So What? How To Use AI as a Sales Copilot

So What? Marketing Analytics and Insights Live

airs every Thursday at 1 pm EST.

You can watch on YouTube Live. Be sure to subscribe and follow so you never miss an episode!

In this episode of So What? The Trust Insights weekly livestream, you’ll learn how to use AI as your sales copilot to boost efficiency and close more deals. You’ll discover how to generate an AI-powered sales dashboard, prioritize leads, and identify next best actions. You’ll also learn how to automate these actions in your CRM for a seamless sales copilot experience, freeing you to focus on building relationships and closing deals.

Watch the video here:

Can’t see anything? Watch it on YouTube here.

In this episode you’ll learn:

• How to convert a sales playbook into an AI coach
• How to use AI coaching for sales post mortems
• What a live AI sales copilot could look like

Transcript:

What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.

Here’s the edited transcript:

Christopher Penn — 00:00
Happy Thursday, everyone. This is the “So What” Marketing Analytics and Insights live show. It’s with me and John. Katie is at, what is it? The Marketing AI Institute AI for B2B Marketers Summit, where she is delivering her keynote talk on becoming AI-ready. So, which if you… I don’t know if you’ll be able to catch her recording of it or not, but if you want to catch what she’s talking about, it is in our new AI Readiness Marketing Strategy Kit. Go to Trust Insights AI Kit to get it for free. This week is part two of our talk on AI and Sales. We’re going to talk about how to use AI as a sales copilot. Nothing to do with Microsoft Copilot. I should that, but it is a popular term.

Christopher Penn — 01:17
So John, as a sales leader, what things would you like AI to be able to tell you about the condition of the sales pipeline, for example?

John Wall — 01:30
Yeah. The greatest thing there would be automated processes that map our existing procedures where we want to be and where we want to do. Because one of the biggest challenges we have are leads and prospects where we engage at some point, and then for some reason, it fizzles and it sits out there, and to have a backstop where, after two weeks, it goes back and says, “Okay, what’s going on, you know, with this lead and where are things going?” And also, to qualify and score and keep track of just raw leads. Like, just talking about the paper we dropped, it’s gone berserk. We’ve had hundreds of leads come in the past two days. And so, we do have some systems in place where those get prioritized. Like, HubSpot’s done a decent job with categorization.

John Wall — 02:12
I get notice of the ones I’ve already engaged with, so I’m able to cherry pick. But doing more on that front. Yeah, I’m just really excited for what you’ve got because all this stuff—we’ve seen people getting 100X the effort. You know, being able to automate stuff that was normally a pain and now have that just happen is huge. So, yeah, I’m looking forward to this.

Christopher Penn — 02:34
Okay, so let’s start by first saying this. You need to have done the sales playbook. If you have not done this part, today’s episode will not work for you. Like, nothing we do today can be done without this. So, if you missed last week’s episode, it’s on our YouTube channel. Go to Trust Insights AI YouTube. You must have the consolidated sales playbook because in here are things like ideal customer profiles, the pitches that, and the processes that you have in your sales process for each stage of your pipeline and things. Again, if you don’t have that, you can’t do anything.

Christopher Penn — 03:09
So, I think the first place I would want to start, and we know that in general CRM software is fairly bloated, the reporting is challenging, and, or, you know, some systems are like, “Oh yeah, insert additional coins for this or that,” or the other thing. So, we’re going to start by saying, let’s just start with a basic snapshot. What we want to think about here with Copilot is, we want to have a copilot be something that flies along with us, that from time to time you can maybe give it a little bit of control. But you’re still the pilot. I’m going to load in some JSON data. So here’s where this data came from. I took our actual HubSpot data from our actual HubSpot CRM, exported it as a CSV spreadsheet.

Christopher Penn — 03:57
Now, we’ve talked in the past, you cannot give spreadsheets to generative AI. It’s just a terrible idea. Never do that. What you can do is you can hand generative AI a spreadsheet that has been converted into what’s called JSON JavaScript object format. And what I’ve done is I’ve made a synthetic version of our HubSpot pipeline that has none of the actual person identifying information or the people who are actually working the sales, because that’s just basic data privacy. But what you can see in this file is things like it is our solutions, it is our fake conversations. So, you know, John, Chris, and Katie are in here, and we have all these fake companies with fake stakeholders, but modeled on our very real HubSpot data.

