In this live session, automation coach Anna Bui is walking us through the exact step-by-step process she used to scrape leads from LinkedIn using Appify, enrich them with data, and send them through a highly targeted outreach workflow, all while keeping it lean and cost-effective for early-stage experiments. This isn’t theory. It’s real-world, been-through-the-struggles, here’s-what-actually-worked automation.
Anna doesn’t just build tools, she builds systems that learn and adapt to your personality. You'll see how she overcame common roadblocks (and not-so-common AI frustrations) using N8N, Airtable, Clay, and Claude, plus how to personalize your entire funnel so it doesn’t sound like a robot wrote it.
Anna is a rising star in the AI and automation space - a coach at Clay Bootcamp, an ex-project manager at Speak On Podcast, and a hands-on innovator who’s turning Reddit posts, transcripts, and third-party data into fully automated content and outreach systems. She’s not just smart , she’s scrappy, fast, and endlessly creative. And in this session, she’s showing you everything.
Get Anna's workflows here: https://n8n.io/workflows/7034-convert-linkedin-post-reactions-into-qualified-leads-with-ai-and-apify/
Learn more about what Anna does: https://www.claybootcamp.com/
About Leveraging AI
- The Ultimate AI Course for Business People: https://multiplai.ai/ai-course/
- YouTube Full Episodes: https://www.youtube.com/@Multiplai_AI/
- Connect with Isar Meitis: https://www.linkedin.com/in/isarmeitis/
- Join our Live Sessions, AI Hangouts and newsletter: https://services.multiplai.ai/events
If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
GMT20250821-154502_Record (00:00):
Hello and welcome to another live
(00:03):
episode of the Leveraging AIPodcast, the podcast that shares
practical, ethical ways toleverage AI to improve
efficiency, grow your business,and advance your career.
This is Isar Metis, your host,and I am really excited today.
I'm really excited for twodifferent reasons.
Reason number one, we haven'tdone one of these lives in a
very long time, well, not verylong, and I guess for most
people, maybe not that much, butabout a month and a half since
(00:24):
we've done the last live, and weused to do this every single
week, but I was on vacation inIsrael, and that's a very
different time zone, and itwould've made the whole thing,
uh, uh, a lot more complicated,but we're back.
We're gonna be back doing thisevery single week unless I'm
traveling.
So if you are with us live,first of all, thank you for
being here.
Uh, if you're not with us liveand you're listening to this as
the podcast or watching this onYouTube afterwards, uh, we're
going to be back to doing thisevery single Thursday at noon
(00:47):
Easter time with an amazingexpert, uh, that's gonna teach
you how to do something verypractical and very effective
with AI and other tools.
And so, if that's what you wannalearn, come join us because A,
you get to hang out with thecool people, and B, you'll be
able to ask questions, which ifyou're just listening to this
afterwards, you cannot.
which leads me to.
The second reason I'm excited isour topic today.
(01:07):
You heard me say on this podcastmany times before that you can
have a business withoutmarketing.
You can have a business withouthr.
You can have a business.
Without finance, you can have abusiness.
Without operations, you cannothave a business without clients,
because without clients youdon't have a business.
That's the whole point.
And so to get clients you needleads and to get leads is not
(01:28):
necessarily an easy thing.
And today, in today's episode,we're going to dive deep into
how to use AI tools togetherwith NA 10, which is my favorite
automation tool to.
Get leads, meaning look throughrelevant people on LinkedIn and
reach the information about themso you know who they are, and
create relevant personalizedoutreach messages that drive
(01:49):
engagement, that lead to a muchhigher conversion rate, which
leads to clients, which is whatyou need in order to run your
business.
And so if you are in business,that should be of a high
interest to you because as Imentioned, this is the bloodline
of every single business.
And our guest today, Anna Bowie,has built a really cool, amazing
automation that does each andevery one of the steps that I
(02:11):
just described previously.
Basically getting you from, Idon't have anybody to, I have a
conversation with a long list ofpeople Now, in addition to that,
Anna is a.
clay.com expert and she has acourse where she teaches people
how to use clay.
Those of who don't know clay,it's an incredible tool that
does a similar process.
It does, uh, shows you how to,or it allows you to grab people
(02:34):
from different sources and thenyou reach the information, then
reach out to them.
So if you're asking yourself,why does Anna need a separate
process other than clay?
Well, first of all, I would lether explain probably more in
detail, but the main reason isclay is expensive.
And unless you get a very goodreturn on ROI on your clients,
then there might be a differentway and it also is a good entry
point.
So said.
So the process she developedallows you to do this without
(02:56):
risking more or less anything.
It's practically free to use, touse, uh, NA 10 and with a few
tokens, uh, running through theAPIs, you can do the process.
Figure out how effective it'sworking for you, finesse it.
And then go to Anna and take thecourse on how to implement it on
Clay.
So with all of that in mind, Iam really, really excited to
welcome Anna to leveraging ai.
(03:18):
Anna, welcome to the show.
Anna. Bui (04:03):
Thank you so much, Issa and, uh, so glad that
everybody joining our LA Linkline right now when I'm gonna
show you the workflow and gonna,you know, dive deep into, you
know, not just the technical,but also all the problem and all
the headache I got from buildingthis, uh, flow as well.
And that's just a little bitabout myself.
I was working as a projectmanager where I try to d into
automation just to make the teamlife easier at my previous
(04:26):
company.
And then I got that, I got intolike, you know, Zapier and make
very popular.
And then, I make a shift to Nend and I become literally an
end it end enthusiastic.
Uh, I finally become recently anN end creator, uh, officially.
So.
You probably will see my pro,my, uh, verified profile later
on.
(04:46):
But, um, basically I got, um, Ibecome a coach at Clay Bootcamp
where I teach people and showpeople skill on how to use the,
like clays as the two.
And now we are doing automation,especially with NNN because it's
also my expertise as well.
So I'm really looking forwardto, uh, continue our session now
and then later on when people,if anyone having any questions
regarding clay or automation, Ilove to, uh, answer them as
(05:07):
well.
Isar Meitis (05:08):
Awesome.
And relating to the people whoare joining us live on LinkedIn
and or on Zoom.
feel free to ask questions.
So I'm monitoring the chat onboth conversations.
If you have any questions, uh,please just pop them in the
chat.
I will get them to Anna'sattention and we're gonna answer
all your questions.
Thank you obviously so much forbeing here.
I'm sure you have other stuff todo on, uh, Thursday morning,
afternoon night, depending whereyou are in the world.
(05:29):
and with that in mind, it wouldbe great if you introduce
yourself in the chat.
Just say, where are you from?
What are you looking to get outof this session?
Uh, just so that people, uh,can, you can meet each other and
network as well and share linksto your LinkedIn if you're on
Zoom and so on.
Uh, but Anna, let's get started.
Let's dive right in.
Anna. Bui (05:45):
Yes.
Uh, I'm just gonna share my, uh,screen quickly.
GMT20250821-154502_Recor (05:48):
Please do.
Anna. Bui (05:50):
Okay, perfect.
So this is my workflow.
First of all, you can grab thisworkflow on this template right
here, right now it's the three,but I'm trying to build more and
more.
but here is the workflow.
This is a template where you canjust go in and grab this for
free.
is, is this exactly, is it it isthis workflow, but I just add
data to the, uh, dummy data tothat one.
But again, the, what thisworkflow does, okay.
(06:13):
I had to scale it back a bit.
When, um, I joined ClayBootcamp, Nathan Lippy, which is
our CEO, he encouraged me topost more on LinkedIn.
I never touched LinkedIn before.
I never, like on a, I have aprofile.
I never touch it, I never postanything.
And since I like, okay, I'llstart posting just the things
that I built, like the N eightend workflow or my opinions
screen guarding to, you know,the platform itself or like
(06:34):
different automation.
I just, you know, using theplatform to showcasing my
knowledge and what I builtbecause I think it's, I don't
have any thought.
I would just like, here's what Idid.
And, The impact that's from allthe posts and then the
encouragement from all the, fromeverybody within my LinkedIn
post is definitely overwhelming.
Hence, I was building thisworkflow just because, okay, I
have not like a viral post, Ihave a quite popular post and I
(06:58):
now consider LinkedIn is a warmlead, right?
I believe it's you in sale.
You probably agreed with me thatthat is, that is your nurture
sequence.
That is where you nurturepeople, that's where you
interact with people.
That's where prospect find you.
That is basically where peopleheard about you and they come to
you.
Uh, lots of inbounds coming in.
So I got this, somewhat popularpost.
I can, not gonna say it'spopular, but basically is have,
(07:21):
This is this one.
So I have, uh, 79 people,interact with, like, interact
with me.
These are reaction.
and I was like, okay, I have allthese people in interested in,
the, um, in what I, in my post.
So I was just wondering, let'sjust consider them as leads.
Let's consider'em at leads.
And then from, because they areindeed people that, you know,
(07:41):
interact with my post.
So maybe some of them wouldprobably, enjoy my content or
probably become my ICP.
Again, I'm not, I'm, I'm not abusiness owner, so this is all
like, um, you know, just, I'mjust cons, uh, I'm just take
myself into, um.
A position of a, of an an, of anagency owner or a small business
owner and my ICP people that arelike small business or small
(08:02):
agency or solo entrepreneursthat need help with automation.
So that would be my ICP.
Hence from that, I just want tosay, okay, all the people that,
interact with my post, theymaybe like 50 something percent
of them are probably my ICP.
So how can I actually capturethem and then, nurture them from
so far and so forth and probablybecome a prospect and then,
closing the deal and also sortof things.
(08:23):
That was my initial thought withthis workflow.
any questions?
Anything Yar?
GMT20250821-154502_Recordin (08:28):
No.
So two 10, 2 cents.
First of all, on a very highlevel, what is NA 10 or make the
tools that you mentioned?
So they are tools, automationtools have been around for a
very long time.
Right?
So I've used Zapier the firsttime around 2015, so it's about
a decade.
And, and what automation toolsknow how to do this?
They know how to move data fromone.
Software to the other.
So you have a contact onLinkedIn.
You can grab the information,put it in your CRM, you can take
(08:50):
the data from your CRM, put iton Excel file, and then in, in
your outbox, right?
It's like these kind of things.
What these tools did not haveany ability to do is to reason
think, research all the thingsthat AI is good at.
Another combination of thesetraditional automation tools
together with AI capabilitiesmake them superpowers if you
know how to use them.
