Full transcript
0:00I'm about to give you the complete
0:01tutorial for using social signals in
0:03Trigger Five, which is a platform I've
0:05developed myself. And since implementing
0:07it in my own sales and marketing, I've
0:08used it to generate millions in
0:10pipeline. It's a tool that you can use
0:12to time your outreach to the exact
0:15moment a prospect is thinking about what
0:17you sell, create content that's specific
0:20to your ICP is looking for, and catch
0:22churn before it happens, and many, many
0:25more use cases. So, in this video, I'm
0:27going to show you exactly how to use
0:29Trigger Five and how to leverage social
0:31signals across sales, marketing, and
0:33customer services. So, let's dive in.
0:35The first thing that we need to think
0:36about when it comes to using social data
0:39is what signals do we want to get?
0:41Because when we think about a post,
0:43let's take a LinkedIn post. If a
0:45prospect is engaging with that post,
0:47that might be multiple different signals
0:50inside of that one post. And let me
0:51explain. For example, if you're a
0:53salesperson and it's an ICP individual
0:56engaging with a LinkedIn post, that
0:58might be a signal to you that they are
0:59interested in a certain topic, depending
1:01on what that post is about. Maybe it's a
1:03customer engaging with a competitor,
1:06that's potentially a churn signal. And
1:08maybe we're analyzing millions of posts
1:10talking about a particular product,
1:12which is going to give us some insights
1:13and some signals into where our own
1:15product roadmap should be based on the
1:17frustrations that general users are
1:19thinking and feeling. And so, you can
1:21see here that actually there are
1:22multiple different lenses that we could
1:24apply into social data to pull out our
1:27own social analysis or our own social
1:29signals. So, I'm going to show you
1:31exactly start to finish how we can set
1:33this up using Trigger Five's kind of
1:35custom workflow feature. First things
1:37first, we need to understand the bedrock
1:40of the data. The bedrock is essentially
1:42a place that we're going to go to create
1:45a social listening search. So, we can
1:46monitor things across LinkedIn, X,
1:49Reddit, YouTube, Threads, Instagram,
1:52GitHub, Substack, and so much more to
1:55create the foundation of the analysis,
1:57which is going to give us our signals.
1:59So, if in this case, let's pretend that
2:02I am creating or I work for a company
2:04that sells data enrichment and I want to
2:07monitor anytime someone is perhaps
2:08frustrated about competitors such as
2:11Apollo or ZoomInfo or something like
2:14that. So, the first thing that I'm going
2:15to do is I'm going to pick LinkedIn here
2:17and I'm going to title the search name
2:19as Apollo. And I'm going to search for
2:21anytime someone is talking about Apollo
2:24and ZoomInfo. And so, here we have now
2:27all of the searches coming in, anyone
2:29mentioning these particular keywords
2:31like Apollo, for example, inside of
2:33here. Now, what I might want to do is I
2:35might want to link this to something
2:37like data to make the searches a little
2:39bit more specific, for example. Or what
2:43I want to might want to do is I might
2:44want to say, "Okay, do you know what?
2:45I'm going to do this instead. I'm going
2:47to say, 'Okay, hey, anytime Apollo is
2:49mentioned, link to contact information
2:53or data enrichment or email or cell
2:58phone or whatever it is, that's
2:59important to me.' From here, I can start
3:02to exclude certain keywords such like I
3:03don't want to have anything about hiring
3:05coming in, that's not relevant. And I'm
3:07just going to take a quick little browse
3:08here. This all looks pretty spot-on for
3:10me. So, what I've done here is I've
3:12actually just edited the search and I'm
3:13actually finding anyone talking about
3:16Apollo. So, this is search number one
3:18that I want to do. I could actually
3:19further add some filters to this. I
3:21could say, "Okay, hey, I want them to be
3:22a CEO or founder, head of sales, sales
3:26director, a CRO, perhaps." There we go.
3:28So, now it's like even more, I think,
3:31relevant pieces of content and we're
3:33just going to save this. So, we're going
3:34to set that up and then I'm going to go
3:35ahead and do a similar search, but this
3:37time I'm going to do this on ZoomInfo.
3:39So, now same principle, I'm searching
3:42for anytime the keyword ZoomInfo has
3:45been spoken about. ZoomInfo I don't
3:47necessarily need to do a lot more kind
3:49of like joining keywords, so to speak,
3:51because it's a very unique name, and
3:54therefore anyone talking about ZoomInfo
3:56in the last week is most likely going to
3:58be talking about ZoomInfo. But, I am
4:00just going to get rid of like hiring
4:02content associated with that. And so, we
4:04can see here, just double-check
4:05everything. It all looks fine. So, I'm
4:07going to go ahead and save this. Okay.
