another AI agentthat's kind of managing that
(07:49):
process where it's trying tounderstand in the conversation
where it's at and where thatother AI should go.
SPEAKER_01 (07:57):
It it's interesting.
Uh, it depends on your I thinkyour ICP a bit, your ideal
customer profile, because Idon't want to have a back and
forth conversation on it.
I want someone just to get toit.
Other people do.
And and so I think some so howdo you hand any of the
contextual data for that?
Because that's because Ihonestly, if you did that for
me, I usually reply back, hey,you need to work on your
sequencing.
I have no interest in the fivethings you just sent me.
SPEAKER_00 (08:18):
Yes.
SPEAKER_01 (08:19):
Just tell me what you do, and I'll tell you if I'm
interested in it now.
Like, like uh, talk to me aboutyour kind of AI piece with that,
because that that personamatters.
Uh, because the person who doeswhat you just described will
never get me out on call, butthe person who's very brief and
probably will.
SPEAKER_00 (08:34):
So no, and and you hit the nail on the head.
Uh, one of the ways we built itwas uh to be flexible for the
client's needs in the sequencinglength.
So we can actually set up thesequence to be two messages.
One nice, warm and fuzzyrapport-building message, and
then a straight talk.
Here's what I do well.
(08:55):
This is how I can help you.
What do you say?
We hop on a call.
About half of our clients usethat to great success.
But in sales, I'm you you're insales as well.
You know how important mirroringis.
There are also some of yourprospects that are gonna fall
into that um, you know, the theisolated, what do you call it,
(09:16):
um, the exclusionary categorythat actually do want to have
the conversation and they don'twant the direct pitch.
They don't want to hear aboutit, they want to chat about what
they're posting about orsomething before they get to,
okay, now I'm ready to hear whatyou're all about.
And so even for those clientsthat have the short cadence of
(09:36):
sequence, um, it still helps tohave the human oversight because
the human can step in and say,you know, it doesn't feel like
the the vibe in the room is aright to pitch, you know, our
normal short straight pitchmessage already to spot in the
conversation.
SPEAKER_01 (09:55):
What one of the uh greatest connection requests
automations ever ran was onlythis.
All I wrote was ooh, oh likefour O's and an H.
Dun dun dun.
Tell me more about company.
SPEAKER_00 (10:07):
Ooh.
Tell me about this.
Tell me more about ABC Company.
SPEAKER_01 (10:10):
Yeah, just tell me it's ooh, tell me more about
this.
Which gives them kind ofinterest.
Like, oh, are they interested inme?
What's going on?
And so uh I've ran that one toeffect.
I've also, I'll tell you what,the other one I would tell
people to take this is if youdon't know, it's better to say
nice to connect.
Do you want the long or shortversion of me?
Introduction.
Yeah, and like that.
It invites them to say noversion or short version, cool.
(10:35):
Quick reply, right?
Long version, great.
Um uh or do you want to pitchslap me or should I pitch slap
you?
Something like that.
SPEAKER_00 (10:45):
Like like I think the place the place that would
miss is that the the people thatdo appreciate rapport building
pivoting to a salesconversation, but they wouldn't
accept it if you offered it tothem.
Because if you said, Hey, do youwant to have a quick, you know,
three to five messages back andforth building rapport, and then
(11:05):
I'll tell you about my company,they're gonna say, you know, I I
don't want to go through thatprocess.
SPEAKER_01 (11:10):
Now, if you just there's a there's a group that
would be like, Hey, do you wantthree to five messages of us
rapport building, or would yourather just network for 10
minutes on a call?
Yeah, that like something likeyeah, like look at we can decide
to disconnect if it if itdoesn't make sense, though.
That is effective.
Now, also some that eliminatesthe serious side, to be fair.
SPEAKER_00 (11:29):
The point is you can AI they can play your each way.
That will get you meetings, butnot necessarily qualified
meetings.
And that's another thing that wetry and do is we try and not
waste our clients' time becauseour clients' time is very
valuable.
Um, and so we don't want to justset them networking meetings
that might go somewhere.
We need to uncover in thatsituation, in that conversation,
(11:51):
the fact that they needsomething from us and that we
can provide it, we can solvethat problem for them.
SPEAKER_01 (11:57):
Now, this goes and that's a really good valid
point.
So in our ICP, our stuff istypically we use the podcast as
uh as a top of the funnel formany.
So I'm knowing their positionand the fact that they're they
have some level of success, wecan help them.
It's just they're gonna theyneed it or not.
And that but we with the podcastin place, it it still helps
because it is it is a there's apiece that they're gonna always
(12:18):
have that conversation.
And it's usually never wasted ifmy yeah, they'd still be a good
guest.
They have this.
So it there's a lot ofvariables, I guess is my point
that goes into this.
Um, let's just take a pivot herefor a minute for you.
Just in your current business,what's kind of the biggest
metaphoric tie that you'restruggling with right now?
I think to kind of cut, getover, get, you know, to go next
level.
SPEAKER_00 (12:38):
The biggest tie is honestly being a service
company.
Um, I do not want to be aservice company anymore.
I want to pivot into being asoftware company.
And that will cut all sorts ofties because, you know, the
valuation of software companies,the scalability, the, the ease
of running, the requirement ofstaff.
(12:59):
Um, and we are very close tothat level.
You know, you and I have hadconversations, Thomas, before
about white labeling my softwarefor your agency, other agencies,
and they using their labor asthe copywriters, because that's
kind of integral to the ZenithProject offering.
You have that human in the loop,but other agencies can use
(13:22):
Zenith Project software to getthis massive level up uh with
the AI suggesting 80% of thehental uh heavy mental uh
gymnastics that you have to doto come up with very clever and
strategic sales conversationmessaging.