2 cents about make versus Zapierversus n Aden make is probably
(09:12):
the easiest to use.
Zapier is a step above that withmore capabilities, but mostly, a
lot more connectors.
But to be fair, at this point,it doesn't really matter because
you can almost connect anythingyou want, uh, one way or
another.
Uh, and NA 10 is more of thegeeky brother of these two
tools.
on one hand the learning curveis steeper, like it's more
(09:33):
complicated to use and that'swhy it shouldn't probably be
your first entry point.
But because the way it's built,it's significantly more
flexible.
You can do a lot more thingswith it.
You're practically almostunlimited with what you can do
with it because it can run codeand it can get any kind of web
hook and.
The other reason why it'sbecoming very, very popular is
that it's open source andthere's a huge community who's
(09:55):
developing additionalcapabilities for it.
And you can self-host it,meaning you can actually grab
the code, host it on your ownhosting server, and then your
data doesn't go to make order.
Zapier just stays in the boxthat you're hosting.
and it makes it all cheaperbecause now you're paying$6 a
month for hosting, doesn'tmatter how many automations
you're actually running.
So there, there are manydifferent reasons why to use,
uh, NA 10.
(10:16):
Uh, that being said, I wouldn'tstart with NA 10 just because
it's more complicated, unlessAnna, you disagree.
Anna. Bui (10:23):
I, I agree with you.
I started with Zapier.
I started with make, and thankyou so much for scaling things
out.
I was just like, jump rightstraight in because no, no, no.
So
GMT20250821-154502_Recordi (10:30):
now, now we can jump straight in.
so, and you can actually gothrough the steps on how this
thing works.
Anna. Bui (10:35):
But, uh, I completely agree with you.
The reason why I avoid Zapier,because it is tend to get more
expensive.
Yes.
It's had a lot more connector.
It is gonna make your first stepinto automation, a lot easier
because it's verystraightforward.
It's just step by step like awaterfall.
It can do a lot, but it is gonnaget very expensive the more,
workflow you're trying to run.
Yeah.
So that's why I, same as you,um, with everything that you
(10:57):
just list.
That's why I choose N Ends as myfavorite.
you know, automation platformform.
Like you can self host it or youcan have this, clouding, as
well.
And one of the thing I probablyin the future you'll probably
see more and more is the MCPserver.
You can host the MCP within any,within the cloud or like, or
like, uh, having on your local,on your local hosting itself.
So it's definitely a verypowerful platform, but.
(11:21):
One of the things that peoplekeep reaching out.
Sorry, before we jumping into,the flow, uh, people reach out
to me and ask how can they startwith NNNI think the best way to
start is just grab one of thetemplate that you see on the
community.
They have, they have more than4,800 template.
So just you can search it outand then even with your tech
(11:41):
stack as well, just grab atemplate and play with it.
And the more you play with it,the more you feel.
That is very straightforward.
The community of N End is verypowerful as well.
Everybody contributing.
You will see a lot of peopleposting about their workflow,
about their problem and everyonethat I ever encountered that,
enjoy and building things on anyend, they're very kind.
So please feel free to justreach out to them asking a bunch
(12:03):
of questions.
But, I highly recommend you justjump straight in and just start
playing with the template.
It doesn't have to be perfectbecause they are template.
You supposed to modify them tosuit your need.
So the more you build a workflowthat solve your own problem, the
more you learn it very quickly.
GMT20250821-154502_Recording_ (12:19):
A hundred percent.
I'll add one more thing and thenwe'll dive into the actual flow.
Uh, when you do grab thesetemplates, you don't have to use
the, the entire template.
So you can consider the buildingblocks of the template, say, oh,
this segment, these four steps,they do this thing as part of
the 20 steps that the automationin the template do.
And I can use these four stepssomewhere else because I need to
(12:39):
do this one step for a differentthing.
So this way you can mix andmatch between different
components and differenttemplates, knowing very, very
little and still buildingsomething that works because
you're taking these legos thatsomebody has already built into
a part of a solution andcombining them together into a
complete solution.
But now let's dive into ouractual workflow.
We, we definitely given enough,uh, intro to this.
Anna. Bui (13:00):
Yeah.
So the workflow is gonna splitinto three session.
We have the trigger right here,and then we have the whole
process, as you can see.
Starting from here and then thisis the end, like the end result
is gonna have the record createdon your Airtable or whatever
database, uh, storing you use.
For me, I use Airtable'causeit's very straightforward and
easy for beginner.
So basically, these are the twoact appify actor.
(13:22):
So, um, so this AFI actor isserved, to scrape the link post
for you.
Like the script, the linkreaction for you is basically a
like a little special key thatgoing like unlock the dog link
in and like, okay, this is postreaction, let's just unlock it.
Get all, get everything out likea little teeth because, well we
all know how LinkedIn, be veryparticular with people with
(13:43):
automation and scraping.
So they are like a little teethunfortunately.
And um, there's two actor areusing.
So the first one is, okay, Ihave my post with 79 reaction.
How would I able to get theprofile of the 79?
like those 79 people?
And here is the one that I use.
Oh, sorry, this is the firstone.
so this is, yeah, reaction.
(14:05):
Oh, sorry.
It should be this one.
My apology.
So this one is the link portreaction scraper.
So this is your ap, your Appifyconsole.
And I'm just gonna show you twothings that you need to, you
just need to look into the firstone.
Just a second.
GMT20250821-154502_Recordi (14:18):
Just one second.
To those of you who don't know,appify is a platform and you can
get multiple different toolsthat are all API based that you
can use in automations andapplications that you're
building and so on.
And they are geared to do thingsthat are very, very specific.
So you go into Appify and thenyou can find, again, many of
these different things.
This particular one knows how toscrape LinkedIn for post
(14:40):
reactions.
Anna. Bui (14:42):
Thank you so much.
Um, yeah, appify is, is have alot of different, uh, API like
a, uh, endpoint Lexus.
So you can even find like, okay,Google map and vendors or like,
Instagram followers, like allthose sort of things.
So please.
And they have a lot of, youknow, actors like, serve
specific need.
But for this one we have the, wefocusing on the link and we have
(15:02):
the first of all, the link postreaction scraper.
So we scrape, uh, scrape thereaction of that particular
link, post the information thatyou probably need to see, is
that the, first of all, theinput parameter?
So basically we need to sendthis parameter to, um, to this,
the endpoint.
So basically you go to input andyou will see here it had exist.
(15:25):
And this information for postURL, you can find it in here.
So when you go to this link,this particular, uh, number is
here is the, unique, uh, numberfor this, uh, link in post.
So you add it there.
So this amplify will startextracting the reaction from
that, from this one.
GMT20250821-154502_Recording (15:45):
So again, for those of you who
don't see, it's the end of theURL of your post, right?
So those of you are not watchingthe screen because you're
listening to the podcast asyou're driving.
it's the end of the URL.
There's like a long numbercharacter kind of like thing.
You grab that, that's the whatappify needs to know what post
to script.
Anna. Bui (16:04):
Yeah.
Perfect.
And so basically this is, I'mjust call, so this e so okay.
In edit and you have lot ofnative note and all the native
note I using in here is, youknow, you request for, um.
An H TT P request.
You have a wait note if, so,this is our native note within
NN and in the first one we aregonna use is the, uh, HTPP
(16:26):
request.
So this is the endpoint on theURL to call to cause this, to
cause this appify, actor.
So you, you can see right hereis actually the, um, it's
actually the uni, the link, um,yeah, yeah.
Link to this, um, to thisendpoint.
And then for, and then afterthat you have your, you have
(16:47):
your token.
So, um, when you open an a, anappify account, you will have
your unique token and you justadd it at the end.
So that mean it's gonna use acredit, it's gonna use a credit
up your account.
And then as you can see here, Iput money here so you only use a
credit of your account to startusing this actor.
Does that make sense?
GMT20250821-154502_Record (17:08):
yeah, yeah, yeah, yeah.
Anna. Bui (17:10):
And then from here, we are gonna send, as you can
see here, we sent, uh, we sendthe, we add the port, URL here,
and then the page number is one.
So going back to this actor, theinformation that you would like
to add in, as you can see here,is page number.
So page one, return one to 100.
So a reaction.
So that means, if you have apost at under 100, you just need
to post at one, or by default,then it will be the page number
(17:33):
one.
But if you have a very viralpost, you have like, 500 people,
um, interact with it, then youhave to do like page three and
four and so on.
But to keep it simple is gonnabe one, and it's going to
extract all the 79 reaction fromthis link post through here.
As you can see.
Yeah, there's, there's an
GMT20250821-154502_ (17:50):
interesting question on LinkedIn.
do these tools comply with theLinkedIn user agreement?
Basically what LinkedIn allowsyou to do, from their
perspective?
Anna. Bui (18:02):
I don't, I'm not sure about that 100%.
I just know that from myknowledge is, whenever you go to
Appify and whatever with link,specific scraping actor sort of
thing, try to use no cookie.
So if you use no cookies, thatmean like whatever action, the
action I'm doing right now,which is scraping my, the
reaction from my LinkedIn postright, is from the person who
(18:23):
built the actor itself, not fromme, because I didn't associate
my LinkedIn account with this,with this scraping processed.
it's just an API call fromanother thing that the other
person built.
That my LinkedIn profile didn'tattach to it.
GMT20250821-154502_Recordi (18:36):
Yes.
So, so there's
Anna. Bui (18:37):
no cookie key associated, there's nothing to
kind of like track my behavior,if that makes sense.
so that is why, so there's,
GMT20250821-154502_Rec (18:43):
there's, what I will say is when it comes
to LinkedIn automation, thereare two aspects.
One is what you can get.
Off LinkedIn, and it is veryhard for LinkedIn to know who is
actually doing it.
Like Anna just said there,there's no way for them to know
who is actually scrapingLinkedIn.
Like they can try to block it,but they can't know it's you,
uh, because it's coming in froma specific application.
The other aspect of of LinkedInautomation is if you want it to
(19:06):
post on your behalf, on yourregular basis, and this is where
you might get in trouble, right?
If you're building automationsto engage on your behalf
automatically, this is where youmight get in trouble because
then it's you, it's you, it'sengaging on your behalf.
And then usually the bestpractices on these tools is just
to use them at whatever levelsthey tell you that they checked
and that are okay.
(19:26):
It's still against LinkedIn'srules and regulations, each and
every one of these tools willtell you, okay, if you do more,
no more than 30 a day, you'llprobably be fine.
