Agentic Community Signals Driving Pipeline

03:56:04.130 –> 03:56:09.280
Julia Nimchinski: And we are transitioning to our demo part in a second.

04:26:36.310 –> 04:26:40.939
Kevin White: Hello! Hello, Fan of Vidyard, too! That’s cool. I’m going right after Vidyard. That’s great.

1395
04:26:41.390 –> 04:26:42.349
Kevin White: Good stuff.

1396
04:26:43.100 –> 04:26:48.320
Julia Nimchinski: I’m trying to figure out. I I have to say that. Is it count intelligence?

1397
04:26:49.000 –> 04:26:50.440
Julia Nimchinski: Isn’t it real? Kevin?

1398
04:26:50.780 –> 04:26:51.570
Kevin White: Oh, yeah.

1399
04:26:52.230 –> 04:26:55.852
Kevin White: Human Kevin. Not human intelligence. No.

1400
04:26:56.820 –> 04:27:02.640
Kevin White: yeah. I maybe I I could use some help on the demo from Kevin intelligence, so we’ll see how it goes.

1401
04:27:03.380 –> 04:27:04.240
Julia Nimchinski: Let’s dad.

1402
04:27:04.665 –> 04:27:07.220
Kevin White: Cool, so we should just go for it. Get started.

1403
04:27:07.880 –> 04:27:21.351
Kevin White: Alright one sec. Let me just make sure oops didn’t wanna not share my video. I want to share my screen. Okay? So I know I got 15 min, and I promise not to make this too much slide where?

1404
04:27:21.680 –> 04:27:44.270
Kevin White: there we go. Okay. So the 1st thing I wanted to start on is just like for the audience here. It’s just like a primer on what common room does, because I feel like it sets the context for how agents work on top of all this data. And so this is the one slide I use to describe what common room does. It’s con. It’s a platform for generating pipeline. That’s kind of like the common through line.

1405
04:27:45.081 –> 04:27:57.310
Kevin White: But the the 3 pillars of it are one capturing signal buying signals, from every possible corner of the Internet possible, including using AI to capture, bespoke signals.

1406
04:27:57.682 –> 04:28:19.607
Kevin White: And then unifying the person and the accounts behind those signals. So that you know that the same person that’s visiting your website is the same person who’s just changed jobs. Or that’s like commenting on social or whatever, and so unifying that all into a single user profile, and then rolling that up to the account level and then also, you know, waterfall enrichment?

1407
04:28:19.940 –> 04:28:29.949
Kevin White: on the back end of all that. So not only do you know that it’s this person, but you have all the details you need to contact them. Create segments, build different like micro campaigns and stuff like that.

1408
04:28:30.170 –> 04:28:48.820
Kevin White: And so if you combine these 2 things, you can then, you know, activate, based on. You know what someone’s done what signals an account is showing you and help reps really prioritize their their outbound motion. But then also help marketing help all these other things. So the last part here is like kind of putting these 2 things together and taking action on them.

1409
04:28:48.820 –> 04:29:15.490
Kevin White: And so if you think about it like, this is a nice like foundational layer for agents to like, go do their thing on top of all this data. And so that’s kind of like where what? I’m going to be more demoing today versus like the the infrastructure of how to make all this happen. So it’s kind of like how we can use agents to go capture signal how we can use agents to orchestrate different pipeline plays and service up the best leads for reps.

1410
04:29:15.490 –> 04:29:33.529
Kevin White: and then how those reps can do things like cut account research time. By, you know orders of magnitude or send I hate to use the term aisd, but send like a personalized, outbound message using AI to generate that message powered by all these signals. So that’s kind of like what I’m going to demo today.