4:10So, now what we have is we have the
4:12Apollo search that we just set up. We
4:13have over 50 results basically coming
4:15back in here, and we have ZoomInfo. So,
4:18at this point, we don't have any
4:20signals. We've just set up what I call
4:22like some intent-based searches. Now,
4:25what I might want to do is I might want
4:26to take this a step further. And so, I
4:29might want to be like, "Okay, I want to
4:32track, for example, the CEO of
4:35ZoomInfo." So, I'm going to go and grab
4:38his LinkedIn URL, and I'm going to set
4:40up another search, and I'm going to call
4:41this Henry ZoomInfo here. And I'll
4:44explain why I'm doing this in a moment.
4:46So, simple. Let's check the post in
4:48there. Validate that. There he is. Set
4:51this up, and now I'm searching for that.
4:52So, I've got three really good searches.
4:54I'm monitoring anyone anytime someone's
4:56talking about Apollo, anytime someone's
4:58talking about ZoomInfo, and then anytime
5:00that Henry, who's the CEO of ZoomInfo,
5:02posts, I will grab that post coming in.
5:05So, now we need to move on to phase two,
5:07which is pulling the signals out of
5:10these searches, because, like I said at
5:12the start, there's multiple different
5:14searches that we could or signals that
5:16we could pull from each search within
5:18here. So, we're going to go build some
5:19custom flows. And we're going to call
5:21this competitor ops. So, Triggerfy
5:24basically allows you to custom create
5:27your own signals. So, here I'm going to
5:29say, "Okay, I'm going to select the
5:30platform. We did some keyword searches,
5:32and it was these two searches that I set
5:34up." So, what this is saying now is
5:36anytime these two posts or these two
5:38searches get any new posts coming in,
5:40it's going to trigger this workflow. I'm
5:42then going to use an agent, and we're
5:43going to use Sonar. And we're going to
5:45say, "Analyze this post and let me know
5:49if the person is frustrated or upset or
5:54angry, etc. with either Apollo or
5:58ZoomInfo, the contact UTM providers."
6:01And then I'm going to feed it the post
6:02text. There we go. Um we're going to get
6:04this to give us a true or false. True if
6:08they are, false if they aren't. Okay.
6:10So, let's just run a quick test and see
6:12what this basically has to say on this
6:14particular one. So, what I'm doing is
6:16now I'm just checking that that agent
6:18field is basically working and operating
6:20the way that I want. Cuz what I'm having
6:22here is I'm basically using an agent to
6:24qualify whether this person is
6:27frustrated because if they are
6:29frustrated, then it's going to be a
6:31really good signal for me. And if they
6:33aren't, that's not to worry, there is
6:34still a signal that I that I can pull
6:36from this. So, this is false. Okay,
6:38cool. So, they're not frustrated. So,
6:39I'm going to build an if condition out
6:41of the back end of this. And if they
6:43aren't, so I'm just going to say, uh if
6:46the agent output equals false, contains
6:49false, then it's going to go down this
6:50route. And when we think about this, so
6:52they're not angry with the ZoomInfo or
6:54frustrated or annoyed or anything like
6:56that. So, therefore, I don't probably
6:58want to reach out to them. I mean, you
6:59could do, but they're not necessarily
7:01what I would deem as like a a hot lead
7:03because they if they are frustrated, we
7:05want to reach out to them, right? We
7:06want to engage with them because we we
7:09can have the opportunity of selling to
7:11them. And remember, we're doing this
7:12pretending I'm a competitor to ZoomInfo
7:14or Apollo. Now, if they aren't
7:16frustrated, then that's okay. If we look
7:19at the actual post that we did there,
7:21it's still people are engaging with
7:23topics around people talking about
7:25ZoomInfo and Apollo. So, if that is the
7:28case, then I might want to reach out to
7:29them because there might be some
7:30relevancy with people who are engaging
7:33with content around data contact
7:34providers. And I can reach out saying,
7:36"Hey, like I can see, you know, you're
7:38like looking in or interested in things
7:41such as ZoomInfo, blah blah blah, and
7:42you can kind of cross-correlate
7:44the joinings there, and they'll be a
7:46slightly warmer lead for you. So, what
7:48I'm basically going to say here is, let
7:50me type in this. If they're not, I want
7:52to go ahead and I want to grab the likes
7:56from the post. So, it said you can say
7:58like here, I want to grab up to 1,000
7:59people who are liking this post. Now,
8:02what that's going to do is that's going
8:04to go and grab all of them. So, then
8:05what I need to do is I need to filter
8:09down across this data that the likes
8:12node is returning me. So, basically what
8:14that says is out of the 1,000 people
8:15that have liked the post, it's now going
8:17to go through it one by one. So, if you
8:19zoom in here, we can say for each item,
8:22what I want to then do is I want to run
8:25a person enrichment. And I'm going to
8:27add the LinkedIn or the URL of that
8:30person. And if I just scroll down, I
8:32want to find it on the looped data,
8:34which is here, the profile URL. Add that
8:36in. So, now I'm going to get their full
8:38kind of like information, which means
8:40from here I could check to see if they
8:42are ICP. If they are ICP, I then
8:45obviously could send them into something
8:47like a Heyreach or something like that.