Christopher Penn — 04:42
So I’m going to start by putting the JSON file of our HubSpot real data in, and we’re going to give this a prompt to say, “Turn this into a sales dashboard”. Right? Let’s just get a look at our pipeline. So the prompt is, “Using HTML, CSS, Tailwind, create a summary infographic of our current pipeline based on the JSON import so sales leader John Wall and CEO Katie Robert can instantly see the status of our sales pipeline.” So, I’m going to add that, and I’m also, of course, going to add in the sales playbook because that has all the things like what each stage of our deals means. If I don’t put this in, then this is an exercise in futility so. And I’ll also give some brands, brand style guidelines just so it makes something that isn’t completely ugly.

Christopher Penn — 05:28
So I’m going to go ahead and run this, and what it’s going to do is it’s going to generate raw HTML. Now, raw HTML is not going to be nice to look at. However, we can copy and paste that, and then just load that in a browser window. Why we would do this is again because sometimes what’s built into sales products, CRM, Salesforce, I’m looking at you, are the most convoluted, complicated, overthinking dashboards. When really someone like you or Katie just like, “What’s going on?” Like, that’s all I need to know, what’s going on?

John Wall — 06:05
Yeah, this is it. To be able to get a picture of where you’re at, you don’t need 35-page dashboards.

Christopher Penn — 06:12
Well, that’s the thing, is I have, when I log into HubSpot, you go to the dashboards and reports page, and it’s just this wall of dials, and it looks like it reminds me of what they do in Star Trek or in Doctor Who where they just add dials just to make it look complicated.

John Wall — 06:30
Right, yeah, well. And yeah, we could go on forever about, you know, dashboards and how that works as far as upper-level management versus people actually making change.

Christopher Penn — 06:40
Exactly. So, let’s see how it’s doing here. Oh, that’s interesting. It’s actually starting to put in the deal data itself. So we can see the sales pipeline snapshot open, total open deals, total active deals pipeline by stage. But I see it’s actually putting in summary-level data that we might be able to manipulate, which I did not ask for. But you know what, that’s pretty cool. I’ll take it. While it’s doing this, John, what are the things that you currently have on your sales dashboard that, and how often do you look at the dashboards that are built into HubSpot?

John Wall — 07:17
Yeah, it’s a pretty simple process for us because we have four checkpoints and then closed and loss. So, it’s, I mean, it gets used every other day or better, because I just go in there and say, “Okay, what’s on the to-do list? You know, who do we need to check in with, where are they at?” And it does have the HubSpot deal record, has all the key points as far as amount and stage and the referrals, you know, where the stuff came from, source codes. But after that, it’s really not that complicated. You know, it’s where a small enough org that once I know what’s going on with the 15 or 20 things that are live, that’s about it.

John Wall — 07:56
You know, it’s when you get, once you have that second sales rep, like, that’s when it really starts to kind of get crazier because you’ve got records that you know the other person doesn’t even know what they are. So now you need to start having more rigid process and be able to figure out what’s going on. But yeah, for right now, it’s just, you know, it does work in that it’s, “Here’s the picture of everything that’s going on and where it’s at.”

Christopher Penn — 08:17
Right. So, this is what it snapped together. It said “sales pipeline snapshot. Total open pipeline value $587,000, 40 active deals, 393 in proposal, 193 in opportunity, 18 newer unqualified leads, leaders action items, needs attention, deals with no activities in 30 days.” We have Health Bridge Pharma, Global Edu Connect, and Pixel Spark Digital Health Bridge Farm in particular, no activity on that, which is kind of bad. And then, high value deals: Here is Momentum Marketing Group, Cyber Secure, Ascendia Health. Obviously, like I said, these are not real companies. And then what’s supposed to be closing this month? August 2025. So I will say this is a lot easier on the eyes than the hideously bright orange everything in HubSpot. If you had this dashboard to look at, John, what if anything would this help you with?

John Wall — 09:15
Yeah, well, having the “needs attention,” that’s great. The stuff that is stalled, there’s been a violation of process. So something needs to get done with those—is great. The only thing that’s missing from there that I can see so far is having them broken down by stage because the ones where we have scopes already out and awaiting signature review, those are more important than the ones that, you know, they read a white paper and we’ve got a call on the docket for next week. You know, those can kind of sit and simmer. Whereas we do have the high-priority ones that, you know, we’re continually babysitting.