1411
04:29:33.790 –> 04:29:44.299
Kevin White: And might as well just jump into it. Okay. So I kind of have this checklist. I don’t know if people can see it. I have this checklist that. I created a notion of, like all the different things

1412
04:29:44.300 –> 04:30:06.018
Kevin White: that I wanted to go through to to keep me in check. So let’s hopefully, hopefully, I can. Hopefully, the demo AI gods, or whatever with me, Kevin. Intelligence is with me and I can. I can or do these things. Without any too many hiccups. So let’s start with the 1st one which is doing account research. And so like, this is a

04:30:06.320 –> 04:30:32.310
Kevin White: a demo of what demo environment, of what common room looks like. You can see. We have, you know, a pretty healthy amount of accounts in here that might be in a book of business, or whatever. And so just by having these accounts and loading them into common room. We’re gonna do a lot of things to identify like what’s happening at this account. At the account level, but then also, like the contacts within that account. The one thing that I wanted to show here that’s really helpful is like

1414
04:30:33.032 –> 04:30:48.200
Kevin White: account. Research used to be this thing that was like very tedious and mundane, and like having to go sift through all these different things and Google searches, or whatever to like, find out, like what you should know about this account. That is like a really good job for an AI agent to go do for you.

1415
04:30:48.310 –> 04:30:52.260
Kevin White: And so you can see here we’ve already have our agent that has gone and crawled

1416
04:30:52.578 –> 04:31:03.250
Kevin White: vercel, and is returned back these different, like buckets of research that my team might care about. And we we do this out of the box. But then it’s also super customizable and flexible to your own

1417
04:31:03.250 –> 04:31:28.089
Kevin White: business model and go to market motion so you can see here. You know we’ve we’ve pulled out certain things about every account. If I go into any other account, it’ll kind of show this step, too. And if I go into detail, you know, we can see like, Oh, this is Vercell’s business model. Maybe I should like for a rep. I want to know if they have a freemium product or something, because that applies to like how I would go outbound to that account. So this is cool. But then the kind of fun magic

1418
04:31:28.090 –> 04:31:29.856
Kevin White: demo part of it is

1419
04:31:30.360 –> 04:31:53.309
Kevin White: is to like, just do this from a fresh account. So if I find something here, I’m actually going to go to the next page, I think there’s better one better examples here that haven’t been done. And you can see like these, some of these have these like sparkles. That means that we’ve done the account research there. So I’m going to pick one. Hopefully, this is not a hopefully, this is a good example. Actually, square is probably a better example and pick pick square. It’s better to pick public companies because it does look at like earnings calls, and stuff like that.

1420
04:31:53.600 –> 04:32:05.629
Kevin White: And if I click on overview here, you can see that we don’t have this AI research. Essentially, our AI agent is going out and fetching that signal on square, and then it’s going to return it back to us, and hopefully.

1421
04:32:05.873 –> 04:32:29.049
Kevin White: this 60 seconds. Hopefully, no one’s keeping a time watch for me, but it should. It should show up very soon here. But be while it’s doing that, and while the agent is doing its thing, I’m gonna go in the background and show you kind of like what’s happening behind the scenes here. So if I go to settings I can click on this roomy AI. This is our agent settings and look at capture. And so so anything that’s turned on here is kind of what’s showing in that other view.

1422
04:32:29.050 –> 04:32:40.769
Kevin White: So if I look at the business model essentially like this is just a prompt that is totally customizable. And you know we have a whole prompt library that’s helpful for people to use. If you, if you want to look on common.

1423
04:32:41.060 –> 04:33:06.667
Kevin White: Maybe I can link that out at some point. Essentially, we just have, like a very simple, prompt, and what’s happening at each of the these, for each of these accounts is, our AI agent is going out doing real time, search and reasoning, and then pulling back a answer to this, so that you know, and and returning it to that vercel, or whatever account, forget what the one I just looked at was square square. So if I go back maybe we’ll see. Okay, it’s not there yet.

1424
04:33:06.939 –> 04:33:17.159
Kevin White: but you will see something like this that’s returned on each each account once it’s gone and done all that kind of processing. Maybe. There’s too many prompts happening here

1425
04:33:17.369 –> 04:33:40.139
Kevin White: now. The other thing I want to show, and I’ll maybe come back and show you the magic in a sec is that not only do these things come out of the box but you can modify and create new ones. And so if I go back to our roomie AI research agent here. I can go and look at like there’s 1 prompt for primary competitors.

1426
04:33:40.619 –> 04:33:57.789
Kevin White: You can see here. This is just same similar prompt. It’s like, Go find this primary competitors of this organization. But one thing I might want to add to that is and build upon, that is like, find me negative reviews of that primary competitor. So I have. I have another prompt here that’s not yet turned on.