Yeah.
SPEAKER_01 (13:41):
Yeah.
And I think so.
I think the the we had talkedabout this a little offline.
So you could be at just anotherlead gen company.
Uh what you guys are developingspecifically to me, and and I
just kind of we're openly justkind of talking about the
solution of this, what you'redoing.
I think what you could offer toother AI lead gen companies as a
plug-in for theirs would be thebetter play.
(14:02):
Because you're like, hey, weactually give you the contextual
stuff beyond what's going to bein anything you're doing.
So your stuff just works better,like a middle layer, if you
will.
Uh, and then that way you're notever in trouble with any of the
social media platforms foryou're just taking data and
you're doing something with itto give advice to somebody else
trying to do something.
That middle layer becomesdefinitely makes you more, you
(14:22):
know, uh viable.
And and people can like, youknow, obviously, hey, listen,
you know, it's eight dollars peruser that you you have on yours,
and now you you have these, youknow, de-risk revenue micro all
over, and they're just pluggingin your stuff to make their
stuff more effective.
I think that there's a there's amarket for that because some of
the stuff we described is sonuance-based.
Like, who's Thomas?
Who's Ryan?
(14:42):
Who wants this?
What mood's he in?
What's his post been?
That they've been more aboutthis or that.
The the point is if you can getthat nuance really narrowed down
to what that person might like,that'd be humorous, are they not
like all these things?
What's their current intent?
That that would be applicableagainst anybody's own version of
their services of it of it.
Anyway, so I I think as youlook, you know, it's the
pickaxe, right?
For the gold rush, if you will.
(15:03):
Uh there's uh rush, you know, Iyou and I talked about this too.
Like, we're I'm pivoting morelike I just want to verify
companies can do what they saythey can do because there's so
many who can't.
Yeah, you get burnt.
And so like that, so you look atthese pivots, but how do you
feel about that?
Like kind of being a you canhave your own version of the
thing, but like you can havetechnology that could be applied
(15:25):
to all the Apollo data that'sout there, all the seamless data
that's out there.
Yeah, I mean, like so so I gotbuild is something more valuable
than just lead gen.
SPEAKER_00 (15:34):
I I'm a fan of vertical vertical integration,
you know.
I if if I can capture every partof the the the customer
lifecycle, then we're gonna trybecause you know having all of
that in one ecosystem is gonnabe way more convenient for the
client.
Um plugging our software intosomething like Apollo is
(15:56):
actually gonna be so easy, justbecause that's a an API feed
between the databases.
Um, and that is the vision forZenith project, is it's going to
basically be a CRM that you haveall of your enriched data flow
into, and then we do yourLinkedIn, we do your email, we
do your Facebook, we do yourtexting, and all of it has that
(16:18):
human layer in it because AIisn't ready to automate yet.
Like when I get an AI messageand it's like two paragraphs of
AI text, I don't read it and I'moffended by that person because
they didn't read it either.
I know they didn't read it, andthey're expecting me to take
five minutes to to to writetheir AI message.
(16:40):
It's over.
SPEAKER_01 (16:41):
I tell it drive me nuts too.
Like this literally justhappened before my meeting here.
Like, I'm trying to meetsomebody to help us with a
client for employer branding,just what it is.
And the person was like, Hey,I'll give you a call, or and I'm
like, I never do phone calls.
I didn't even anyway.
And then I'm like, they're like,Why can't you call talk for two
minutes?
And I'm like, I'm about to go ona podcast.
And they're like, that'ssketchy.
And I'm like, you want me tocall you for a long-term
(17:03):
partnership because you're in acar.
I'm like, set up time, get withme face to face.
It's like some people have thislike sketchiness, this kind of
like the they they they've likethey read into what you do too
much based on based on yourinitial interactions, right or
wrong.
This this person who did this,but I'm kind of like, you give a
perspective.
(17:23):
If you send an AI messageinitially, and it's clearly like
so in depth and perfectlywritten, it it's a dehumanized
without calling to the fact thatit's been dehumanized.
You're kind of like, oh,whatever.
Like this is just yes, I'm notreally important to you, is what
you're communicating to me.
Yes.
And and honestly, like Rubylowercase typos, I throw them
(17:44):
all in there.
Like, listen, I'm actuallywriting this shit.
I'm not, but my AI writes me intypo format.
SPEAKER_00 (17:49):
When I see one of my copywriters make uh a spelling
mistake, I'm like, good, thatdoesn't look like AI, you know.
SPEAKER_01 (17:57):
I use AI to say, listen, all lowercase uh use dot
dot dot instead of dashes, uh,throw occasional text version
typo in there.
You are versus, you know, usethe like write it like you did
in text, like I wrote for mymobile phone.
SPEAKER_00 (18:11):
That's good.
I I also one of the biggestrules for my copywriters, and
this is for everyone that's youediting AI text, is deletion is
your friend with AI.
AI will include context that thereader already knows and that
isn't needed to be saidexplicitly.
Um, and it will also justelaborate on things that need no
(18:34):
elaboration.
And so if you take a an eraserto AI text, it's gonna seem way
more personal and human.
SPEAKER_01 (18:41):
The uh the pros of AI sometimes, like, dude, just
say what you have to say and bemove on, like right.
Yeah.
Um but I will tell you no matterhow you do it, AI is an
incredibly effective tool forhelping you with brevity.
If you say, just give me brevityon this piece, it's still
probably longer and moreeffective and less ums and ems
and thems and the words thatdoes it.
Um talked about though.
(19:03):
This this is just from the fromthe purposes of staying on brand
or a podcast of cutting ties.
You're you don't want to be aservices company, shocker.