Christopher Penn — 09:53
Okay, I think that’s a really good idea. So, let’s take that feedback, and let’s say, “Okay, figure out next best action. We’ve got our sales playbook, we’ve got our pipeline. Let’s figure out what the next best action is and have the tool attempt to.” Let’s list them in outline format by deal name, opportunity stage, and next best action. Explain each next best action, and we’ll have it do it in HTML and Tailwind so that it’s not just console output. But one of the things that I often find, and I remember this from my sales days, is trying to figure out like, “What am I supposed to do today, or what’s the next step with this deal?” Like I, I have no idea. And again, this is going back to our sales playbook from last week where you have the sales process mapped out.

Christopher Penn — 10:47
Here’s the prospecting, qualification, discovery, and things key metrics, and then you have the sales tactics. So, at each stage of the pipeline, here are sales tactics. So if you have that in your sales playbook, then you can come up with, “A, here’s a recommended next best action for each of these things.” Each of the deals that are in the pipeline. How do you, how do you as a human meet, manage next best action?

John Wall — 11:13
John, that I just use the force with that. You know, I didn’t go through. We go down the list and yeah, there’s notes in HubSpot. So, we do have checkpoints you, and we use Google Drive for a lot of stuff because for us, it’s pretty simple. You know, we do initial screening call just to prove that they’re for real, and they have budget, and that we may be able to work together. The second stage is to bring in your Katie for technical and review. As far as, okay, you know, we realize a project could happen, but is it do we feel we can be successful once we’ve got the whole team on, and then we actually write up a formal document, and that’s kind of the classic sales chasing the signature and making sure we get through their procurement process and all that stuff.

John Wall — 11:53
So it’s not super complicated as far as what the steps are. So, as long as I can see what stage they’re at, I know what we’re supposed to be doing next. But yeah, otherwise, it’s kind of clunky, you know, because it’s, I’m relying on email and the HubSpot notes field to give me a specific point as far as, like, “Okay, here’s where we are today, and what we need to get done tomorrow.”

Christopher Penn — 12:12
Have you worked with any CRM software that builds that, has a built-in next best action to as a copilot to tell a person, “Hey, here’s what you should be doing?”

John Wall — 12:23
No, I, you know, I’ve worked with simpler systems that were just kind of like, “Okay, the next checkpoint is X.” But yeah, I’ve never seen anything that was smart enough to actually, you know, kind of say like, “Okay, you should be sending this email, or you know, giving you really more specific and accurate next steps.”

Christopher Penn — 12:40
Okay, well, let’s see what it came up with. So, it says “next best actions” in the sales pipeline. Pixel Spark Digital AI Content Lead Gen Solutions. Next best action: Execute gentle nudge. Value reinforcement play from the sales play. We’re going to make this a little bit bigger so we can all see what we’re doing, looking at here. The deal has stalled. Send a polite follow-up email that provides new value, link to a new, relevant blog post. Inquires about their decision-making timeline. Playbook rationale. Re-engages the prospect without being pushy. Keeps trust inside top of mind. Help uncover hidden objections as in playbook play 6.4.3. Global EduConnect. Next best action: Proactively address legal review. Schedule brief call with their primary contact. See if they need clarification for their legal team. Often have Katie join if they need strategic stuff.

Christopher Penn — 13:24
Fintech innovators. Next best action: Confirm next steps in timeline. Follow up emails the primary contact. So how does if you imagine if this was our actual pipeline, John, as a sales leader or as a sales lead, does this give you a sense of what you would want to do?

John Wall — 13:46
Yeah, it gives you a clear approach on to like, “Okay, what’s the next lever to pull and where to go?” And the huge value I see with this is when you get to the point where you’ve got too many leads going. You could have a virtual assistant that could scrape through this and do the, you know, kick off the tracks that don’t require any extra effort. You know, take all the easy work, the less challenging work off the plate.

Christopher Penn — 14:10
Why? Why a virtual assistant?

John Wall — 14:14
Yeah, no, it could definitely be an AI agent. I mean, automating it would be even better. But yeah, as of today, BA would be how we would handle that.

Christopher Penn — 14:22
No, I would make it.

John Wall — 14:25
Yeah, I’m all on board. I’m in.

Christopher Penn — 14:28
One of the things though that I see also useful here is because the next best action is pretty well canned. There actually is a field, a default field in HubSpot called Next Step. But the software doesn’t populate with anything. It just says “next step” and expects you, the human, to fill it in. I could easily see, since on the back end, in the JSON file we have the deal IDs having a tool like N8N, literally just go into HubSpot, say, “Okay, I’ve got my next best action recommendation from the LLM. Let’s put it in the records.” Like, “Let’s put it in the actual next step field within the record.”

