Taylor Haren teaches you how to automatically create varied email messages based on job title of your site visitors.
0:00
So this is going to be how to send different emails to people, different email
0:06
variations
0:07
based on their job title.
0:10
And so our B2B is sending us data inside of this clay table via the clay
0:15
integration.
0:16
And we've been collecting this data for a while.
0:18
And so we have a lot of data to work with.
0:19
And so sends us all of this information.
0:22
One of them is title.
0:24
And so here we can add a filter that makes it really easy.
0:28
We can change our messaging.
0:30
There's a lot of different things that we can do.
0:32
And so let's say I only want to talk to salespeople, marketing type people, or
0:40
people who are
0:42
only manager level and above, and so on.
0:45
There's a lot of different things that we can do.
0:47
And so I think the best way to do that is to add an enrichment.
0:51
And if you type in AI, you'll see OpenAI GPT generate text.
0:58
And I think the best way to do this is to say, if title
1:19
has to do with marketing or sales and is a manager or above, then output.
1:34
Yes.
1:37
Only output.
1:43
Yes.
1:46
Or no.
1:52
And it shouldn't need too much more than that.
1:55
We'll see.
1:56
We use 4.0.
2:00
And then let's just run this on.
2:02
Actually, I'm just going to run it on particular rows.
2:06
So let's see.
2:07
Run.
2:08
So I'm going to do Run column, choose different rows, starting from B7 65.
2:13
We're going to run 20 rows.
2:24
And so what's great about doing it this way is that it's just super
2:27
straightforward.
2:28
Where if you try to use clays formulas, I don't think they can pick up on the
2:31
nuance
2:31
of it, right?
2:33
Where here, they totally can, right?
2:34
And so as we look through it, right, marketing manager, maybe that's one it
2:39
could have done
2:40
better.
2:41
But if you run that one, under a fiddle, pick it up next time.
2:46
But account strategist, sales manager, sales communications, VP of revenue,
2:53
senior leadership,
2:55
so it gets them all there.
2:57
It's not picking up people like CEO and founder though, so we can continue to
3:00
prompt it just
3:01
a little bit more.
3:04
Or is an executive or leader of a company.
3:21
Another thing too, I've noticed with AI is that if you capitalize these things,
3:26
it seems
3:26
to do better.
3:41
So now let's see how that one works.
4:01
Now we'll try it again.
4:06
Hey, it's just super good.
4:11
So you just keep prompting it until it gets it perfect.
4:14
The other thing that we can do right is we can give it examples so we can add
4:17
examples
4:17
from previous rows, which is extremely helpful.
4:21
I don't know why I just did that.
4:26
Add from previous rows.
4:29
Choose row.
4:30
To give an example, let's do the one that I want to give an example.
4:42
And what it should say is yes.
4:46
We can add another example.
4:51
And we can just pick the ones that it didn't do a good job with.
4:55
So Marky Manager, see it got it this time.
5:00
Non-resident fellow.
5:06
What even is that?
5:07
Non-resident fellow.
5:10
Answer for that.
5:28
Oops.
5:33
My bad.
5:34
I'm going to put in 5.7.69.
5:37
There.
5:38
Answer should be no.
5:41
Chief Financial Officer Technically, it should get that one.
5:48
Program Analyst should be no.
5:59
So you can basically just keep prompting it like this until it's perfect.
6:06
Okay.
6:09
I just needed a few examples.
6:23
And then boom.
6:25
Now director, sure there.
6:30
Yep.
6:31
Yep.
6:32
CFO, sure.
6:33
Because we did tell it anybody, director level and above leadership and boom.
6:37
Now it is perfect.
6:41
So now what we can do is any rows after this, we can have a conditional formula
6:44
that looks
6:45
at this and it says hey, if this says yes, then we're going to send them this
6:48
messaging
6:49
instead.
6:50
If it says no, then we're going to send them this different messaging or maybe
6:53
we don't
6:53
Send them any messaging at all.
6:55
[ Silence ]