Let's talk about that a littlebit.
There's a lot of people outthere that are you know, it's
it's how you get startedtypically as an entrepreneur.
You start doing things yourselfthat you were gonna get paid for
someone else paid you to do.
Totally normal.
(19:24):
Um tell me how you're makingthat move though.
Like it's more than thing justto go to technology.
Are you do you have mentors?
Like, what actually happens foryou to go from services to more
of a SaaS?
SPEAKER_00 (19:36):
Great question.
Great question.
My main mentor, uh, who you arefamiliar with, Lee Coulter, um,
he has been pushing me to gosoftware, the software route
since the beginning, since Itold him about my idea to create
this AI driven outreach company.
Um, so I definitely have out uhmentors in the software to
(19:56):
service to software space.
Um, the other thing.
Is that proof of concept isreally necessary when selling a
software.
You have to be able to showpeople that they can recreate
the results that you createdwith the software.
And if you haven't proven itout, then you can't show them
anything.
And so that's one of the reasonswe've been a service company for
(20:19):
the past year and a half.
And now we have incredibly richresults and testimonials from
our clients who have used ourservice when we were using our
software and have seen greatresults from it.
And so I can show that toagencies and say you can
recreate these testimonials.
You can be a franchise,basically, of Zenit's project.
(20:41):
It's similar to how you knowMcDonald's is the corporation of
McDonald's is not a restaurantcompany.
They're a franchise company.
Oh, they're a real estatecompany.
Yeah, they are.
SPEAKER_01 (20:53):
It's a real estate company that has things on it.
SPEAKER_00 (20:56):
This is true.
This is true.
That's that's how Burger Kingactually figured out how to put
their restaurants, is they justput them near McDonald's because
McDonald's already did all theresearch on the real if you want
to go figure out where to go buyhouses or buy land, go find
where the next quick trip is.
SPEAKER_01 (21:12):
And for the next uh you see Walmart, 20 years from
now, that place will have 20x topeople around it.
So go get go get a lot of fiveacres somewhere because
someone's gonna want to go buyit for something because they
they exactly it's like exactlyand yeah, go ahead.
You want a rehab, go find aStarbucks that's in a location.
Like, why is that here?
Because there's adegentrification coming to this
(21:34):
area they already know about.
And there you go.
SPEAKER_00 (21:37):
They they've done all the research, they have the
teams, they have the resources.
Uh, believe them.
SPEAKER_01 (21:43):
That's one of my favorite phrases.
Sorry, Ryan, conscious timehere.
Just tell first of all, peoplewhere they should go to check
you out, check out the product,uh, and and and you know, get a
hold of you.
SPEAKER_00 (21:51):
Absolutely.
So zenithproject.co uh is ourwebsite.
You can feel free to go readabout us there.
Um, you can connect with me onLinkedIn.
My uh handle is Ryan Miller HQ.
Um and my email is Ryan atzenithproject.co uh and you can
reach me there.
(22:11):
Uh those are great ways.
All right.
SPEAKER_01 (22:14):
Let's leave the listener with like uh just three
pieces of advice.
Something you should always dowhen leveraging AI and reach
out, one and something youshould never do, and then
surprise me with the third.
SPEAKER_00 (22:26):
Okay.
So something you should alwaysdo with outreach of AI is read
the message before you send it.
And read read it, read it outloud.
Repeat that.
Yes, read the message before yousend it.
Because AI does hallucinate, andyou might get into some pretty
embarrassing situations withwhat AI says on your behalf.
(22:49):
Um, or have someone read it foryou, you know, like CF Ride.
Yes, absolutely.
Yes.
Um, so something to never dowith um AI outreach, I would
say, is uh use out-of-the-box umChat GPT messaging.
Like put some thought into whatprompting, what what iterations
(23:11):
of the AI messaging you want toget to.
Because if you use the generic,you know, if you use M dashes uh
in your messaging, you're thatis going to come across really
bad to potential prospects.
They're used about GDPT, right?
SPEAKER_01 (23:24):
No matter how many times it's helped to stop doing
that, it still puts them inthere.
I'm like, can you never use adash again and leave all emojis
dead forever?
And it's like, yeah, here'semoji with a thumbs up.
SPEAKER_00 (23:35):
I'm like, god damn it.
The the M dash is, I don't knowwhy AI likes it so much.
SPEAKER_01 (23:39):
I think it's actually uh proper, proper uh
grammar is why I've oh it wasfun new phenomenal.
SPEAKER_00 (23:45):
Yeah.
SPEAKER_01 (23:46):
Someone who like uh what was a my co-founder, his
wife is like a linguist orwhatever, and she's like, that's
actually the proper way to doit.
No one ever does it.
So now her her stuff looksGPT-ish.
I'm like, well, you gotta quitdoing it.
SPEAKER_00 (23:58):
I know.
I you I used N dashes, sothere's three types of dashes
there's n dashes and m-dashes.
M-dashes are basically used inplaces of commas or to um you
know split a sentence fragmentout.
Um, and they they're phenomenal.
They're so nice to read.
Uh they're they're wonderful towrite if you know how to.
(24:19):
They're kind of hard to, but um,but yeah, I I kind of miss using
m-dashes, but I I don't I don'tuse them anymore.
And I have all of my AI'sprograms to not use them.
SPEAKER_01 (24:30):
All right.
So okay, so one is to read it.
What was the thing you should uhis that always do or never do?
SPEAKER_00 (24:35):
Not do?
Yeah.
Yeah, the the thing you shouldalways do is read it.
The thing you should not do isuse m-dashes uh or use
out-of-the-box AI.
You know, uh definitely developyour own AI.
The third thing to uh surpriseyou.