And then you can decide whatyour level of risk to decide how
much you wanna push thatenvelope.
Anna. Bui (19:38):
Yeah.
And um, I recently, um, I'm on acall with, kga.
I probably, butcher his name,but basically he is, um, like an
expert in the Clay Bootcamp whenit's come to, uh, Lincoln
Outreach.
And he told me from very, thevery beginning, it's gonna, it
is there will be a risk, therealways gonna be a risk when we
come to, uh, LinkedIn outreach.
it's not really something that,um, Lincoln promote and not
(20:00):
something that li Lincoln wantpeople to do, like those sort of
automation or, um, you know,interact on, on your behalf,
other sort of things.
So there will be a risk if youdecided to do a sort of linking
outreach campaign, but that willbe on your own risk and you
probably have to accept that forthis one.
go back to what AU justmentioned, is an application
from another person, et cetera.
(20:21):
So I guess I was just.
Being like hidden.
I hope, I guess.
Yeah.
Okay.
So
GMT20250821-154502_Recordin (20:25):
the first step we use Appify to
scrape the people who actuallyengage with the prompt.
What happens then?
or that engage with the post,sorry.
Anna. Bui (20:32):
Yeah.
engage with the post.
And after that we have like 77items coming in.
It's like, oh my god, it's toomuch, too overwhelm.
So that is why we have this oneis a loop over item, which is
another native, uh, loop, sorry,another native note in n at end
it's basically split, split thething in batches.
So it's break down the 77 itemto just do this thing once, one
(20:54):
item at a time.
So that mean we gonna, it'sgonna, one item's gonna run
through this whole processed.
And when it's done, then thesecond coming in and the third
coming in and the fourth comingin.
So it is basically try to avoidover dry like overflow, over
flooding, sorry, over flood.
So, um, the system and to have,and also let the, um, each
person have time to, sorry, eachof the items have time to
(21:17):
actually process and go throughwhatever, you desire the
outcome.
Awesome.
And yeah, so here the first itemcoming in, I always recommend
people to try this set note.
So this is another, native nodein NN is just basically to clean
your data.
So whenever you how to say this,the more I work with this data
(21:37):
and the more I work withscraping.
I realize that you have to startcleaning your data from the very
beginning so that from to thevery end, it's not gonna guide
you.
get you a lot of headaches.
The same thing when you startbuilding your play table.
Always make sure your data isclean.
So I, I guess like for everybodywho enjoy data and then data
analysis or data strategies,they, you, they, you probably
agree with what I'm saying.
I try to clean things.
(21:57):
Well, let, let's
GMT20250821-154502_Rec (21:57):
explain, let's explain in simple English
what that means.
Cleaning the data in thisparticular use case, so we got
scraped information fromLinkedIn.
We're now sending the firstitem, right?
Because we're gonna do this oneby one.
So the information that comesis, profiling, first name, last
name, company, you know,everything.
It knows that it can pick fromthat.
What does the data cleaner do?
Anna. Bui (22:17):
Yeah, so this basically, is just to spit up a,
a.
How to say this?
Split out the, as you can seehere, the data is, there's a lot
of data here.
They have like the profilepicture, like everything like
that.
We don't need that.
You only need what important.
So this is just basically cleanit out and able to just give you
what you desire.
So I just want, so profile urn,I just, I just want to get the,
(22:41):
okay.
Uh, had to explain it a a littlebit.
We have the Lincoln URL, whichis your own link.
and your own profile.
Link Profile, yeah.
Profile link.
And then you have the URN, whichis the unique number A
associated with your accountwith your LinkedIn account.
So the URL can be changed, likeyou see like with my name Anna
Bui, something, something, maybeI can change this to a different
(23:02):
name.
So the URL can change, but theURN will never change.
So that is.
This is like a primary data ifyou want to do some sort of
lookup.
So it's because it's reliable,it's reliable data.
The your end.
Okay, going back, I just want ajob title.
I just want link in URL, I justwant this and the name.
That's all I need for the nextintegration of scraping to use,
(23:24):
which is just a very clean,straightforward data.
So it's not gonna mess up thesystem or like mess up the se
sequence of scraping because nowit's like, okay, I'm not gonna
give you all this crazy thing.
I'm just gonna give you thisfull, straightforward
information right here becausePerfect.
GMT20250821-154502_Recording (23:39):
So it, it's does two things.
A, it filters the lots of datayou got into just the little
data you need.
And the second thing it picks.
What particular type of datayou're gonna get.
So like you said, the URL maynot be the right thing.
The URN is the right thing.
So that's what I want totransfer forward, as far as the
ID of the person.
So this is basically as thefirst step of everything else
(23:59):
because this will set the stageto these are the attributes that
I'm actually going to use, uh,moving forward in my automation.
Awesome.
Perfect.
Great explanation.
next step?
Anna. Bui (24:09):
Yeah.
Um, just want to, I just want toadd one more thing.
The reason why I have to cleanit from the very beginning,
because it's all these othernode can actually use this for
reference.
Yeah.
So it's gonna be like a veryreliable, uh, data set.
And then after that I'm check ifthere's any duplications in my,
um, in my Airtable.
So as you can see right now,it's blank.
So I'm just gonna check okay.
If.
(24:29):
if it is actually blank.
so this is mean that if there'sno data, I would say this is
data.
So this means the data lang.
So that's, there's likeinformation here is true.
Then this mean it's gonna startcreating, it's gonna start with
all this increment.
And then if is already exist,then it's nothing to do.
But basically, uh, just tosummarize, this is a check note.
(24:52):
Just it's just basicallychecking if this person already
on my database is this person,haven't, then Okay, let's start
with the next one.
Yeah.
GMT20250821-154502_Record (25:00):
Makes sense.
Anna. Bui (25:02):
So now is check that okay, this person haven't exist.
That mean is gonna, is gonnacreate a new record on, on this
new record.
It's gonna have, it's gonna addinto all this feel on Airtable
here.
It's gonna start mapping it out.
Right now it's nothing.
But when it's run, you willactually see the information
coming in and.
Is, you know, this is Airtablenote that, um, with NNN it's
(25:22):
integrating in NNN.
So it's very straightforward.
You just kind of just map thingsout.
And as you can see here with theURN, I just put it like this and
it's referenced the note.
That's why you clean the data.
GMT20250821-154502_Recording (25:31):
So again, for those of you are not
watching, and don't knowAirtable, Airtable is like Excel
with a better user interface ifyou want.
It's just a database where youcan define rows and columns and
you can put data in them and toconnect the two together.
So now we have four pieces ofinformation on each individual
that we scraped.
You basically connect theAirtable node, and that will
bring into NA 10, the fourcolumn headers that already
(25:54):
exist in Airtable.
You can do the same exact thingwith Excel, and then all you
have to do is drag theparameters that you brought from
the scraper.
And drag them into each andevery one of the columns.
So you drag the name to thename, the id, to the id, the
link to the link, whatever itis.
Uh, and that's it.
And that's how you set up.
And now what's gonna happen isas this first data comes in, it
is going to check, does thatdata already exist?
(26:17):
It's probably going to checkbased on the URN that we now
know.
That's the unique identifier ofthe person.
If that does not exist, it willcreate a new line in the
database.
Uh, or in this case on the, uh,Airtable table.
By the way, I, I'll say one morething you say, okay, why, why do
I even need to check?
Why do you care if there'sduplicates?
there's two reasons.
A duplicates are never a goodthing when you are creating a
(26:37):
database.
But the other thing is, some ofthe few steps will actually send
you to different large languagemodels, which means you're gonna
be paying for tokens.
And if you can minimize theamount of tokens you're paying,
uh, then you're just gonna saveyourself money and time.
Anna. Bui (26:49):
Oh yeah, definitely.
I, I've been there, done that.
I scraping one person atmultiple times because I was
testing it out.
Highly recommend you havesomething to check, to look up
your, uh, your data so youdon't, um, for sure.
Waste second.
Okay, now for the fun part, thereason why, I have people asking
me, okay, now you have thisperson pro.
So from the previous, appifyactor is give us this person,
(27:11):
uh, profile ID and then the, thewebsite, the UL.
Right now we actually need toenrich that.
Now we actually have to scrapethis person, personal link, uh,
LinkedIn profile.
Hence we will move to, s sorry,this actor right here.
So it says, two actor use inthiss workflow.
The first one is to scrape thereaction of the post and the
second one to actually enrich,the individual profile that
(27:35):
interact with your, yourLinkedIn.
If that makes sense.
GMT20250821-154502_Record (27:39):
Yeah, it's good.
Yeah.
Yeah.
So, so the, the what we gotbefore, because what we scraped
is the post is only theinformation that appears on the
post.
Now we want to go to the actualperson's profile and then we can
pull whatever we want.
We can see their company name,their company size, their
industry they're in, what theywrote about themselves, uh,
their title, where they workedbefore.
(28:00):
All the stuff that appears onprofile you can bring in by now
scraping their profile, which isthe current step.
Anna. Bui (28:06):
Yeah.
And for this one, uh, for thisactor, all you need to do is the
pro the profile, uh, BRL.
And I have it here.
I just added from this cleandata note added in the, um.
You know, add in the actor andthe actor is running and if gave
out all the information.
So the previous one just gave uslike very basic information, um,
like the name, the uls or theprofile, all those sort of
(28:27):
thing.
But this one, as you mentioned,is giving like your full name
and the headline and theyexperience the company that they
work in.
Basically it scrape this wholeperson LinkedIn profile.
And, uh, from that, it is justgonna, so the, from all this
information, it is kind of likespit up as as one item.
So I want them to combinetogether, combine the full name,
(28:49):
stuff like that, all theinformation that I need.
These, aggregate node, it quitestraightforward.
Just basically just merge allthis information, like all this
long list into like one cleanerlist.
And then from that we are gonnahave our AI.
Classify if this person isactually part of our ICP.
So based on the information thatis that we scraped from this
person's, profile, we are gonnalet the AI know, like, okay,
(29:12):
based on the problem that I'mgiving you, as you can see here,
you are an AI classifier thatdetermines if this person is the
ideal, is the ICP or not.
So I add in my ICP definition,like the people that I, the
roles that need automation.
Basically I paste my ICP here,and I also gave, uh, gave it an,
uh, output format.
is this ICP, yes or no?
The reason why this was an, uh,this person is an ICP and it
(29:36):
just need to give, give outthose information based on the
incoming data, um, that, fromthe scraping itself, S, but this
is more like into the promptingengineer and everybody have
their different ICP.