John Wall — 15:07
Yeah, yeah. And because then you can even have a dashboard run that so you can see them all, you can see by next task what they’ve got.

Christopher Penn — 15:16
We could also do goodness of fit, which is something that is very difficult for us to figure out. So let me go back to our pipeline analysis, and here I’m going to say, “Let’s do some analysis and diagnosis of our pipeline. Match each opportunity against the ICPs, return the deals and companies the closest match and the goodness of fitness on a scale of zero to 100 as to whether the company deal fits the against fits one of the seven ICPs, return your results in HTML, CSS, Tailwind.” So, this is now sort of going a level deeper to look at. If you’re, if you were a sales leader, you’re looking at this and going, “We got $4,390,000 in proposal, and $190,000 opportunity and 18 new leads.”

Christopher Penn — 16:17
You might say, “Gosh, we’re, we know we have like you said, a thousand new prospects in our database from our LinkedIn paper and like 800 new prospects for the new AI readiness kit. Why isn’t this bigger? Like, why is this, this. Why is, why is the pipeline so bad? Or was it from Glenn Gary? Glenn Ross? The leads are weak. The leads are weak. We might want to run a diagnosis.” So, let’s call this diagnosis and let’s take our sales records as it’s generating and say, “How good a fit is our pipeline?” Because we, last week when we did this, we have our ideal customer profiles, we have all these, and then we have the buyer Personas in here. So, we know who our ideal customers are, and we could measure what’s in a pipeline. Go. You know what.

Christopher Penn — 17:17
Part of the reason the pipeline might be stalled is that the companies that are in the pipeline, just they’re not the right customers. They’re not going to do the job for us. So let’s paste this in and look at our pipeline diagnosis and we go down here, like, “Alpha Investments”. Oh, that’s niche solution Seeker. But it scores only a 35. So it’s like, yeah, that’s not great, right? We have a lot of, we have a lot of ICPs, ones, two, we have some, a lot of fives. And so we might even rejigger this to say, like, you know what, those ICP5s, the niche solution seeker, they don’t have money. I Ignite Fitness Studios. You know, I, I know most fitness studio chains, they do not have great margins.

John Wall — 18:04
Right. They’re not looking to spend, and so that’s the stoplight indicator is against just ICP, ICP match.

Christopher Penn — 18:11
Yeah, I guess a fit of ICP. So we could equally do this again for things like budget, for things like deal age. Just getting a, a green, yellow, red of, like, how good a fit is this company or any given companies based on our criteria.

John Wall — 18:28
Yeah, that’s a fantastic way to screen.

Christopher Penn — 18:31
How else would you want to look at our pipeline?

John Wall — 18:37
You know, other slices are. If you can get any commonality. That’s something that we’re always kind of wondering about. You know, is there anything as far as the types of orgs in there, can it tell us who else we should be talking to in the database based on the folks we’ve been able to get into the process?

Christopher Penn — 18:55
Well, we covered that a lot last week with the ICPs themselves. Like, these are all the types of companies that we work with. Are you saying that you might have stuff that are laterals outside of our ideal customer profiles?

John Wall — 19:07
Yeah, and really that idea of, you know, the best proof are the folks that have actually stepped up and have budget and are going to go to deals. You know, I mean, that’s even better than just the database and the house list. Yeah, that’s really more deal analysis and. Yeah, I hadn’t thought of that. That opens up a whole door too, of like just looking at the wins, you know, what kind of stuff can you get from the wins? Because now you’re not, you know, you’ve got all the stuff in the raw pipeline that there’s a bunch of it that never makes it to close.

Christopher Penn — 19:38
Okay, well, let’s ask it this. Using our existing ICPs on the sales pipeline, what other kinds of companies, who would be looking to market to, who might be most lucrative? I mean, I think that’s a reasonable question to ask. And again, I can’t emphasize this enough. I’m going to beat everyone over the head of this. You have to have all this data predefined so that the AI knows what is exists already. If you attempt to do this without it, you are in hallucination territory that looks a lot like Keith Richards.

John Wall — 20:09
Yeah, completely. No idea what’s going to come out and don’t. I don’t think you can just jump on it.