1427
04:33:57.790 –> 04:34:14.669
Kevin White: That is saying, okay, we’ve already gone and researched this one bit about this account. We’ve we found who their primary competitors are. Let’s build upon that signal that we’ve captured. And this is kind of like where it becomes more agentic when you’re Daisy chaining these things together as like, let’s go find their primary competitors and return

1428
04:34:14.669 –> 04:34:29.445
Kevin White: and then go find negative reviews about those primary competitors. And maybe we want to use that as like a hook in an email. And I’m gonna tie this all together in the actual like actioning Demo, too. So that’s what this prompt is doing. And so if I wanted to I could just simply

1429
04:34:30.230 –> 04:34:36.390
Kevin White: turn this prompt on on the last screen. It was not on so negative

1430
04:34:36.490 –> 04:34:47.460
Kevin White: competitive views. I could turn this on. I’m not going to do it for this, but I do have it in another instance, and then real quick. Before I do that, let’s go back and see if a square is populated yet.

1431
04:34:47.460 –> 04:35:10.329
Kevin White: if not oh, there we go! Great! Yes, so we have the all of that research signal now pulled into the square account. So like maybe it was a little longer than 60 seconds, but pretty pretty fast, much faster than a human going and doing it. So I’m going to say that one is done, and then also this next one is capturing bespoke enrichment signal. That was the primary competitor thing I showed. So great, let’s check this.

1432
04:35:10.330 –> 04:35:23.360
Kevin White: Okay? And then I’m gonna close. I’m gonna keep that open. It’s fine. Okay, so this next instance I’ve had. I had now have both of those fields, the primary competitors and the negative reviews already populated for all of many of these accounts in here.

1433
04:35:23.619 –> 04:35:50.695
Kevin White: And so let’s say, I want to build a micro campaign, a micro segment for a certain competitor, and, like my team, wants to use those. Anyone who is competing against that competitor like use their negative reviews as a hook to like, get them to, you know. Read the email that I’m sending out so hopefully, that makes sense. What? What? One thing that’s really powerful with common room is like now that we’re pulling in that signal from. If I pick on that last name here.

1434
04:35:51.320 –> 04:36:14.319
Kevin White: We can now filter by any of these other. Any of this data that’s returned. So I can build micro campaigns and segments based off of this information. It’s like it’s like finding, going and finding and enriching signals, I guess. And so I wanna so say, I want to find anyone who’s competing with intercom. So I can click on, add filter here, and I can type in primary competitors.

1435
04:36:15.364 –> 04:36:21.209
Kevin White: And then this will allow me to just do a open text field and say, intercom here.

1436
04:36:21.369 –> 04:36:45.590
Kevin White: And this is, gonna give me 109 organizations that have intercom listed as a primary competitor. And so if I click on one of these like gorgeous. Hopefully, I actually haven’t looked at this before so hopefully. It’s accurate. But if I look at competitors, our primary competitors oh, here it’s down here. You can see that these are listed intercoms listed here, and it gives sources so great that was accurate.

1437
04:36:45.912 –> 04:37:12.049
Kevin White: And then we also have we also have this primary competitors. Sorry. Negative competitor reviews. You can see here that it’s like pulling different reviews from different places and giving also sources there. So let’s pick another account here. Maybe I wanna target. And I know someone a customer. So I’m purposefully picking this and so I want to reach out to somewhat customer using that negative competitor review as a hook.

1438
04:37:12.712 –> 04:37:41.010
Kevin White: So what I can do is I can select Sam here I can click into Sam, and this will give me all of the signals and information that we have about Sam and I can have a common rooms. AI agent. We call it roomy AI draft a message. And I call this one the competitor, Nick. So if I click on this. It’s gonna pull up and generate an outbound message that I can copy, paste, modify, do whatever I want, and send that to Sam. And you can see here. It’s kind of cool. It’s like pulling it that, you know, intercom is expensive. This is a quote

1439
04:37:41.009 –> 04:37:50.270
Kevin White: from that intercom someone talking poorly about intercom. And we can say, like, Hey, like this, is someone saying something bad about intercom?

1440
04:37:50.290 –> 04:38:08.010
Kevin White: Yeah, you know, you’re a competitor to them. And you have this competitive advantage of self serve motion. Here’s how you can use. I’m using common room in this case. Here, you can use common room to like exploit that negative sentiment from that competitor. So it’s like, kind of stitching together all these different things to create, this really bespoke email. And if I do it for someone else.