So always make sure that youcustomize it to suit your need
to able to give up the correctinformation.
GMT20250821-154502_Recor (29:53):
Quick, uh, help for people to test this
out.
So if you run this here on theautomation, every time you run
this, you're gonna pay fortokens.
It's very, very cheap.
Like you can probably run this athousand times and it's gonna
cost you a dollar.
But, uh, if you wanna make itcompletely free, uh, use any of
your existing AI tools or theone that you're actually gonna
use for this.
So if using Chat g pt, use chat,g, pt, cloud, Gemini, whichever
(30:14):
you're gonna use for the API goto that tool and then give it
the information from thatperson.
Take a screenshot from LinkedIn,drop it in there and use your
prompt that you're gonna use inthis automation here and just
test it out and just keep onimproving the prompt until you
get the proper answer everysingle time.
And then start running itthrough the automation, because
then testing it is gonna costyou nothing.
Anna. Bui (30:35):
Yeah.
And one of the sessions that Ihave at Clay Bootcamp from
Spencer Teil, he is excellentand he was breaking down to us
how to actually.
And engineers is prong.
So at first I was like, oh, AIprompting easy.
Not a big deal.
The way that he break it downhelped me a lot with this
project, because you are notjust gonna give the AI
(30:55):
instruction, you're gonna giveit knowledge as well about your
per, your ICP.
And not to mention, you have togive up, a required format that
it need to follow, an examplethat it will need to see and
reference.
So the more so the informationthat you're giving into ai, the
more, structure and the moredetailed information go to the
ai, the better the result.
And as you can see here, um, Iask it to like, just give me two
(31:18):
output, which is, is this anICP, true or fails?
And the reason why you thinkthis person is not an ICP.
That's just it.
Very simple.
GMT20250821-154502_Rec (31:26):
Awesome.
Okay.
Anna. Bui (31:28):
And then lastly is that I'm gonna have this, okay,
now we have the information ofthis person.
This person is first of all, um,couldn't find this person.
Uh, email address, which couldhappen because sometimes the
scraper unable to scrap theemail address.
That's some, you know, it's ahit or miss.
Um, sometimes it able to findit, sometimes not.
But basically ICP is not an ICPthen the reason, so right now we
(31:51):
just need to update the recordthat we create from the very
beginning with the reasoning andthen the checker, if it's an ICP
person or not.
And that was the whole flow.
Now all is coming, going back tothe loop note and then the loop
on.
Do integration and integrationuntil it scrape out all the, it
did to all the 77 items.
GMT20250821-154502_Rec (32:12):
Awesome.
First of all, incredible,really, really powerful
capability that again, uh, we'llprovide the link to this, uh, in
the show notes.
So anybody who wants access tothis, Anna has, the way I found
her is because I saw her sharingthis, this thing on LinkedIn.
I'm like, oh my God, this isawesome.
I want everybody to know.
And so that's why she's on theshow.
So she was, Gracious enough toshare this with the world.
(32:33):
And we're gonna share this onthe show notes of this episode
as well.
Uh, so you can get access to itand copy it.
And like Anna said, not justcopy it, but also learn from it.
So understand what each and oneof the components.
So what you can do is you canactually open this on your
computer and watch the video asyou're looking at the actual
thing and dive into each andevery one of the components so
you can understand what it does.
And that was the main reason whyI wanted to dive into this and
(32:54):
show you step by step.
I want to touch on a couple ofmore things, uh, and then I'll
go into a few of the questionsthat, that are coming from, from
LinkedIn.
But, aspect number one that Iwanted to touch on is if you are
still working on your prompt forthe agent that classifies,
whether it's a.
(33:16):
Potential customer or not?
Right.
Does it, is it aligned with yourICP or not?
One of the things that you cando, and I have a very similar
process myself and I have amaybe option, so when it's not
sure it's spinning out a maybeand that enables the AI to
basically okay with not beingsure.
And I want it not to be ahundred percent sure when it's
not a hundred percent surebecause otherwise I may miss a
(33:37):
client, it decides that it's nota client.
And so what I do with my maybeson my table is I go and manually
inspect them.
Right?
And so that gives you, uh, theopportunity not to lose people
that actually might be goodpotential clients for you.
And so every time it's not ahundred percent sure I.
Allow it to say, well, I'm not ahundred percent sure and then I
(33:58):
can go and check myself.
So that's number one.
Uh, the second thing that I wantto ask you, Anna, uh, and I know
the answer, but I want you toexplain in theory, because we
don't have it in front of us,what's the next step?
So let's say I have all of them.
How can I create an engagingoutreach to them that will
actually capture their attentionand will allow me to start a
conversation with them?
Anna. Bui (34:18):
Oh, okay.
That, that is a great question.
So at first, lemme just, uh,show you how this workflow,
let's just test it to see howit's run.
So right now you see that'sblank, right?
Yeah.
And after I run this wholeworkflow, as you can see here,
GMT20250821-154502_Recording (34:34):
so for those of you not watching,
you can kind of see the thingrunning.
It shows you what step it'sdoing right now.
And now we have the first fieldor the first line populated.
On, Airtable and it's gonna keepon doing this.
I think it goes through theentire 76.
And for H one, we will initiallysee just the basic information
coming in from the first step,and then we're gonna see the
enrichment.
If it is the ICP, it's becausethat's how the automation runs.
Anna. Bui (34:58):
So I think the, just to answer your, uh, your
question is the way to build theoutreach is, how to say this,
this is where Clay come to play.
So you have this, you have this,this, all these people profile
UL, right?
Is that correct?
You have all this information soyou can, so with Clay, I mean
you can use other two as well,but I have to dig in a bit more.
(35:18):
But what with clay is that theyhave, you have clay gen and all
of this information that you areable to scrape from this person,
is gonna, click and able toenrich this person, uh, data
also based on your UOL as well.
And then from that you're gonnause the clay gen, which is like
a, an ai, agent in clay toactually write out the post for
you, uh, to write out theoutreach for you, because it's
gonna based on all the, um, allthe personalized, parameter of
(35:41):
this person.
So for example, this person is,a title is AI automation and
developer.
Maybe you will write somethingabout like, hey, swale, I saw
that you also an AI andautomation developer, and then
it starts writing it out.
And that is where Clay come toplay.
Because the reason, the reasonwhy I recommend Clay for that
is, is able to interact with somany different, variables, like
different specific variable,personalized variable of this
(36:04):
person.
And it can create, um, thisoutreach.
One thing that I recommend isthat you should have a draft
outreach that you would like theAI to follow.
Hey, make sure you use thecorrect name or like.
This should be an example emailthat you will use, like this one
email that you find is very goodenough for outreach.
You let the AI see that exampleand then from that it's going to
interact with differentvariables that is, was able to
(36:26):
scrape from play itself, thatable to enrich from this person
information, which is theprofile out to use that to write
out the, uh, outreach, withinClay itself.
And then from that, clay canactually send the outreach to,
it can send it to um, it cansend it back to NN or it can
send it to landless or smartlead or something like that.
So that is like the differentprocess, uh, process if you want
(36:46):
to combine.
again, just one more thing.
If you want to combine n it endclay and is that let n it end do
the things that, to save thecredit from clay, for example.
as you mentioned, it's gonnacost you like two,$3 to scrape
1000 people to get the profileUL from that.
You export that as CSV, uploadit to Clay as a new table and
then use enrichment, functionof, of clay to actually extract
(37:07):
the profile you are with evenmore personalized information
and then write the email fromthere and that when you see the
email outreach from that personwithin Clay, shoot it out to
somewhere else.
So that how you kind of have tobe strategic of how you would
like to have your outreach veryhighly personalized.
And with clay, the more narrowand filtered your list, the
better that mean, like the listthat you're going to put into
(37:29):
clay has to be 100% somethingthat you would like to target
the list that you absolutelywant to nail.
Then you're gonna use clay.
And Clay gonna exceed with that,with the outreach and writing
very personalized, email.
Okay.
Awesome.
I will
GMT20250821-154502_Recordin (37:41):
say one, one more thing.
I think you can still do it herebecause all you have to do is
build a second agent, a secondAI agent in your NA 10 process.
That does exactly the samething, right?
You can give it all theinformation from the previous
steps.
You can say, here's the name,here's the title, here's what
they wrote about themselves inthe profile.
Here's the size of the company.
Like all the stuff that youhave.
Here's the template of the emailthat, that I want you to use, or
(38:06):
the template of the message thatI want you to send, uh, to base
it on.
And then I want you topersonalize that based on all
the information that we'vecollected in the previous steps.
And you'll do the same thing.
And then you can put the outputin Airtable, so you can have a
column in Airtable.
That would be the recommended,thing to send.
I agree with you that it ispotentially easier to do it in
clay.
and like you said, clay can thenconnect to a gazillion other
(38:28):
places that can continue, uh,the process.
Uh, I'm not taking anything awayfrom Clay.
I think it's an awesome tool.
Uh, and like you said, it's notnecessarily a bad idea to
combine the two together to getthe best of both worlds.
A lot of people are saying thankyou and this is great, and this
has been awesome, and that theyreally like the flow.
If, and I, and I agree.
I think this is a, uh, first ofall, it's a, it's an awesome
flow, but it's also, uh, you,you did a very good job in
(38:50):
explaining how it works.
Uh, if people wanna learn moreabout you, learn from you, work
with you, take your courses, uh,hire you to build automations
for them, what are the best waysto do that?
Anna. Bui (39:01):
please feel free to reach out to me on LinkedIn and
from that you'll see all theresources.
and I will keep posting thingsconstantly.
and maybe again, uh, on the,maybe on the next episode,
hopefully I can tell you,actually build out the, um, as
the AI agent that can write theemail for you.
Uh, I was just thinking aboutscalability, you know?
Yeah.
But anything all good, but like,when you start scaling things,
you probably want something morerobust.
(39:21):
But yeah, thank you so much forhaving me, and I'm so glad that
everybody in, you know,hopefully you enjoy the session
and find it.
Um.
Helpful.
GMT20250821-154502_Record (39:31):
Thank you so much.
This was absolutely fantastic.
Thanks everybody for joining us.
Thanks for, uh, being veryactive.
There's a lot of chats happeningon LinkedIn and on Zoom and
people asking great questions.
I didn't bring all of them up.
I answered some of them myselfif I knew, uh, the answers.
Uh, but thanks everybody forjoining us again.