Christopher Penn — 20:15
Exactly. Let’s see, here’s a report. Oh, it’s decided. This is going to make me report. So, it apparently thinks I want web pages for everything. Now it’s just going to start generating because I’ll spin up a new ops HTML file here so I have a place to put it. Let’s see. It’s really thinking through this sort of thing. It’s impressive. Oh, it’s going by the different opportunity categories. That’s interesting. All right, let’s take this whole thing because it partially mangled the response, and let’s put it into on screen here. So, it said, “Hey. Our current pipeline reveals four highly lucrative adjacent market opportunities. They’re not entirely new ICPs, but rather high potential specializations.” Prioritizing these segments will allow us to generate to leverage our proven strengths. The regulatory specialist, the agency enablement partner, the old economy digital transformer, and the enterprise AI strategist.

Christopher Penn — 21:26
And let’s make this a little bit bigger here so we can see what we’re doing. The regulated industry specialist companies operating highly regulated environments. Supporting evidence from the pipeline. These are the deals. Why it’s lucrative. High cost of failure, complex problems. Budget availability. The agency enabled partner. You have all these agencies. Why it’s lucrative. Force multiplier. One agency partner can become a channel for dozens of end client projects. They find the clients, we provide the expertise. I really like that. High lifetime value, scalable offerings, Marketing on sale. Double down on the trust in our name. Create targeted content on ethical AI and healthcare AI, financial compliance. Well, Katie’s gonna be doing that. Develop a formal trusted sense partner program. Offer referral fees. The old economy digital transformer manufacturing, B2B industrials, WIDEs, lucrative operational budgets, greenfield opportunity and tangible ROI. And the enterprise strategist.

Christopher Penn — 22:28
Engage clients, solves problems owned by the C suite. High strategic value, full stack trust insights value. So there’s your pipeline adjacent stuff, John.

John Wall — 22:39
Yeah, that’s really interesting. I’m surprised that the AI agency1 comes in lower, but that does make sense. I guess as you dig in deeper to the deals that those profiles are the ones that tend to get all the way to close, you know, makes sense.

Christopher Penn — 23:00
Okay, so next, what do you want to know as a sales leader? We’ve now gone through, we have better reporting. Right. So that, that automatically helps. We have next best actions for our entire pipeline of what we should do. We have pipeline ICP fit to see what kind of companies in our pipeline and how well they fit our ICPs. And now we have a go hunt, go fishing for a new business.

John Wall — 23:28
Yeah, you know, it’d be cool to. Can we run a backstop, run something, you know, look at the losses that are more than six months old or three months old. And show me the ones that based on what’s closed, that should have closed, you know, get.

Christopher Penn — 23:40
I don’t think I generated that in the fake data.

John Wall — 23:42
Oh yeah, there’s no. The closed stack is not over there.

Christopher Penn — 23:45
Yeah, I think I only have. I’m like, let me just double-check that to make sure I’m not lying. But I’m. I think I only faked.

John Wall — 23:54
Yeah. Fake the live stuff.

Christopher Penn — 23:56
Yeah. There’s no closed one or closed lost in here, unfortunately.

John Wall — 23:59
Okay. Yeah, because that would be one thing is, you know, re-engagement because, like, there’s tons of clients out there where went through the whole process and then suddenly the champion got laid off. Or they were like, “Oh, we don’t have the budget for this quarter.” So to be able to find stuff that could be ripe now, you know, that we had missed through the past run would be good.

Christopher Penn — 24:21
Oh, it’s like re-engagement. Like, win them back.

John Wall — 24:24
Yeah, totally. Oh yeah. I guess with wins, you could do that too.

Christopher Penn — 24:28
Upsells.

John Wall — 24:29
Yeah, upsell, right. We don’t know what’s upsold versus, like, who bought a six-month project, and then didn’t come back for more stuff. But yeah, so then that creates another interesting question. Lifetime value, you know, of. Of what’s in there. Who, who are the best over time because we lose sight of that. You know, we kind of. We have some clients, we think we’re doing great because they’re doing all these tiny projects, but then we have other ones that, you know, they kind of come in once every other year with a giant project. Like, actually, who are those are the. The best people to be working with. Like, of the customers we have, who are the ones we should be trying to.

Christopher Penn — 25:06
Okay, well, I’m having it generate 10 closed one and lost files. Say, great. From those one and lost. Let’s build a backstop report. What’s in the backstop report?

John Wall — 25:22
Yeah, backstop. It’s always stuff that got dropped or left behind. And can we go back and bring something back to life? Is there something that again got dropped? Because as a sales manager, that’s always the biggest one is the, “Hey, we gave you this list,” and they said, “Oh, well, I called everybody, nobody called me back.” But is that the case? And is there anything that they missed in that?