1441
04:38:08.336 –> 04:38:23.350
Kevin White: Let’s go back here, and click on Rick. Maybe it’ll be a separate message and let me look at this this one. So it’ll generate that message specifically for Rick. And you can kind of see it’s in the same format here.

1442
04:38:23.650 –> 04:38:41.479
Kevin White: And so to show you what that looks like behind the scenes, and I only have a couple of minutes left here. I’m gonna go into settings, and this is on. Now. Our agentic like, go find and generate an Aisdr type of message. For this person. If I look at this, I can look at Competitor Competitor Nig.

1443
04:38:41.700 –> 04:39:02.819
Kevin White: I think was the name of it. Yes, edit, and you can see here that we’re pulling in those signals. From that we went out and fetched from roomie AI. So the negative competitor reviews are here. We have their business model. Here we have risk analysis here. These are all those custom agentic signals that we’re pulling in at the account level and using that to inform our outbound email. So like, really, really cool stuff here

1444
04:39:03.379 –> 04:39:30.219
Kevin White: and then let me go back to this and make sure. So we built a micro campaign. This is that segment of accounts using intercom, using AI to filter by that segment and then identify our best bet prospects. That was the people who I actually should show this. If I look at gorgeous, for example. You can see their contact score here, and these are like the best people to reach out to, because they’ve had these certain signals. So it’s kind of like surfacing up the best contacts. Reach out these accounts.

1445
04:39:30.620 –> 04:39:38.819
Kevin White: And then we generate an outbound contextual message. Great check and then the last thing I wanted to show

1446
04:39:39.330 –> 04:39:58.899
Kevin White: is how you can. Just if you say you’re doing this and running, you’re you’re you’re finding that like you’re going through each contact, and you’re clicking on different contacts, and you’re just adding them to an outreach sequence. And you’re doing them all kind of like on autopilot. It doesn’t make sense for a rep to just like, go through and like, do this one by one, although you can do it kind of in bulk like this.

1447
04:39:58.900 –> 04:40:12.089
Kevin White: But let’s say you want to automate that whole end to end process. So for that, we have these like really powerful workflows, where I can just say, create a new workflow. I can say, run this for a set of contacts as I’m gonna be emailing contacts or putting contacts through a sequence.

1448
04:40:12.545 –> 04:40:25.190
Kevin White: I’m gonna add that filter for primary primary competitor here, I mean, let me zoom in a little. So people can see add a filter primary competitor contains. Oops.

1449
04:40:25.410 –> 04:40:31.290
Kevin White: Hold on clicking around funkily here, intercom 1 min left.

1450
04:40:31.440 –> 04:40:47.802
Kevin White: and then I’ll add a filter for like making sure that they’re, you know, an economic buyer, director, level, title, or above, and then the last step here is I can add them to a you know, whatever sequence. But I’m going to pick outreach sequence here, and then I can select that same

1451
04:40:48.130 –> 04:41:06.375
Kevin White: demo, or whatever sequence. Select my mailbox and then save this, and then anytime someone qualifies for that, they’ll be automatically put into that sequence. So you can really do like a crawl walk, run, approach, using this method. So I think I got everything checked off, and then the last thing I just wanted to say is,

1452
04:41:06.990 –> 04:41:32.220
Kevin White: you know, a plug for common room. You can try this out for free we for anyone viewing this, I’m offering a 30 day trial up from 14. So that’s great. And then a 5,000 credits, if of using like AI what’s it called tokens. So and then also, I talk about this stuff all the time on Linkedin. Sometimes I have an AI agent do that for me. I call it Kev intelligence and check me out on Linkedin. So that’s it.

1453
04:41:33.520 –> 04:41:34.160
Julia Nimchinski: Wow!

1454
04:41:34.390 –> 04:41:35.300
Kevin White: Perfect timing.

1455
04:41:35.730 –> 04:41:41.190
Julia Nimchinski: Amazing timing. I’m not sure that this is like, actually not. It’s some kind of automation, Kevin.

1456
04:41:42.710 –> 04:41:45.480
Julia Nimchinski: Oh, it’s a pleasure to see you all right. Thank you.