Um well, let's see.
Does it have to be about AI orcan it be about MI?
(24:57):
I mean, it'd be cool if it was,but you know.
Okay, let's let's think of an AIone really quick.
SPEAKER_01 (25:02):
Give me like some maybe maybe a surprising thing
you didn't know you could dowith AI that you're some people
most some of your customers arelike, oh, I didn't know you
could do that.
Like, what would be an answer tothat?
SPEAKER_00 (25:12):
My my favorite use of AI is um reformatting massive
sets of data.
I this is kind of specific to meanyway, because I'm a data nerd,
but you can put in massiveamounts of data and ask it to
reformat in a specific way, andit'll do it perfectly.
Um, so you know, datareformatting or slight changes
(25:33):
to massive amounts of text issomething that AI does for
breakfast.
SPEAKER_01 (25:37):
Yeah, I I'll extend that.
You know what I use AI for is Ihave it I export our
transactions and our accountingand books and say, look at this,
and what do you see?
Like what what what do you seehere?
And what what's it cost to runthis company and when did when
should we hold reserves?
It's amazing.
So um uh listen, Ryan, thanksfor so much coming on today.
(25:59):
I appreciate it.
Absolutely uh get a hold ofRyan, Ryan, give him your uh
your place to go one more timeif you don't mind.
Zenithproject.co thanks Ryan forcoming on today.
I appreciate you.
You too.
Thanks, Thomas.
Listen, everyone who's stilllistening, get out there, go cut
a tie to whatever's holding youback.
Make sure to follow the podcast,subscribe, and if you really
liked it, do five stars, nothingless.
(26:19):
Five and only five.
That's all we accept.
Okay, thanks for listening.
Welcome to the Cut the Tie Podcast.
Hello, I'm your host, ThomasHelfrick, and I'm on this
mission to help you own your ownsuccess by cutting the ties to
whatever it is holding you backin life.
Um I'm I'm really passionate asan entrepreneur myself that
people do this because it's it'svery disappointing as people
kind of get to where they wantedto go in life and they realize
that, oops, that wasn't what Ireally wanted.
(00:21):
And there's lots of stuff totell them back from it.
So as we meet entrepreneurs likewe are today with Ryan Miller.
Ryan, how are you?
I'm doing wonderful.
Thanks for having me.
You know, as we meet guys likeyou, we're gonna learn about
your journey and who you are.
So let's start with that.
Who are you and what is it youdo?
SPEAKER_00 (00:37):
Thanks, Thomas.
Uh, my name is Ryan Miller.
I my background's in sales, andI about two years ago realized
that I think I could automate uhthe worst part of sales with AI.
So I started an AI company.
Um, we have a startup, uh, we dooutreach, and it's all AI
driven.
Um, that also allowed me toleave my home in Florida and
(01:01):
move to Florianopolis, Brazil.
And so I live in Brazil rightnow.
I'm an expat.
And uh we are making hugeadvancements in in how outreach
can be automated and howefficiency can be driven and um
really changing the game of theeffectiveness of outreach.
So that's that's me in anutshell.
SPEAKER_01 (01:21):
I mean, that it I think it's cool you're doing the
uh, you know, the live somewhereelse, different cost basis,
different experience in life.
I think that's beautiful.
Uh you know, lead generation isyou know, we're in this, we're
in the game to some degree withit, with consulting and some
execution.
It's it's competitive, but it'sa weird way to say competitive
because there are differentvariants.
There's people who do like$30 amonth fully automated, and it I
(01:43):
don't know what that is and ifit works.
And then there's consultanciesthat charge tens of thousands.
Uh where do you guys fall in theniche of this?
SPEAKER_00 (01:52):
That's a great question.
You know, uh, one thing I'velearned over the past two years
of doing outreach, um, and I wasdoing outreach far before it was
automated.
You know, I did outreach, I didcold calling, I didn't door
knocking.
Uh, if I had a dollar for everytime a door has been slammed in
my face, you know, uh, I'd havea lot of dollars.
And um, I've come to realizethat everyone these days, or a
(02:16):
lot of those cheaper options,are using kind of the old IBM
playbook.
Um, it was IBM, it wasSalesforce.
It's the playbook of if we sendout enough messages and get it
in front of enough people, we'regoing to find those people that
have enough pain that they willwork with the first person that
you know comes through theirdoor.
(02:38):
And that isn't a good strategy.
I mean, it worked for them, um,but these days everyone is using
that strategy.
Everyone is spamming everyone.
And I don't know about you,Thomas, but I get about a
hundred messages per day in myLinkedIn with people shoving a
pitch down my throat.
And they all are saying the samethings, and I ignore all of
(03:01):
them.
And so Zenith Project is tryingto cut, break that mold, and I'm
sure you guys are doingsomething similar with your
outreach service.
Um we reach out based on theirposts.
We try to personalizeeverything, everything.
There's zero templating atZenith Project.
Everything is uh AI driven.
So we use AI to help personalizeit.
(03:24):
Um, but then everything is humanreviewed because if you don't
have that human element and andkeep a human layer in between
your messaging and the personreceiving that messaging, you're
gonna go right into the the thespam bucket along with everyone
else reaching out with that IBMstrategy.
Do you do you see somethingsimilar in your experience?
SPEAKER_01 (03:46):
I I see a lot of mass emailing.
I actually see a lot of over AIautomation going on too.
Uh we used in the tech.
Uh I I I turned on the AIautomation one day to see how
the thing replied, and I forgotabout it for like a week.
And it came in and like, man,that thing made me sound like an
expert in employer branding.
I'm like, holy cow, I what isthis?
And I even just emailed the guy,I'm like, my AI took off here.