Christopher Penn — 25:47
Okay, build your report in HTML with this and Tailwind. So, since it made the lovely files, let’s see if it can create a backstop report on who we lost. I like the idea of this because there’s. There is always. But, like, we always say, “Is they always come back.” Right?

John Wall — 26:17
Yeah, yeah, even. Oh, and I thought about that. I don’t know if you have contact info, but it jumps company to company. Right. Because ultimately we’re selling to people, and there’s tons of people who love what we do. They see the value, but it’s just not the right time at their org for whatever reason. Like, they’re not in enough of a position of power to make a decision, or they don’t have the budget this time, or you know, there’s a million reasons why this is not the time that can be done. But the person is a champion, is on board and wants to do something. We have plenty of folks, you know, we have the “no budget crew” that we make fun of who, you know, every time they move jobs, they come back and want to get something rolling, and they never can.

John Wall — 26:57
But you know, even those people are champions. And so yeah, if we can find deals where, you know, we’ve made some inroads in the past, and now we can go back and see if it’s finally time. Yeah. Because they always come back.

Christopher Penn — 27:09
Okay, well, let’s take this backstop report first and see how that looks. We’ll put it into the window here. So we have backstop report to analyze every loss. You’ll understand the root cause loss and provide a playbook strategy for in potential recovery data well corp, predictive modeling pilot lost on price, budget, future prevention, and recovery plan, the selected vendors, less experienced, set of 4-month follow-up task, re-engage with The Gentle Nudge Play 6.4.3, offering new relevant thought leadership piece, be prepared to be the first call when cheaper solution under delivers, that happens a lot. Someone goes for the cheap solution and then six months later they’re like, “So the cheap solution didn’t work.” And then we’re like, “Yes, the expensive solution is still available, however, unfortunately, it’s more expensive.”

John Wall — 27:55
Right? Exactly.

Christopher Penn — 27:57
Innovest internal champion left, right. So during discovery, actively map out and build relationships as noted says six months follow up, execute the content-driven ICP outreach play 6.1.4 to the new leadership, the goal is to re-establish contact and re-eval it the original need treating as a new warm lead, first chapter books lost, internal DIY effort. Yep, that happens a lot. Internal IT projects face delays or fail to be business needs. Set a three-month follow-up check in, ask how the internal project is progressing, and offer to progress. Provide a second opinion or to help them get unstuck.

John Wall — 28:30
Yeah, those—it’s funny, those are always an interesting role of dice because you still never know who’s vested in the internal project when you come knocking. You know, there’s some people that it’s their intention to bury you when you come asking offering a second opinion, because that’s the last thing they want to see. But hopefully, you know, sooner or later they get to the bottom of the bucket and can’t make up excuses, and somebody will come in who wants to make things right and be effective.

Christopher Penn — 28:55
Exactly. New CMO comes in, it’s like, “What are you guys doing?”

John Wall — 28:59
Why have you been working 10 months on something that you know nothing’s gone out the door?

Christopher Penn — 29:03
Yep. Momentum Dynamics Enterprise Attribution Modeling lost to entrenched incumbent large firm. Yep. Long-term nurture. The large firm will likely deliver a generic solution, periodically send the client contact our cutting edge thought leadership on specifically attribution modeling, position ourselves as the true specialists they can turn to when the generic solutions are good enough. And then finally, Global Logistics Corporation Process optimization project pause due to external factors, the reason for loss is not rejection of value. Add the contact to long-term nurture sequence in Maudex at a firm. CR test CRM task to re-engage in late Q4 2025 ahead of their stated Q1 2026 restart the outreach. Should reference the prior discussion, ask they’re beginning to plan for the new year. There are a couple of leads in our pipeline and I know who they are, that this is the exact description.

Christopher Penn — 29:47
It’s just that the gender of AI just changed all the names. Right. That I can say, like, immediately after this live stream, I need to go email them, say, “So your fiscal year’s starting in a month. What do you want to do?”

John Wall — 30:00
Right? Yeah, yeah. The folks that said, “Hey, we don’t have money this, you know, this year, but come back and talk next year.”