8:48So, the first thing that I'm going to do
8:50now is off the back of this person
8:52enrichment, I'm going to say if this
8:54person enrichment, the job title
8:56contains, let's just say for the purpose
8:58of this, marketing. We'll just say that
9:00that is ICP. Then what I want it to do
9:03is I want to get the email. So, if you
9:04type in email, we've got two different
9:07options. We've got Prospero, and we also
9:08have Surf. Surf has six providers inside
9:11of it, so it runs its own waterfall. So,
9:13I'm going to use that one, and I'm going
9:14to go ahead and type in the URL, and
9:16just make sure you're selecting the
9:17correct ones from the person enrichment.
9:19So, there we go. First name, last name,
9:22and then domain here, company name.
9:25Perfect. So, now I'm going to go get
9:26their email, which is validated. And I
9:28can just delete this. We don't need that
9:30there. And then once that's completed, I
9:32could then send them through into
9:34something like Instantly, for example.
9:36So, you want to go ahead create a
9:37campaign in there and then configure
9:39that. And that is one signal from anyone
9:42who is engaging with content talking
9:44about my competitors. Now, if it's true,
9:47which means if we go back to the
9:49original agent question here, which was
9:50are they frustrated? I'm going to use
9:52another agent again and I tend to use
9:544.6 for most things and I'm going to say
9:56this person is frustrated with Apollo or
10:01ZoomInfo and you provide me with one
10:05short sentence on why they are
10:08frustrated with these tools. For
10:10context, I want to reach out to them in
10:13a LinkedIn message saying "Hey, and then
10:16I'll do like name and then this is where
10:19I'd insert your response. So, it needs
10:23to flow." So, I'm going to say example
10:25would be "Hey, name. Um I can see that
10:29or I saw that you're a little frustrated
10:33with how Apollo customer service is
10:36working. Slow to reply." There we go.
10:38And that would be the example, right?
10:40And I probably would want to iterate on
10:41this a little bit and I might just say
10:43"Don't worry about the name, just give
10:46me that one-line short sentence." And
10:50then I'm going to add the post text.
10:51There we go. Perfect. And then hit save.
10:53And then from there, what I can do is a
10:55simple uh add them to like HeyReach
10:58where I'd build a campaign. I would add
11:00like a custom variable where we would
11:01send through the response from Agent
11:04One, which I might just retitle to like
11:06uh frustration agent, for example. And
11:09by the way, you can add your all your
11:10own API keys in there, so you can not
11:12get charged any credits. Configure your
11:13HeyReach campaign. And now, with one
11:16workflow, I'm basically getting two
11:18different signals that I could create
11:20from this. So, this is very GTM
11:23orientated and very GTM focused in terms
11:26of like sales perspective. Now, what you
11:28can also do, you know, and I've
11:30explained this in other videos is you
11:32can use this to analyze content to
11:34create content agents. We can use it to
11:36analyze stuff around customer success to
11:40see you know who's engaging with who,
11:41who's talking about what and we can pull
11:43and effectively use these workflows here
11:45to custom create any signals that we
11:48would want to find because we have all
11:49of these different tools that are
11:50arsenal built on top of this social
11:54listening data. So, if you want to get
11:56started and start setting this up, the
11:57Triggerfy link will be in the
11:59description. The top link, so click you
12:01can get your first signal list set up in
12:03minutes and it's free to get started
12:05with. So, again just go ahead and click
12:07on that link below in the description.
12:09And if you want to see how I get
12:10unlimited leads for my business using AI
12:12agents, then watch this video next.