I know you have other stuff todo.
Come join us next week everyThursday, 1:00 PM Eastern.
We're gonna do something likethis with another amazing expert
(39:51):
like Anna.
and thank you obviously, Anna.
This was amazing.
First of all, thank you forworking on this and thank you
even more for coming and sharingit, uh, with us.
Anna. Bui (39:58):
Yeah, thank you so much.
Hello and welcome to another live
(00:03):
episode of the Leveraging AIPodcast, the podcast that shares
practical, ethical ways toleverage AI to improve
efficiency, grow your business,and advance your career.
This is Isar Metis, your host,and I am really excited today.
I'm really excited for twodifferent reasons.
Reason number one, we haven'tdone one of these lives in a
very long time, well, not verylong, and I guess for most
people, maybe not that much, butabout a month and a half since
(00:24):
we've done the last live, and weused to do this every single
week, but I was on vacation inIsrael, and that's a very
different time zone, and itwould've made the whole thing,
uh, uh, a lot more complicated,but we're back.
We're gonna be back doing thisevery single week unless I'm
traveling.
So if you are with us live,first of all, thank you for
being here.
Uh, if you're not with us liveand you're listening to this as
the podcast or watching this onYouTube afterwards, uh, we're
going to be back to doing thisevery single Thursday at noon
(00:47):
Easter time with an amazingexpert, uh, that's gonna teach
you how to do something verypractical and very effective
with AI and other tools.
And so, if that's what you wannalearn, come join us because A,
you get to hang out with thecool people, and B, you'll be
able to ask questions, which ifyou're just listening to this
afterwards, you cannot.
which leads me to.
The second reason I'm excited isour topic today.
(01:07):
You heard me say on this podcastmany times before that you can
have a business withoutmarketing.
You can have a business withouthr.
You can have a business.
Without finance, you can have abusiness.
Without operations, you cannothave a business without clients,
because without clients youdon't have a business.
That's the whole point.
And so to get clients you needleads and to get leads is not
(01:28):
necessarily an easy thing.
And today, in today's episode,we're going to dive deep into
how to use AI tools togetherwith NA 10, which is my favorite
automation tool to.
Get leads, meaning look throughrelevant people on LinkedIn and
reach the information about themso you know who they are, and
create relevant personalizedoutreach messages that drive
(01:49):
engagement, that lead to a muchhigher conversion rate, which
leads to clients, which is whatyou need in order to run your
business.
And so if you are in business,that should be of a high
interest to you because as Imentioned, this is the bloodline
of every single business.
And our guest today, Anna Bowie,has built a really cool, amazing
automation that does each andevery one of the steps that I
(02:11):
just described previously.
Basically getting you from, Idon't have anybody to, I have a
conversation with a long list ofpeople Now, in addition to that,
Anna is a.
clay.com expert and she has acourse where she teaches people
how to use clay.
Those of who don't know clay,it's an incredible tool that
does a similar process.
It does, uh, shows you how to,or it allows you to grab people
(02:34):
from different sources and thenyou reach the information, then
reach out to them.
So if you're asking yourself,why does Anna need a separate
process other than clay?
Well, first of all, I would lether explain probably more in
detail, but the main reason isclay is expensive.
And unless you get a very goodreturn on ROI on your clients,
then there might be a differentway and it also is a good entry
point.
So said.
So the process she developedallows you to do this without
(02:56):
risking more or less anything.
It's practically free to use, touse, uh, NA 10 and with a few
tokens, uh, running through theAPIs, you can do the process.
Figure out how effective it'sworking for you, finesse it.
And then go to Anna and take thecourse on how to implement it on
Clay.
So with all of that in mind, Iam really, really excited to
welcome Anna to leveraging ai.
(03:18):
Anna, welcome to the show.
Anna. Bui (04:03):
Thank you so much, Issa and, uh, so glad that
everybody joining our LA Linkline right now when I'm gonna
show you the workflow and gonna,you know, dive deep into, you
know, not just the technical,but also all the problem and all
the headache I got from buildingthis, uh, flow as well.
And that's just a little bitabout myself.
I was working as a projectmanager where I try to d into
automation just to make the teamlife easier at my previous
(04:26):
company.
And then I got that, I got intolike, you know, Zapier and make
very popular.
And then, I make a shift to Nend and I become literally an
end it end enthusiastic.
Uh, I finally become recently anN end creator, uh, officially.
So.
You probably will see my pro,my, uh, verified profile later
on.
(04:46):
But, um, basically I got, um, Ibecome a coach at Clay Bootcamp
where I teach people and showpeople skill on how to use the,
like clays as the two.
And now we are doing automation,especially with NNN because it's
also my expertise as well.
So I'm really looking forwardto, uh, continue our session now
and then later on when people,if anyone having any questions
regarding clay or automation, Ilove to, uh, answer them as
(05:07):
well.
Isar Meitis (05:08):
Awesome.
And relating to the people whoare joining us live on LinkedIn
and or on Zoom.
feel free to ask questions.
So I'm monitoring the chat onboth conversations.
If you have any questions, uh,please just pop them in the
chat.
I will get them to Anna'sattention and we're gonna answer
all your questions.
Thank you obviously so much forbeing here.
I'm sure you have other stuff todo on, uh, Thursday morning,
afternoon night, depending whereyou are in the world.
(05:29):
and with that in mind, it wouldbe great if you introduce
yourself in the chat.
Just say, where are you from?
What are you looking to get outof this session?
Uh, just so that people, uh,can, you can meet each other and
network as well and share linksto your LinkedIn if you're on
Zoom and so on.
Uh, but Anna, let's get started.
Let's dive right in.
Anna. Bui (05:45):
Yes.
Uh, I'm just gonna share my, uh,screen quickly.
GMT20250821-154502_Recor (05:48):
Please do.
Anna. Bui (05:50):
Okay, perfect.
So this is my workflow.
First of all, you can grab thisworkflow on this template right
here, right now it's the three,but I'm trying to build more and
more.
but here is the workflow.
This is a template where you canjust go in and grab this for
free.
is, is this exactly, is it it isthis workflow, but I just add
data to the, uh, dummy data tothat one.
But again, the, what thisworkflow does, okay.
(06:13):
I had to scale it back a bit.
When, um, I joined ClayBootcamp, Nathan Lippy, which is
our CEO, he encouraged me topost more on LinkedIn.
I never touched LinkedIn before.
I never, like on a, I have aprofile.
I never touch it, I never postanything.
And since I like, okay, I'llstart posting just the things
that I built, like the N eightend workflow or my opinions
screen guarding to, you know,the platform itself or like
(06:34):
different automation.
I just, you know, using theplatform to showcasing my
knowledge and what I builtbecause I think it's, I don't
have any thought.
I would just like, here's what Idid.
And, The impact that's from allthe posts and then the
encouragement from all the, fromeverybody within my LinkedIn
post is definitely overwhelming.
Hence, I was building thisworkflow just because, okay, I
have not like a viral post, Ihave a quite popular post and I
(06:58):
now consider LinkedIn is a warmlead, right?
I believe it's you in sale.
You probably agreed with me thatthat is, that is your nurture
sequence.
That is where you nurturepeople, that's where you
interact with people.
That's where prospect find you.
That is basically where peopleheard about you and they come to
you.
Uh, lots of inbounds coming in.
So I got this, somewhat popularpost.
I can, not gonna say it'spopular, but basically is have,
(07:21):
This is this one.
So I have, uh, 79 people,interact with, like, interact
with me.
These are reaction.
and I was like, okay, I have allthese people in interested in,
the, um, in what I, in my post.
So I was just wondering, let'sjust consider them as leads.
Let's consider'em at leads.
And then from, because they areindeed people that, you know,
(07:41):
interact with my post.
So maybe some of them wouldprobably, enjoy my content or
probably become my ICP.
Again, I'm not, I'm, I'm not abusiness owner, so this is all
like, um, you know, just, I'mjust cons, uh, I'm just take
myself into, um.
A position of a, of an an, of anagency owner or a small business
owner and my ICP people that arelike small business or small
(08:02):
agency or solo entrepreneursthat need help with automation.
So that would be my ICP.
Hence from that, I just want tosay, okay, all the people that,
interact with my post, theymaybe like 50 something percent
of them are probably my ICP.
So how can I actually capturethem and then, nurture them from
so far and so forth and probablybecome a prospect and then,
closing the deal and also sortof things.
(08:23):
That was my initial thought withthis workflow.
any questions?
Anything Yar?
GMT20250821-154502_Recordin (08:28):
No.
So two 10, 2 cents.
First of all, on a very highlevel, what is NA 10 or make the
tools that you mentioned?
So they are tools, automationtools have been around for a
very long time.
Right?
So I've used Zapier the firsttime around 2015, so it's about
a decade.
And, and what automation toolsknow how to do this?
They know how to move data fromone.
Software to the other.
So you have a contact onLinkedIn.
You can grab the information,put it in your CRM, you can take
(08:50):
the data from your CRM, put iton Excel file, and then in, in
your outbox, right?
It's like these kind of things.
What these tools did not haveany ability to do is to reason
think, research all the thingsthat AI is good at.
Another combination of thesetraditional automation tools
together with AI capabilitiesmake them superpowers if you
know how to use them.
2 cents about make versus Zapierversus n Aden make is probably
(09:12):
the easiest to use.
Zapier is a step above that withmore capabilities, but mostly, a
lot more connectors.
But to be fair, at this point,it doesn't really matter because
you can almost connect anythingyou want, uh, one way or
another.
Uh, and NA 10 is more of thegeeky brother of these two
tools.
on one hand the learning curveis steeper, like it's more
(09:33):
complicated to use and that'swhy it shouldn't probably be
your first entry point.
But because the way it's built,it's significantly more
flexible.
You can do a lot more thingswith it.
You're practically almostunlimited with what you can do
with it because it can run codeand it can get any kind of web
hook and.
The other reason why it'sbecoming very, very popular is
that it's open source andthere's a huge community who's
(09:55):
developing additionalcapabilities for it.
And you can self-host it,meaning you can actually grab
the code, host it on your ownhosting server, and then your
data doesn't go to make order.
Zapier just stays in the boxthat you're hosting.
and it makes it all cheaperbecause now you're paying$6 a
month for hosting, doesn'tmatter how many automations
you're actually running.
So there, there are manydifferent reasons why to use,
uh, NA 10.
(10:16):
Uh, that being said, I wouldn'tstart with NA 10 just because
it's more complicated, unlessAnna, you disagree.