Christopher Penn — 30:06
Exactly. So, the backstop report again, very cool, very straightforward, and I want to emphasize this is not, like, super high-tech. This is one spreadsheet file from the CRM that has the notes data, as well as all the other company data, converted into a JSON file that we have been just asking stock out of the box Google Gemini these questions and asking it to build these reports in HTML with our brand style guidelines. That’s all this is. This is not high-tech. This is not, you know, hundreds of thousands of dollars in technology and consulting. This is literally just asking questions and having it put in it in something that doesn’t look like garbage for output. That’s it. If we wanted to convert this into a true copilot.

Christopher Penn — 30:58
Our next steps would be to take the sales playbook, take our CRM data, and take the prompts that we’ve been using them, and put them into actual workflows like the N8N environment or Python code or whatever, and instead of trying to process all the data at once, because obviously for Trust Insights, I would love that. I would love it. If we had the problem of, like, “We have 10,000 deals.” There’s no way that Gemini can process all 10,000 all at once. We don’t have that problem. Gemini can handle our. It still fits within the month, 1 million token windows, but you would instead have a system that programmatically goes and does each deal individually, one at a time, and then passes the updates to and from an LLM back into your sales CRM.

Christopher Penn — 31:45
So in terms of a live copilot, as deals, in fact, one of the things you can do in a system like N8N is you can have triggers themselves where the trigger is something that happens in HubSpot, and the N8N system listens to HubSpot. So like when a deal is created, that could trigger a next best action flow to say, “Okay, I’ve looked at the deal, I looked at what you’ve put in it, here’s your next best action.” And obviously, you can create the. If we look up here, you can update tickets, you can create deals, you can create contacts, you can create engagements, tickets and things like that. Or you could just tie in like a Gmail thing, and say, “Hey, I see the deal. Here’s your next best action.”

Christopher Penn — 32:34
And then when you assign it to a person in HubSpot, that person gets the email saying, “Hey, your next best action on this deal is to do this.”

John Wall — 32:42
Yeah. And there’s also value, even for the stuff that doesn’t close, you know, as something dies, to have it go through and say, “Okay, we do a check back in six months and at a year,” you know, things like that to take care of all that, you know. Otherwise, you have somebody that’s having to set up all these tasks and appointments further out in the calendar. That’s completely work that can easily be automated.

Christopher Penn — 33:02
Exactly. So within a system like N8N, you would just be enabling these reports and these analyses. It would just be automatic rather than manually doing it. But even if you don’t do that, if you just do the manual thing that we’ve just done now over the last 35 minutes, if you do that at the end of the week, I mean, it didn’t take that long. I. This week, due to time constraints, I didn’t have a chance to pre-bake anything. So we did this entirely live. And we have what, five really solid reports from our sales pipeline that will take you under 20 minutes a week to look at and go, “Okay, these three deals, we need to talk about them at this week’s sales meeting. Here’s what we need to do. For each of these new deals in the pipeline.”

Christopher Penn — 33:52
The, the ones that are at the bottom here that are just not good fits, like kick them off the deal board. If they’re not going to buy, kick them off the deal board because we’re just not going to provide value for them. Here’s the next four webinars we have to do, right, to say, like, yep, we need a regular regulated industry webinar, AI for regulated industries. And then the, the, I guess the cleanup and maintenance, like, who can we win back?

John Wall — 34:19
Yeah, that’s all solid stuff. But again, a plug for. Just as you said, this is all possible because we have the process and procedures set up. That’s the thing. If you are not to this point and you want to know how can we configure our CRM and set up a book of policies and procedures so that we’re gathering this data so we can finally take advantage of this stuff, we help out with that stuff, and we can get you where you need to go. So you’ve got actionable data.

Christopher Penn — 34:46
Exactly. Anything else on your wish list as a sales leader, John?

John Wall — 34:51
Yeah, no, I just wish there was like an arm twist button, you know, that I could hit and just make a signature appear on a document. That would be the real ultimate power. But no, it’s, you know, it’s like every single deal is, there’s human behavior, there’s emotion, there’s empathy. So having these tools to simplify and build a process, it allows you to spend the time on the difficult stuff to qualify and to see, you know, what’s holding you back while you’re not getting that signature as opposed to spending your whole day, like setting up tasks six months from now and, you know, drilling through the database trying to find that one that you forgot, you know, last week and now it’s buried in the pile and you can’t get to it.

John Wall — 35:33
So it’s, yeah, this is all great stuff to make you more effective and put you where on the problems that you need to take care of and get everything else off your plate.

Christopher Penn — 35:43
Yep. So I imagine as Katie is watching the replay of this, she will be asking where the Trust Insights version is that uses our actual data. So after we hang up from the live stream, I know exactly what I’m going to be doing.