(04:08):
Here's my calendar, like if youwant to chat.
And he's like, ah, okay.
I was like, man, I have like thethe technology doesn't sometimes
give me the boundary to say,hey, limit your replies to two
or three, but push them to thisone thing, right?
Like, like answer the call toaction and say, hey, listen, why
don't we just have aconversation?
Like, just keep pushing peopleto my link.
(04:28):
Like it, the some of the textout there.
So I'm seeing this across theboard.
Um, I'm actually okay being AIautomated out in the outreach
when I see it, as long as thekind of message sequencing is
fairly like clear.
Like, listen, I'm just reachingout to people.
This is what I do.
If you want to find time, great.
Otherwise, uh, here's somecontent I'll post.
I'm cool with that message allday long.
Yeah, all right, I'll jump in.
If I have an interest, I'll jumpwith you.
(04:50):
But but it's when it's like toopersonalized, or like like no
one cares that much on a firstit's like that doesn't work
either.
SPEAKER_00 (04:57):
So that's true.
Um, we we've had to tone downsometimes the AI.
Some of the things we see a lotis the AI is very enthusiastic,
uh, and it will pull in all ofthe context that it knows about.
SPEAKER_01 (05:10):
Um, and one of the things I get a pub before my
speech recently you lost a childand have cancer, and and how we
tell you how that relates toyour post on nanoscience.
SPEAKER_00 (05:18):
Like, I know.
Well, and that that's the memeon LinkedIn.
Whereas like, oh yeah, my myparents both died in a car
crash.
This is what it taught me aboutsales.
SPEAKER_01 (05:29):
Like, that's that's not the kind of back up like
four years, we were we were onOpenAI's beta before it was chat
GPD, and we were turned thisthing on for automated
commenting.
And and I looked at one of mycomments one day and it said
that my comment literally was mydad was also an NFL star and
died in a plane crash.
I'm like, my dad's alive and wasan optometrist.
(05:51):
Um, it was like, holy cow, liketurn that thing off.
But the funny thing is, I stillsee some kind of version of that
in this current AI where it justtry to like enthusiastically
relate.
Uh so talk about your strategy alittle bit of how you're the
reason I bring that up, not forjust for humor context, but for
what you guys do, uh is youknow, you have you talking about
(06:12):
your journey in sales a littlebit, but tell me how you're
solving that a bit because thatthat's that's the nuance, I
think, where it happens, whereit works, or it doesn't.
SPEAKER_00 (06:20):
Yeah, and I'm happy to uh uh pull back the curtain a
little bit.
Uh everyone listening, feel freeto steal this idea.
Um, so what we realized is thatAI is very good at accomplishing
small goals.
You know, it's you've probablyall heard, you know, uh it's
like a you know, a new intern.
You can give it small tasks andit completes those rather well.
(06:42):
And so the the concept is tryingto break a sales conversation up
into smaller tasks where theobjection, objective is actually
evolving throughout theconversation.
So at the start of a salesconversation, a cold sales
conversation on LinkedIn, you'regonna have a very different
objective for the AI messagegeneration than at the end.
(07:05):
Because if you try and have thesame objective throughout the
conversation, it's just going tobe falling back on the pitch,
basically.
It's gonna say, I know this ishow I can get convinced people
to set a meeting.
I'm gonna send this messaging.
That's not what you want to do.
You wanna start by getting themin involved in the conversation,
(07:26):
asking curiosity-drivenquestions, building rapport, and
then pivot that to asking abouttheir challenges and try and
guide them to the subject thatyou want to talk about.
So we have built a structurethat actually feeds these
evolving objectives to the AI.
And we have another AI agentthat's kind of managing that
(07:49):
process where it's trying tounderstand in the conversation
where it's at and where thatother AI should go.
SPEAKER_01 (07:57):
It it's interesting.
Uh, it depends on your I thinkyour ICP a bit, your ideal
customer profile, because Idon't want to have a back and
forth conversation on it.
I want someone just to get toit.
Other people do.
And and so I think some so howdo you hand any of the
contextual data for that?
Because that's because Ihonestly, if you did that for
me, I usually reply back, hey,you need to work on your
sequencing.
I have no interest in the fivethings you just sent me.
SPEAKER_00 (08:18):
Yes.
SPEAKER_01 (08:19):
Just tell me what you do, and I'll tell you if I'm
interested in it now.
Like, like uh, talk to me aboutyour kind of AI piece with that,
because that that personamatters.
Uh, because the person who doeswhat you just described will
never get me out on call, butthe person who's very brief and
probably will.
SPEAKER_00 (08:34):
So no, and and you hit the nail on the head.
Uh, one of the ways we built itwas uh to be flexible for the
client's needs in the sequencinglength.
So we can actually set up thesequence to be two messages.
One nice, warm and fuzzyrapport-building message, and
then a straight talk.
Here's what I do well.
(08:55):
This is how I can help you.
What do you say?
We hop on a call.
About half of our clients usethat to great success.
But in sales, I'm you you're insales as well.
You know how important mirroringis.
There are also some of yourprospects that are gonna fall
into that um, you know, the theisolated, what do you call it,
(09:16):
um, the exclusionary categorythat actually do want to have
the conversation and they don'twant the direct pitch.
They don't want to hear aboutit, they want to chat about what
they're posting about orsomething before they get to,
okay, now I'm ready to hear whatyou're all about.
And so even for those clientsthat have the short cadence of
(09:36):
sequence, um, it still helps tohave the human oversight because
the human can step in and say,you know, it doesn't feel like
the the vibe in the room is aright to pitch, you know, our
normal short straight pitchmessage already to spot in the
conversation.