Anna. Bui (10:23):
I, I agree with you.
I started with Zapier.
I started with make, and thankyou so much for scaling things
out.
I was just like, jump rightstraight in because no, no, no.
So
GMT20250821-154502_Recordi (10:30):
now, now we can jump straight in.
so, and you can actually gothrough the steps on how this
thing works.
Anna. Bui (10:35):
But, uh, I completely agree with you.
The reason why I avoid Zapier,because it is tend to get more
expensive.
Yes.
It's had a lot more connector.
It is gonna make your first stepinto automation, a lot easier
because it's verystraightforward.
It's just step by step like awaterfall.
It can do a lot, but it is gonnaget very expensive the more,
workflow you're trying to run.
Yeah.
So that's why I, same as you,um, with everything that you
(10:57):
just list.
That's why I choose N Ends as myfavorite.
you know, automation platformform.
Like you can self host it or youcan have this, clouding, as
well.
And one of the thing I probablyin the future you'll probably
see more and more is the MCPserver.
You can host the MCP within any,within the cloud or like, or
like, uh, having on your local,on your local hosting itself.
So it's definitely a verypowerful platform, but.
(11:21):
One of the things that peoplekeep reaching out.
Sorry, before we jumping into,the flow, uh, people reach out
to me and ask how can they startwith NNNI think the best way to
start is just grab one of thetemplate that you see on the
community.
They have, they have more than4,800 template.
So just you can search it outand then even with your tech
(11:41):
stack as well, just grab atemplate and play with it.
And the more you play with it,the more you feel.
That is very straightforward.
The community of N End is verypowerful as well.
Everybody contributing.
You will see a lot of peopleposting about their workflow,
about their problem and everyonethat I ever encountered that,
enjoy and building things on anyend, they're very kind.
So please feel free to justreach out to them asking a bunch
(12:03):
of questions.
But, I highly recommend you justjump straight in and just start
playing with the template.
It doesn't have to be perfectbecause they are template.
You supposed to modify them tosuit your need.
So the more you build a workflowthat solve your own problem, the
more you learn it very quickly.
GMT20250821-154502_Recording_ (12:19):
A hundred percent.
I'll add one more thing and thenwe'll dive into the actual flow.
Uh, when you do grab thesetemplates, you don't have to use
the, the entire template.
So you can consider the buildingblocks of the template, say, oh,
this segment, these four steps,they do this thing as part of
the 20 steps that the automationin the template do.
And I can use these four stepssomewhere else because I need to
(12:39):
do this one step for a differentthing.
So this way you can mix andmatch between different
components and differenttemplates, knowing very, very
little and still buildingsomething that works because
you're taking these legos thatsomebody has already built into
a part of a solution andcombining them together into a
complete solution.
But now let's dive into ouractual workflow.
We, we definitely given enough,uh, intro to this.
Anna. Bui (13:00):
Yeah.
So the workflow is gonna splitinto three session.
We have the trigger right here,and then we have the whole
process, as you can see.
Starting from here and then thisis the end, like the end result
is gonna have the record createdon your Airtable or whatever
database, uh, storing you use.
For me, I use Airtable'causeit's very straightforward and
easy for beginner.
So basically, these are the twoact appify actor.
(13:22):
So, um, so this AFI actor isserved, to scrape the link post
for you.
Like the script, the linkreaction for you is basically a
like a little special key thatgoing like unlock the dog link
in and like, okay, this is postreaction, let's just unlock it.
Get all, get everything out likea little teeth because, well we
all know how LinkedIn, be veryparticular with people with
(13:43):
automation and scraping.
So they are like a little teethunfortunately.
And um, there's two actor areusing.
So the first one is, okay, Ihave my post with 79 reaction.
How would I able to get theprofile of the 79?
like those 79 people?
And here is the one that I use.
Oh, sorry, this is the firstone.
so this is, yeah, reaction.
(14:05):
Oh, sorry.
It should be this one.
My apology.
So this one is the link portreaction scraper.
So this is your ap, your Appifyconsole.
And I'm just gonna show you twothings that you need to, you
just need to look into the firstone.
Just a second.
GMT20250821-154502_Recordi (14:18):
Just one second.
To those of you who don't know,appify is a platform and you can
get multiple different toolsthat are all API based that you
can use in automations andapplications that you're
building and so on.
And they are geared to do thingsthat are very, very specific.
So you go into Appify and thenyou can find, again, many of
these different things.
This particular one knows how toscrape LinkedIn for post
(14:40):
reactions.
Anna. Bui (14:42):
Thank you so much.
Um, yeah, appify is, is have alot of different, uh, API like
a, uh, endpoint Lexus.
So you can even find like, okay,Google map and vendors or like,
Instagram followers, like allthose sort of things.
So please.
And they have a lot of, youknow, actors like, serve
specific need.
But for this one we have the, wefocusing on the link and we have
(15:02):
the first of all, the link postreaction scraper.
So we scrape, uh, scrape thereaction of that particular
link, post the information thatyou probably need to see, is
that the, first of all, theinput parameter?
So basically we need to sendthis parameter to, um, to this,
the endpoint.
So basically you go to input andyou will see here it had exist.
(15:25):
And this information for postURL, you can find it in here.
So when you go to this link,this particular, uh, number is
here is the, unique, uh, numberfor this, uh, link in post.
So you add it there.
So this amplify will startextracting the reaction from
that, from this one.
GMT20250821-154502_Recording (15:45):
So again, for those of you who
don't see, it's the end of theURL of your post, right?
So those of you are not watchingthe screen because you're
listening to the podcast asyou're driving.
it's the end of the URL.
There's like a long numbercharacter kind of like thing.
You grab that, that's the whatappify needs to know what post
to script.
Anna. Bui (16:04):
Yeah.
Perfect.
And so basically this is, I'mjust call, so this e so okay.
In edit and you have lot ofnative note and all the native
note I using in here is, youknow, you request for, um.
An H TT P request.
You have a wait note if, so,this is our native note within
NN and in the first one we aregonna use is the, uh, HTPP
(16:26):
request.
So this is the endpoint on theURL to call to cause this, to
cause this appify, actor.
So you, you can see right hereis actually the, um, it's
actually the uni, the link, um,yeah, yeah.
Link to this, um, to thisendpoint.
And then for, and then afterthat you have your, you have
(16:47):
your token.
So, um, when you open an a, anappify account, you will have
your unique token and you justadd it at the end.
So that mean it's gonna use acredit, it's gonna use a credit
up your account.
And then as you can see here, Iput money here so you only use a
credit of your account to startusing this actor.
Does that make sense?
GMT20250821-154502_Record (17:08):
yeah, yeah, yeah, yeah.
Anna. Bui (17:10):
And then from here, we are gonna send, as you can
see here, we sent, uh, we sendthe, we add the port, URL here,
and then the page number is one.
So going back to this actor, theinformation that you would like
to add in, as you can see here,is page number.
So page one, return one to 100.
So a reaction.
So that means, if you have apost at under 100, you just need
to post at one, or by default,then it will be the page number
(17:33):
one.
But if you have a very viralpost, you have like, 500 people,
um, interact with it, then youhave to do like page three and
four and so on.
But to keep it simple is gonnabe one, and it's going to
extract all the 79 reaction fromthis link post through here.
As you can see.
Yeah, there's, there's an
GMT20250821-154502_ (17:50):
interesting question on LinkedIn.
do these tools comply with theLinkedIn user agreement?
Basically what LinkedIn allowsyou to do, from their
perspective?
Anna. Bui (18:02):
I don't, I'm not sure about that 100%.
I just know that from myknowledge is, whenever you go to
Appify and whatever with link,specific scraping actor sort of
thing, try to use no cookie.
So if you use no cookies, thatmean like whatever action, the
action I'm doing right now,which is scraping my, the
reaction from my LinkedIn postright, is from the person who
(18:23):
built the actor itself, not fromme, because I didn't associate
my LinkedIn account with this,with this scraping processed.
it's just an API call fromanother thing that the other
person built.
That my LinkedIn profile didn'tattach to it.
GMT20250821-154502_Recordi (18:36):
Yes.
So, so there's
Anna. Bui (18:37):
no cookie key associated, there's nothing to
kind of like track my behavior,if that makes sense.
so that is why, so there's,
GMT20250821-154502_Rec (18:43):
there's, what I will say is when it comes
to LinkedIn automation, thereare two aspects.
One is what you can get.
Off LinkedIn, and it is veryhard for LinkedIn to know who is
actually doing it.
Like Anna just said there,there's no way for them to know
who is actually scrapingLinkedIn.
Like they can try to block it,but they can't know it's you,
uh, because it's coming in froma specific application.
The other aspect of of LinkedInautomation is if you want it to
(19:06):
post on your behalf, on yourregular basis, and this is where
you might get in trouble, right?
If you're building automationsto engage on your behalf
automatically, this is where youmight get in trouble because
then it's you, it's you, it'sengaging on your behalf.
And then usually the bestpractices on these tools is just
to use them at whatever levelsthey tell you that they checked
and that are okay.
(19:26):
It's still against LinkedIn'srules and regulations, each and
every one of these tools willtell you, okay, if you do more,
no more than 30 a day, you'llprobably be fine.
And then you can decide whatyour level of risk to decide how
much you wanna push thatenvelope.
Anna. Bui (19:38):
Yeah.
And um, I recently, um, I'm on acall with, kga.
I probably, butcher his name,but basically he is, um, like an
expert in the Clay Bootcamp whenit's come to, uh, Lincoln
Outreach.
And he told me from very, thevery beginning, it's gonna, it
is there will be a risk, therealways gonna be a risk when we
come to, uh, LinkedIn outreach.
it's not really something that,um, Lincoln promote and not
(20:00):
something that li Lincoln wantpeople to do, like those sort of
automation or, um, you know,interact on, on your behalf,
other sort of things.
So there will be a risk if youdecided to do a sort of linking
outreach campaign, but that willbe on your own risk and you
probably have to accept that forthis one.
go back to what AU justmentioned, is an application
from another person, et cetera.
(20:21):
So I guess I was just.
Being like hidden.
I hope, I guess.
Yeah.
Okay.
So
GMT20250821-154502_Recordin (20:25):
the first step we use Appify to
scrape the people who actuallyengage with the prompt.
What happens then?
or that engage with the post,sorry.
Anna. Bui (20:32):
Yeah.
engage with the post.