John Wall — 35:55
Yes, to do list is again, the live stream always seems to generate heat on the to-do list for both.

Christopher Penn — 36:01
Of us, which is the way it’s supposed to go. I mean, the fact is that if we have this stuff documented, we should be able to put in a place, play. Now, what I would like to see, and I don’t know how close we are to it, is I would like to see this automation built into our CRM so that we don’t have to do this in external tools. I mean, it’s great that we can go into Gemini or ChatGPT or Claude or whatever and just crank this out, but I almost feel like they should be doing it for us. Like, what are we paying them for?

John Wall — 36:34
Well, yeah, that’s who. You open a very big Pandora’s box with that one. As far as what’s happening in some of these big CRM companies, you know, it’s. Yeah, I just. With everything happening with N8N, it just seems like end users are going to be doing this stuff far faster and more effectively than sitting around and hoping and praying that the product team gets to this stuff.

Christopher Penn — 36:54
Exactly. And that brings me to the last topic, which is, and I was talking about this on LinkedIn earlier, this is concept, this is technical concept called MCP Model Context Protocol. And what MCP allows you to do is it allows you to call external tools and data from inside a chat window. So as an example, if I bring up Anthropic’s Claude, which is one of the Claude desktop app supports MCPs out of the box, you have to install it. But then once you’ve got that, you, you have something that you can work with. And if I say, “Let me find a, a document here that I can use. Oh yeah, let’s use this one.” Let’s actually use the blog post. I’ll say count the number of. Try my screen share here. Count the number of words in this document using the Word Counter tool.

Christopher Penn — 37:50
The Word Counter tool is an MCP Model context protocol server that I built that does one thing. It does word count. That’s it. Like, it requires no other horsepower than that. Let’s see if Claude remembers that it has a word counting tool. I’m on Sonnet and I’m in the right place. It’s thinking, but a lot of the prompts and a Lot of the things that we talked about today are things that if you were to put them into an MCP server, then you could do like, you know, at Sales Playbook, “What should I do with this deal?” And if you bring in like, you know, you would say like, at HubSpot CRM, show me the deals I need to work on, the three deals I need to work on today.

Christopher Penn — 38:36
And then you could do at, you know, at Sales Playbook, how should I process these deals? It would be able to understand that. So here you could see now, hey, I remembered that I have access to the Word Count tool. I’m going to send the Word Count tool the document. The Word Count tool is going to launch separately, and it’s going to spit back answer that says 708 was what came back from the tool, which is the correct, actual correct number. Interestingly, if you were to do this without the Word Count tool, you’ll get answers between 500 and 1000. Because these tools can’t count.

John Wall — 39:14
Yeah, you just get a hallucination, which is hilarious.

Christopher Penn — 39:18
Like, why don’t you know this? Math is hard. Math really is hard for these things. But I think that’s the thing that I would add in. In is, like, if you were to think about the sales automation, what would it look like to have your sales playbook as an MCP tool right inside of. So if your employees use Chat GPT, which will support MCP soon, you could have it just provide recommendations immediately. Say, “Like, oh, okay, it’s. It’s this.”

John Wall — 39:55
Yeah, that’s. I can see that being very useful. That’s another, you know, just like next-wave sales I did. There’s no catchy term for this yet. You think something would have already stuck as far as next-generation AI-powered sales. I would have to put some cycle time behind that.

Christopher Penn — 40:11
I mean, you have Salesforce Automation, so I guess it would be Salesforce AI.

John Wall — 40:17
Yeah, no, we’re not going that route. We definitely need something else.

Christopher Penn — 40:23
All right, well, yeah, I probably can’t even use the word Salesforce, huh?

---

Need help with your marketing AI and analytics? Get help with your AI strategy » Learn generative AI for marketers with our new course » Prove the ROI of your marketing with our advanced analytics services » Just got a "silly" question you want an answer to? Ask us here »	You might also enjoy: How AI is Changing Marketing Keynote How To Build Better Analytics Dashboards and Reporting Instant Insights: The Beginner’s Generative AI Starter Kit Instant Insights: How to Write an Effective ChatGPT Prompt Artificial Intelligence Consulting
Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, INBOX INSIGHTS. Subscribe now for free; new issues every Wednesday! Click here to subscribe now »	Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new episodes every Wednesday. Click here for Google Podcasts » Click here for Apple Podcasts » Link for all other podcasting software »

---

Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.