SPEAKER_01 (09:55):
What one of the uh greatest connection requests
automations ever ran was onlythis.
All I wrote was ooh, oh likefour O's and an H.
Dun dun dun.
Tell me more about company.
SPEAKER_00 (10:07):
Ooh.
Tell me about this.
Tell me more about ABC Company.
SPEAKER_01 (10:10):
Yeah, just tell me it's ooh, tell me more about
this.
Which gives them kind ofinterest.
Like, oh, are they interested inme?
What's going on?
And so uh I've ran that one toeffect.
I've also, I'll tell you what,the other one I would tell
people to take this is if youdon't know, it's better to say
nice to connect.
Do you want the long or shortversion of me?
Introduction.
Yeah, and like that.
It invites them to say noversion or short version, cool.
(10:35):
Quick reply, right?
Long version, great.
Um uh or do you want to pitchslap me or should I pitch slap
you?
Something like that.
SPEAKER_00 (10:45):
Like like I think the place the place that would
miss is that the the people thatdo appreciate rapport building
pivoting to a salesconversation, but they wouldn't
accept it if you offered it tothem.
Because if you said, Hey, do youwant to have a quick, you know,
three to five messages back andforth building rapport, and then
(11:05):
I'll tell you about my company,they're gonna say, you know, I I
don't want to go through thatprocess.
SPEAKER_01 (11:10):
Now, if you just there's a there's a group that
would be like, Hey, do you wantthree to five messages of us
rapport building, or would yourather just network for 10
minutes on a call?
Yeah, that like something likeyeah, like look at we can decide
to disconnect if it if itdoesn't make sense, though.
That is effective.
Now, also some that eliminatesthe serious side, to be fair.
SPEAKER_00 (11:29):
The point is you can AI they can play your each way.
That will get you meetings, butnot necessarily qualified
meetings.
And that's another thing that wetry and do is we try and not
waste our clients' time becauseour clients' time is very
valuable.
Um, and so we don't want to justset them networking meetings
that might go somewhere.
We need to uncover in thatsituation, in that conversation,
(11:51):
the fact that they needsomething from us and that we
can provide it, we can solvethat problem for them.
SPEAKER_01 (11:57):
Now, this goes and that's a really good valid
point.
So in our ICP, our stuff istypically we use the podcast as
uh as a top of the funnel formany.
So I'm knowing their positionand the fact that they're they
have some level of success, wecan help them.
It's just they're gonna theyneed it or not.
And that but we with the podcastin place, it it still helps
because it is it is a there's apiece that they're gonna always
(12:18):
have that conversation.
And it's usually never wasted ifmy yeah, they'd still be a good
guest.
They have this.
So it there's a lot ofvariables, I guess is my point
that goes into this.
Um, let's just take a pivot herefor a minute for you.
Just in your current business,what's kind of the biggest
metaphoric tie that you'restruggling with right now?
I think to kind of cut, getover, get, you know, to go next
level.
SPEAKER_00 (12:38):
The biggest tie is honestly being a service
company.
Um, I do not want to be aservice company anymore.
I want to pivot into being asoftware company.
And that will cut all sorts ofties because, you know, the
valuation of software companies,the scalability, the, the ease
of running, the requirement ofstaff.
(12:59):
Um, and we are very close tothat level.
You know, you and I have hadconversations, Thomas, before
about white labeling my softwarefor your agency, other agencies,
and they using their labor asthe copywriters, because that's
kind of integral to the ZenithProject offering.
You have that human in the loop,but other agencies can use
(13:22):
Zenith Project software to getthis massive level up uh with
the AI suggesting 80% of thehental uh heavy mental uh
gymnastics that you have to doto come up with very clever and
strategic sales conversationmessaging.
Yeah.
SPEAKER_01 (13:41):
Yeah.
And I think so.
I think the the we had talkedabout this a little offline.
So you could be at just anotherlead gen company.
Uh what you guys are developingspecifically to me, and and I
just kind of we're openly justkind of talking about the
solution of this, what you'redoing.
I think what you could offer toother AI lead gen companies as a
plug-in for theirs would be thebetter play.
(14:02):
Because you're like, hey, weactually give you the contextual
stuff beyond what's going to bein anything you're doing.
So your stuff just works better,like a middle layer, if you
will.
Uh, and then that way you're notever in trouble with any of the
social media platforms foryou're just taking data and
you're doing something with itto give advice to somebody else
trying to do something.
That middle layer becomesdefinitely makes you more, you
(14:22):
know, uh viable.
And and people can like, youknow, obviously, hey, listen,
you know, it's eight dollars peruser that you you have on yours,
and now you you have these, youknow, de-risk revenue micro all
over, and they're just pluggingin your stuff to make their
stuff more effective.
I think that there's a there's amarket for that because some of
the stuff we described is sonuance-based.
Like, who's Thomas?
Who's Ryan?
(14:42):
Who wants this?
What mood's he in?
What's his post been?
That they've been more aboutthis or that.
The the point is if you can getthat nuance really narrowed down
to what that person might like,that'd be humorous, are they not
like all these things?
What's their current intent?
That that would be applicableagainst anybody's own version of
their services of it of it.
Anyway, so I I think as youlook, you know, it's the
pickaxe, right?
For the gold rush, if you will.
(15:03):
Uh there's uh rush, you know, Iyou and I talked about this too.
Like, we're I'm pivoting morelike I just want to verify
companies can do what they saythey can do because there's so
many who can't.
Yeah, you get burnt.
And so like that, so you look atthese pivots, but how do you
feel about that?
Like kind of being a you canhave your own version of the
thing, but like you can havetechnology that could be applied
(15:25):
to all the Apollo data that'sout there, all the seamless data
that's out there.