And after that we have like 77items coming in.
It's like, oh my god, it's toomuch, too overwhelm.
So that is why we have this oneis a loop over item, which is
another native, uh, loop, sorry,another native note in n at end
it's basically split, split thething in batches.
So it's break down the 77 itemto just do this thing once, one
(20:54):
item at a time.
So that mean we gonna, it'sgonna, one item's gonna run
through this whole processed.
And when it's done, then thesecond coming in and the third
coming in and the fourth comingin.
So it is basically try to avoidover dry like overflow, over
flooding, sorry, over flood.
So, um, the system and to have,and also let the, um, each
person have time to, sorry, eachof the items have time to
(21:17):
actually process and go throughwhatever, you desire the
outcome.
Awesome.
And yeah, so here the first itemcoming in, I always recommend
people to try this set note.
So this is another, native nodein NN is just basically to clean
your data.
So whenever you how to say this,the more I work with this data
(21:37):
and the more I work withscraping.
I realize that you have to startcleaning your data from the very
beginning so that from to thevery end, it's not gonna guide
you.
get you a lot of headaches.
The same thing when you startbuilding your play table.
Always make sure your data isclean.
So I, I guess like for everybodywho enjoy data and then data
analysis or data strategies,they, you, they, you probably
agree with what I'm saying.
I try to clean things.
(21:57):
Well, let, let's
GMT20250821-154502_Rec (21:57):
explain, let's explain in simple English
what that means.
Cleaning the data in thisparticular use case, so we got
scraped information fromLinkedIn.
We're now sending the firstitem, right?
Because we're gonna do this oneby one.
So the information that comesis, profiling, first name, last
name, company, you know,everything.
It knows that it can pick fromthat.
What does the data cleaner do?
Anna. Bui (22:17):
Yeah, so this basically, is just to spit up a,
a.
How to say this?
Split out the, as you can seehere, the data is, there's a lot
of data here.
They have like the profilepicture, like everything like
that.
We don't need that.
You only need what important.
So this is just basically cleanit out and able to just give you
what you desire.
So I just want, so profile urn,I just, I just want to get the,
(22:41):
okay.
Uh, had to explain it a a littlebit.
We have the Lincoln URL, whichis your own link.
and your own profile.
Link Profile, yeah.
Profile link.
And then you have the URN, whichis the unique number A
associated with your accountwith your LinkedIn account.
So the URL can be changed, likeyou see like with my name Anna
Bui, something, something, maybeI can change this to a different
(23:02):
name.
So the URL can change, but theURN will never change.
So that is.
This is like a primary data ifyou want to do some sort of
lookup.
So it's because it's reliable,it's reliable data.
The your end.
Okay, going back, I just want ajob title.
I just want link in URL, I justwant this and the name.
That's all I need for the nextintegration of scraping to use,
(23:24):
which is just a very clean,straightforward data.
So it's not gonna mess up thesystem or like mess up the se
sequence of scraping because nowit's like, okay, I'm not gonna
give you all this crazy thing.
I'm just gonna give you thisfull, straightforward
information right here becausePerfect.
GMT20250821-154502_Recording (23:39):
So it, it's does two things.
A, it filters the lots of datayou got into just the little
data you need.
And the second thing it picks.
What particular type of datayou're gonna get.
So like you said, the URL maynot be the right thing.
The URN is the right thing.
So that's what I want totransfer forward, as far as the
ID of the person.
So this is basically as thefirst step of everything else
(23:59):
because this will set the stageto these are the attributes that
I'm actually going to use, uh,moving forward in my automation.
Awesome.
Perfect.
Great explanation.
next step?
Anna. Bui (24:09):
Yeah.
Um, just want to, I just want toadd one more thing.
The reason why I have to cleanit from the very beginning,
because it's all these othernode can actually use this for
reference.
Yeah.
So it's gonna be like a veryreliable, uh, data set.
And then after that I'm check ifthere's any duplications in my,
um, in my Airtable.
So as you can see right now,it's blank.
So I'm just gonna check okay.
If.
(24:29):
if it is actually blank.
so this is mean that if there'sno data, I would say this is
data.
So this means the data lang.
So that's, there's likeinformation here is true.
Then this mean it's gonna startcreating, it's gonna start with
all this increment.
And then if is already exist,then it's nothing to do.
But basically, uh, just tosummarize, this is a check note.
(24:52):
Just it's just basicallychecking if this person already
on my database is this person,haven't, then Okay, let's start
with the next one.
Yeah.
GMT20250821-154502_Record (25:00):
Makes sense.
Anna. Bui (25:02):
So now is check that okay, this person haven't exist.
That mean is gonna, is gonnacreate a new record on, on this
new record.
It's gonna have, it's gonna addinto all this feel on Airtable
here.
It's gonna start mapping it out.
Right now it's nothing.
But when it's run, you willactually see the information
coming in and.
Is, you know, this is Airtablenote that, um, with NNN it's
(25:22):
integrating in NNN.
So it's very straightforward.
You just kind of just map thingsout.
And as you can see here with theURN, I just put it like this and
it's referenced the note.
That's why you clean the data.
GMT20250821-154502_Recording (25:31):
So again, for those of you are not
watching, and don't knowAirtable, Airtable is like Excel
with a better user interface ifyou want.
It's just a database where youcan define rows and columns and
you can put data in them and toconnect the two together.
So now we have four pieces ofinformation on each individual
that we scraped.
You basically connect theAirtable node, and that will
bring into NA 10, the fourcolumn headers that already
(25:54):
exist in Airtable.
You can do the same exact thingwith Excel, and then all you
have to do is drag theparameters that you brought from
the scraper.
And drag them into each andevery one of the columns.
So you drag the name to thename, the id, to the id, the
link to the link, whatever itis.
Uh, and that's it.
And that's how you set up.
And now what's gonna happen isas this first data comes in, it
is going to check, does thatdata already exist?
(26:17):
It's probably going to checkbased on the URN that we now
know.
That's the unique identifier ofthe person.
If that does not exist, it willcreate a new line in the
database.
Uh, or in this case on the, uh,Airtable table.
By the way, I, I'll say one morething you say, okay, why, why do
I even need to check?
Why do you care if there'sduplicates?
there's two reasons.
A duplicates are never a goodthing when you are creating a
(26:37):
database.
But the other thing is, some ofthe few steps will actually send
you to different large languagemodels, which means you're gonna
be paying for tokens.
And if you can minimize theamount of tokens you're paying,
uh, then you're just gonna saveyourself money and time.
Anna. Bui (26:49):
Oh yeah, definitely.
I, I've been there, done that.
I scraping one person atmultiple times because I was
testing it out.
Highly recommend you havesomething to check, to look up
your, uh, your data so youdon't, um, for sure.
Waste second.
Okay, now for the fun part, thereason why, I have people asking
me, okay, now you have thisperson pro.
So from the previous, appifyactor is give us this person,
(27:11):
uh, profile ID and then the, thewebsite, the UL.
Right now we actually need toenrich that.
Now we actually have to scrapethis person, personal link, uh,
LinkedIn profile.
Hence we will move to, s sorry,this actor right here.
So it says, two actor use inthiss workflow.
The first one is to scrape thereaction of the post and the
second one to actually enrich,the individual profile that
(27:35):
interact with your, yourLinkedIn.
If that makes sense.
GMT20250821-154502_Record (27:39):
Yeah, it's good.
Yeah.
Yeah.
So, so the, the what we gotbefore, because what we scraped
is the post is only theinformation that appears on the
post.
Now we want to go to the actualperson's profile and then we can
pull whatever we want.
We can see their company name,their company size, their
industry they're in, what theywrote about themselves, uh,
their title, where they workedbefore.
(28:00):
All the stuff that appears onprofile you can bring in by now
scraping their profile, which isthe current step.
Anna. Bui (28:06):
Yeah.
And for this one, uh, for thisactor, all you need to do is the
pro the profile, uh, BRL.
And I have it here.
I just added from this cleandata note added in the, um.
You know, add in the actor andthe actor is running and if gave
out all the information.
So the previous one just gave uslike very basic information, um,
like the name, the uls or theprofile, all those sort of
(28:27):
thing.
But this one, as you mentioned,is giving like your full name
and the headline and theyexperience the company that they
work in.
Basically it scrape this wholeperson LinkedIn profile.
And, uh, from that, it is justgonna, so the, from all this
information, it is kind of likespit up as as one item.
So I want them to combinetogether, combine the full name,
(28:49):
stuff like that, all theinformation that I need.
These, aggregate node, it quitestraightforward.
Just basically just merge allthis information, like all this
long list into like one cleanerlist.
And then from that we are gonnahave our AI.
Classify if this person isactually part of our ICP.
So based on the information thatis that we scraped from this
person's, profile, we are gonnalet the AI know, like, okay,
(29:12):
based on the problem that I'mgiving you, as you can see here,
you are an AI classifier thatdetermines if this person is the
ideal, is the ICP or not.
So I add in my ICP definition,like the people that I, the
roles that need automation.
Basically I paste my ICP here,and I also gave, uh, gave it an,
uh, output format.
is this ICP, yes or no?
The reason why this was an, uh,this person is an ICP and it
(29:36):
just need to give, give outthose information based on the
incoming data, um, that, fromthe scraping itself, S, but this
is more like into the promptingengineer and everybody have
their different ICP.
So always make sure that youcustomize it to suit your need
to able to give up the correctinformation.
GMT20250821-154502_Recor (29:53):
Quick, uh, help for people to test this
out.
So if you run this here on theautomation, every time you run
this, you're gonna pay fortokens.
It's very, very cheap.
Like you can probably run this athousand times and it's gonna
cost you a dollar.
But, uh, if you wanna make itcompletely free, uh, use any of
your existing AI tools or theone that you're actually gonna
use for this.
So if using Chat g pt, use chat,g, pt, cloud, Gemini, whichever
(30:14):
you're gonna use for the API goto that tool and then give it
the information from thatperson.
Take a screenshot from LinkedIn,drop it in there and use your
prompt that you're gonna use inthis automation here and just
test it out and just keep onimproving the prompt until you
get the proper answer everysingle time.
And then start running itthrough the automation, because
then testing it is gonna costyou nothing.
Anna. Bui (30:35):
Yeah.
And one of the sessions that Ihave at Clay Bootcamp from
Spencer Teil, he is excellentand he was breaking down to us
how to actually.
And engineers is prong.
So at first I was like, oh, AIprompting easy.