Yeah, I mean, like so so I gotbuild is something more valuable
than just lead gen.
SPEAKER_00 (15:34):
I I'm a fan of vertical vertical integration,
you know.
I if if I can capture every partof the the the customer
lifecycle, then we're gonna trybecause you know having all of
that in one ecosystem is gonnabe way more convenient for the
client.
Um plugging our software intosomething like Apollo is
(15:56):
actually gonna be so easy, justbecause that's a an API feed
between the databases.
Um, and that is the vision forZenith project, is it's going to
basically be a CRM that you haveall of your enriched data flow
into, and then we do yourLinkedIn, we do your email, we
do your Facebook, we do yourtexting, and all of it has that
(16:18):
human layer in it because AIisn't ready to automate yet.
Like when I get an AI messageand it's like two paragraphs of
AI text, I don't read it and I'moffended by that person because
they didn't read it either.
I know they didn't read it, andthey're expecting me to take
five minutes to to to writetheir AI message.
(16:40):
It's over.
SPEAKER_01 (16:41):
I tell it drive me nuts too.
Like this literally justhappened before my meeting here.
Like, I'm trying to meetsomebody to help us with a
client for employer branding,just what it is.
And the person was like, Hey,I'll give you a call, or and I'm
like, I never do phone calls.
I didn't even anyway.
And then I'm like, they're like,Why can't you call talk for two
minutes?
And I'm like, I'm about to go ona podcast.
And they're like, that'ssketchy.
And I'm like, you want me tocall you for a long-term
(17:03):
partnership because you're in acar.
I'm like, set up time, get withme face to face.
It's like some people have thislike sketchiness, this kind of
like the they they they've likethey read into what you do too
much based on based on yourinitial interactions, right or
wrong.
This this person who did this,but I'm kind of like, you give a
perspective.
(17:23):
If you send an AI messageinitially, and it's clearly like
so in depth and perfectlywritten, it it's a dehumanized
without calling to the fact thatit's been dehumanized.
You're kind of like, oh,whatever.
Like this is just yes, I'm notreally important to you, is what
you're communicating to me.
Yes.
And and honestly, like Rubylowercase typos, I throw them
(17:44):
all in there.
Like, listen, I'm actuallywriting this shit.
I'm not, but my AI writes me intypo format.
SPEAKER_00 (17:49):
When I see one of my copywriters make uh a spelling
mistake, I'm like, good, thatdoesn't look like AI, you know.
SPEAKER_01 (17:57):
I use AI to say, listen, all lowercase uh use dot
dot dot instead of dashes, uh,throw occasional text version
typo in there.
You are versus, you know, usethe like write it like you did
in text, like I wrote for mymobile phone.
SPEAKER_00 (18:11):
That's good.
I I also one of the biggestrules for my copywriters, and
this is for everyone that's youediting AI text, is deletion is
your friend with AI.
AI will include context that thereader already knows and that
isn't needed to be saidexplicitly.
Um, and it will also justelaborate on things that need no
(18:34):
elaboration.
And so if you take a an eraserto AI text, it's gonna seem way
more personal and human.
SPEAKER_01 (18:41):
The uh the pros of AI sometimes, like, dude, just
say what you have to say and bemove on, like right.
Yeah.
Um but I will tell you no matterhow you do it, AI is an
incredibly effective tool forhelping you with brevity.
If you say, just give me brevityon this piece, it's still
probably longer and moreeffective and less ums and ems
and thems and the words thatdoes it.
Um talked about though.
(19:03):
This this is just from the fromthe purposes of staying on brand
or a podcast of cutting ties.
You're you don't want to be aservices company, shocker.
Let's talk about that a littlebit.
There's a lot of people outthere that are you know, it's
it's how you get startedtypically as an entrepreneur.
You start doing things yourselfthat you were gonna get paid for
someone else paid you to do.
Totally normal.
(19:24):
Um tell me how you're makingthat move though.
Like it's more than thing justto go to technology.
Are you do you have mentors?
Like, what actually happens foryou to go from services to more
of a SaaS?
SPEAKER_00 (19:36):
Great question.
Great question.
My main mentor, uh, who you arefamiliar with, Lee Coulter, um,
he has been pushing me to gosoftware, the software route
since the beginning, since Itold him about my idea to create
this AI driven outreach company.
Um, so I definitely have out uhmentors in the software to
(19:56):
service to software space.
Um, the other thing.
Is that proof of concept isreally necessary when selling a
software.
You have to be able to showpeople that they can recreate
the results that you createdwith the software.
And if you haven't proven itout, then you can't show them
anything.
And so that's one of the reasonswe've been a service company for
(20:19):
the past year and a half.
And now we have incredibly richresults and testimonials from
our clients who have used ourservice when we were using our
software and have seen greatresults from it.
And so I can show that toagencies and say you can
recreate these testimonials.
You can be a franchise,basically, of Zenit's project.
(20:41):
It's similar to how you knowMcDonald's is the corporation of
McDonald's is not a restaurantcompany.
They're a franchise company.
Oh, they're a real estatecompany.
Yeah, they are.
SPEAKER_01 (20:53):
It's a real estate company that has things on it.
SPEAKER_00 (20:56):
This is true.
This is true.
That's that's how Burger Kingactually figured out how to put
their restaurants, is they justput them near McDonald's because
McDonald's already did all theresearch on the real if you want
to go figure out where to go buyhouses or buy land, go find
where the next quick trip is.
SPEAKER_01 (21:12):
And for the next uh you see Walmart, 20 years from
now, that place will have 20x topeople around it.
So go get go get a lot of fiveacres somewhere because
someone's gonna want to go buyit for something because they
they exactly it's like exactlyand yeah, go ahead.
You want a rehab, go find aStarbucks that's in a location.
Like, why is that here?
Because there's adegentrification coming to this
(21:34):
area they already know about.
And there you go.
SPEAKER_00 (21:37):
They they've done all the research, they have the
teams, they have the resources.
Uh, believe them.
SPEAKER_01 (21:43):
That's one of my favorite phrases.
Sorry, Ryan, conscious timehere.
Just tell first of all, peoplewhere they should go to check
you out, check out the product,uh, and and and you know, get a
hold of you.
SPEAKER_00 (21:51):
Absolutely.
So zenithproject.co uh is ourwebsite.
You can feel free to go readabout us there.
Um, you can connect with me onLinkedIn.
My uh handle is Ryan Miller HQ.
Um and my email is Ryan atzenithproject.co uh and you can
reach me there.
(22:11):
Uh those are great ways.
All right.
SPEAKER_01 (22:14):
Let's leave the listener with like uh just three
pieces of advice.
Something you should always dowhen leveraging AI and reach
out, one and something youshould never do, and then
surprise me with the third.
SPEAKER_00 (22:26):
Okay.
So something you should alwaysdo with outreach of AI is read
the message before you send it.
And read read it, read it outloud.
Repeat that.
Yes, read the message before yousend it.
Because AI does hallucinate, andyou might get into some pretty
embarrassing situations withwhat AI says on your behalf.
(22:49):
Um, or have someone read it foryou, you know, like CF Ride.
Yes, absolutely.
Yes.
Um, so something to never dowith um AI outreach, I would
say, is uh use out-of-the-box umChat GPT messaging.
Like put some thought into whatprompting, what what iterations
(23:11):
of the AI messaging you want toget to.
Because if you use the generic,you know, if you use M dashes uh
in your messaging, you're thatis going to come across really
bad to potential prospects.
They're used about GDPT, right?
SPEAKER_01 (23:24):
No matter how many times it's helped to stop doing
that, it still puts them inthere.
I'm like, can you never use adash again and leave all emojis
dead forever?
And it's like, yeah, here'semoji with a thumbs up.
SPEAKER_00 (23:35):
I'm like, god damn it.
The the M dash is, I don't knowwhy AI likes it so much.
SPEAKER_01 (23:39):
I think it's actually uh proper, proper uh
grammar is why I've oh it wasfun new phenomenal.
SPEAKER_00 (23:45):
Yeah.
SPEAKER_01 (23:46):
Someone who like uh what was a my co-founder, his
wife is like a linguist orwhatever, and she's like, that's
actually the proper way to doit.
No one ever does it.
So now her her stuff looksGPT-ish.
I'm like, well, you gotta quitdoing it.
SPEAKER_00 (23:58):
I know.
I you I used N dashes, sothere's three types of dashes
there's n dashes and m-dashes.
M-dashes are basically used inplaces of commas or to um you
know split a sentence fragmentout.
Um, and they they're phenomenal.
They're so nice to read.
Uh they're they're wonderful towrite if you know how to.
(24:19):
They're kind of hard to, but um,but yeah, I I kind of miss using
m-dashes, but I I don't I don'tuse them anymore.
And I have all of my AI'sprograms to not use them.
SPEAKER_01 (24:30):
All right.
So okay, so one is to read it.
What was the thing you should uhis that always do or never do?
SPEAKER_00 (24:35):
Not do?
Yeah.
Yeah, the the thing you shouldalways do is read it.
The thing you should not do isuse m-dashes uh or use
out-of-the-box AI.
You know, uh definitely developyour own AI.
The third thing to uh surpriseyou.
Um well, let's see.
Does it have to be about AI orcan it be about MI?
(24:57):
I mean, it'd be cool if it was,but you know.
Okay, let's let's think of an AIone really quick.
SPEAKER_01 (25:02):
Give me like some maybe maybe a surprising thing
you didn't know you could dowith AI that you're some people
most some of your customers arelike, oh, I didn't know you
could do that.
Like, what would be an answer tothat?
SPEAKER_00 (25:12):
My my favorite use of AI is um reformatting massive
sets of data.
I this is kind of specific to meanyway, because I'm a data nerd,
but you can put in massiveamounts of data and ask it to
reformat in a specific way, andit'll do it perfectly.
Um, so you know, datareformatting or slight changes
(25:33):
to massive amounts of text issomething that AI does for
breakfast.
SPEAKER_01 (25:37):
Yeah, I I'll extend that.
You know what I use AI for is Ihave it I export our
transactions and our accountingand books and say, look at this,
and what do you see?
Like what what what do you seehere?
And what what's it cost to runthis company and when did when
should we hold reserves?
It's amazing.
So um uh listen, Ryan, thanksfor so much coming on today.
(25:59):
I appreciate it.
Absolutely uh get a hold ofRyan, Ryan, give him your uh
your place to go one more timeif you don't mind.
Zenithproject.co thanks Ryan forcoming on today.
I appreciate you.
You too.
Thanks, Thomas.
Listen, everyone who's stilllistening, get out there, go cut
a tie to whatever's holding youback.
Make sure to follow the podcast,subscribe, and if you really
liked it, do five stars, nothingless.
(26:19):
Five and only five.
That's all we accept.
Okay, thanks for listening.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Hey Jonas!
Hey Jonas! The official Jonas Brothers podcast. Hosted by Kevin, Joe, and Nick Jonas. It’s the Jonas Brothers you know... musicians, actors, and well, yes, brothers. Now, they’re sharing another side of themselves in the playful, intimate, and irreverent way only they can. Spend time with the Jonas Brothers here and stay a little bit longer for deep conversations like never before.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by Audiochuck Media Company.