Not a big deal.
The way that he break it downhelped me a lot with this
project, because you are notjust gonna give the AI
(30:55):
instruction, you're gonna giveit knowledge as well about your
per, your ICP.
And not to mention, you have togive up, a required format that
it need to follow, an examplethat it will need to see and
reference.
So the more so the informationthat you're giving into ai, the
more, structure and the moredetailed information go to the
ai, the better the result.
And as you can see here, um, Iask it to like, just give me two
(31:18):
output, which is, is this anICP, true or fails?
And the reason why you thinkthis person is not an ICP.
That's just it.
Very simple.
GMT20250821-154502_Rec (31:26):
Awesome.
Okay.
Anna. Bui (31:28):
And then lastly is that I'm gonna have this, okay,
now we have the information ofthis person.
This person is first of all, um,couldn't find this person.
Uh, email address, which couldhappen because sometimes the
scraper unable to scrap theemail address.
That's some, you know, it's ahit or miss.
Um, sometimes it able to findit, sometimes not.
But basically ICP is not an ICPthen the reason, so right now we
(31:51):
just need to update the recordthat we create from the very
beginning with the reasoning andthen the checker, if it's an ICP
person or not.
And that was the whole flow.
Now all is coming, going back tothe loop note and then the loop
on.
Do integration and integrationuntil it scrape out all the, it
did to all the 77 items.
GMT20250821-154502_Rec (32:12):
Awesome.
First of all, incredible,really, really powerful
capability that again, uh, we'llprovide the link to this, uh, in
the show notes.
So anybody who wants access tothis, Anna has, the way I found
her is because I saw her sharingthis, this thing on LinkedIn.
I'm like, oh my God, this isawesome.
I want everybody to know.
And so that's why she's on theshow.
So she was, Gracious enough toshare this with the world.
(32:33):
And we're gonna share this onthe show notes of this episode
as well.
Uh, so you can get access to itand copy it.
And like Anna said, not justcopy it, but also learn from it.
So understand what each and oneof the components.
So what you can do is you canactually open this on your
computer and watch the video asyou're looking at the actual
thing and dive into each andevery one of the components so
you can understand what it does.
And that was the main reason whyI wanted to dive into this and
(32:54):
show you step by step.
I want to touch on a couple ofmore things, uh, and then I'll
go into a few of the questionsthat, that are coming from, from
LinkedIn.
But, aspect number one that Iwanted to touch on is if you are
still working on your prompt forthe agent that classifies,
whether it's a.
(33:16):
Potential customer or not?
Right.
Does it, is it aligned with yourICP or not?
One of the things that you cando, and I have a very similar
process myself and I have amaybe option, so when it's not
sure it's spinning out a maybeand that enables the AI to
basically okay with not beingsure.
And I want it not to be ahundred percent sure when it's
not a hundred percent surebecause otherwise I may miss a
(33:37):
client, it decides that it's nota client.
And so what I do with my maybeson my table is I go and manually
inspect them.
Right?
And so that gives you, uh, theopportunity not to lose people
that actually might be goodpotential clients for you.
And so every time it's not ahundred percent sure I.
Allow it to say, well, I'm not ahundred percent sure and then I
(33:58):
can go and check myself.
So that's number one.
Uh, the second thing that I wantto ask you, Anna, uh, and I know
the answer, but I want you toexplain in theory, because we
don't have it in front of us,what's the next step?
So let's say I have all of them.
How can I create an engagingoutreach to them that will
actually capture their attentionand will allow me to start a
conversation with them?
Anna. Bui (34:18):
Oh, okay.
That, that is a great question.
So at first, lemme just, uh,show you how this workflow,
let's just test it to see howit's run.
So right now you see that'sblank, right?
Yeah.
And after I run this wholeworkflow, as you can see here,
GMT20250821-154502_Recording (34:34):
so for those of you not watching,
you can kind of see the thingrunning.
It shows you what step it'sdoing right now.
And now we have the first fieldor the first line populated.
On, Airtable and it's gonna keepon doing this.
I think it goes through theentire 76.
And for H one, we will initiallysee just the basic information
coming in from the first step,and then we're gonna see the
enrichment.
If it is the ICP, it's becausethat's how the automation runs.
Anna. Bui (34:58):
So I think the, just to answer your, uh, your
question is the way to build theoutreach is, how to say this,
this is where Clay come to play.
So you have this, you have this,this, all these people profile
UL, right?
Is that correct?
You have all this information soyou can, so with Clay, I mean
you can use other two as well,but I have to dig in a bit more.
(35:18):
But what with clay is that theyhave, you have clay gen and all
of this information that you areable to scrape from this person,
is gonna, click and able toenrich this person, uh, data
also based on your UOL as well.
And then from that you're gonnause the clay gen, which is like
a, an ai, agent in clay toactually write out the post for
you, uh, to write out theoutreach for you, because it's
gonna based on all the, um, allthe personalized, parameter of
(35:41):
this person.
So for example, this person is,a title is AI automation and
developer.
Maybe you will write somethingabout like, hey, swale, I saw
that you also an AI andautomation developer, and then
it starts writing it out.
And that is where Clay come toplay.
Because the reason, the reasonwhy I recommend Clay for that
is, is able to interact with somany different, variables, like
different specific variable,personalized variable of this
(36:04):
person.
And it can create, um, thisoutreach.
One thing that I recommend isthat you should have a draft
outreach that you would like theAI to follow.
Hey, make sure you use thecorrect name or like.
This should be an example emailthat you will use, like this one
email that you find is very goodenough for outreach.
You let the AI see that exampleand then from that it's going to
interact with differentvariables that is, was able to
(36:26):
scrape from play itself, thatable to enrich from this person
information, which is theprofile out to use that to write
out the, uh, outreach, withinClay itself.
And then from that, clay canactually send the outreach to,
it can send it to um, it cansend it back to NN or it can
send it to landless or smartlead or something like that.
So that is like the differentprocess, uh, process if you want
(36:46):
to combine.
again, just one more thing.
If you want to combine n it endclay and is that let n it end do
the things that, to save thecredit from clay, for example.
as you mentioned, it's gonnacost you like two,$3 to scrape
1000 people to get the profileUL from that.
You export that as CSV, uploadit to Clay as a new table and
then use enrichment, functionof, of clay to actually extract
(37:07):
the profile you are with evenmore personalized information
and then write the email fromthere and that when you see the
email outreach from that personwithin Clay, shoot it out to
somewhere else.
So that how you kind of have tobe strategic of how you would
like to have your outreach veryhighly personalized.
And with clay, the more narrowand filtered your list, the
better that mean, like the listthat you're going to put into
(37:29):
clay has to be 100% somethingthat you would like to target
the list that you absolutelywant to nail.
Then you're gonna use clay.
And Clay gonna exceed with that,with the outreach and writing
very personalized, email.
Okay.
Awesome.
I will
GMT20250821-154502_Recordin (37:41):
say one, one more thing.
I think you can still do it herebecause all you have to do is
build a second agent, a secondAI agent in your NA 10 process.
That does exactly the samething, right?
You can give it all theinformation from the previous
steps.
You can say, here's the name,here's the title, here's what
they wrote about themselves inthe profile.
Here's the size of the company.
Like all the stuff that youhave.
Here's the template of the emailthat, that I want you to use, or
(38:06):
the template of the message thatI want you to send, uh, to base
it on.
And then I want you topersonalize that based on all
the information that we'vecollected in the previous steps.
And you'll do the same thing.
And then you can put the outputin Airtable, so you can have a
column in Airtable.
That would be the recommended,thing to send.
I agree with you that it ispotentially easier to do it in
clay.
and like you said, clay can thenconnect to a gazillion other
(38:28):
places that can continue, uh,the process.
Uh, I'm not taking anything awayfrom Clay.
I think it's an awesome tool.
Uh, and like you said, it's notnecessarily a bad idea to
combine the two together to getthe best of both worlds.
A lot of people are saying thankyou and this is great, and this
has been awesome, and that theyreally like the flow.
If, and I, and I agree.
I think this is a, uh, first ofall, it's a, it's an awesome
flow, but it's also, uh, you,you did a very good job in
(38:50):
explaining how it works.
Uh, if people wanna learn moreabout you, learn from you, work
with you, take your courses, uh,hire you to build automations
for them, what are the best waysto do that?
Anna. Bui (39:01):
please feel free to reach out to me on LinkedIn and
from that you'll see all theresources.
and I will keep posting thingsconstantly.
and maybe again, uh, on the,maybe on the next episode,
hopefully I can tell you,actually build out the, um, as
the AI agent that can write theemail for you.
Uh, I was just thinking aboutscalability, you know?
Yeah.
But anything all good, but like,when you start scaling things,
you probably want something morerobust.
(39:21):
But yeah, thank you so much forhaving me, and I'm so glad that
everybody in, you know,hopefully you enjoy the session
and find it.
Um.
Helpful.
GMT20250821-154502_Record (39:31):
Thank you so much.
This was absolutely fantastic.
Thanks everybody for joining us.
Thanks for, uh, being veryactive.
There's a lot of chats happeningon LinkedIn and on Zoom and
people asking great questions.
I didn't bring all of them up.
I answered some of them myselfif I knew, uh, the answers.
Uh, but thanks everybody forjoining us again.
I know you have other stuff todo.
Come join us next week everyThursday, 1:00 PM Eastern.
We're gonna do something likethis with another amazing expert
(39:51):
like Anna.
and thank you obviously, Anna.
This was amazing.
First of all, thank you forworking on this and thank you
even more for coming and sharingit, uh, with us.
Anna. Bui (39:58):
Yeah, thank you so much.
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.
Cardiac Cowboys
The heart was always off-limits to surgeons. Cutting into it spelled instant death for the patient. That is, until a ragtag group of doctors scattered across the Midwest and Texas decided to throw out the rule book. Working in makeshift laboratories and home garages, using medical devices made from scavenged machine parts and beer tubes, these men and women invented the field of open heart surgery. Odds are, someone you know is alive because of them. So why has history left them behind? Presented by Chris Pine, CARDIAC COWBOYS tells the gripping true story behind the birth of heart surgery, and the young, Greatest Generation doctors who made it happen. For years, they competed and feuded, racing to be the first, the best, and the most prolific. Some appeared on the cover of Time Magazine, operated on kings and advised presidents. Others ended up disgraced, penniless, and convicted of felonies. Together, they ignited a revolution in medicine, and changed the world.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.