Subscribe
Share
Share
Embed
Welcome to Selling What's Possible, the podcast that's pushing the boundaries of modern account sales. I'm your host, David Irwin, CEO of Polaris I/O and a veteran with 30 years of experience in successful account sales programs.
In each episode, we'll dive deep into the world of strategic account development, uncovering innovative approaches and fresh perspectives that you may not have considered before. We'll be joined by top sales professionals, revenue leaders, and dynamic innovators who are reshaping the landscape of account sales.
Whether you're navigating the complexities of key accounts or seeking to expand value-driven outcomes for your customers, this podcast is your guide to consistently growing your strategic account relationships.
Get ready to challenge conventional wisdom, explore new methodologies, and unlock the full potential of your account sales strategies. This is Selling What's Possible - where we turn potential into reality.
Selling What's Possible E14: w/ Keith Mescha
Dave Irwin (00:03.352)
Welcome to another episode of Selling What's Possible. I'm Dave Irwin, your host, and joining me today is Keith Mescha, who is Managing Director of UI Technology Consulting. And our topic is going to be around the future of selling with AI, how AI is transforming sales. And I'm delighted to have you here, Keith, because you are one of the people that's been involved in some of the world's biggest commercial system installations from..
Everything related to CRM integrations of data and other systems and to environments that are used by sellers and account sellers and, and do this around not just EY, but also large scale companies. So welcome. It's a great session and a great honor to have you here.
Keith Mescha:
Appreciate it. Glad to be here. Yeah, it's a new world that we're embarking upon and it's great to be a part of an organization that is leading in this space. We like to call ourselves Client Zero on a lot of things. Yeah, it's a tough one because it's funny. was writing a letter to my neighbor's remma graduation party and I used AI to write it. My wife immediately said, you didn't write this. I can tell you used chat GPT to do it.
So how do you avoid that in your sales cycles and over leveraging the tools and making sure that it's still humanized? I saw a bunch of articles on LinkedIn recently from some friends of mine about how do you make it not sound like it was AI generated, even though you're leveraging that tool set. So it's going to be important that it's personalized, but not so personalized that it's not you and it's not genuine from a sales perspective, because we all know sales is relationships and people buy from people they like and respect and trust. However, on the flip side, it's impossible to keep up with all the latest and greatest news and information and insights that come out.
You can't, there's probably 35 conferences a week, feel like these days and you can't be in all of them and you can't read all the papers and the research and the millions of podcasts out there. So how do we leverage these tools to bring the right level of insight, put the humans in the mix there to customize it and make their lives easier versus no action, which is kind of what's happening today. As you show up and you're unprepared, you don't have information, you didn't spend the time. So make sellers lives easier by bringing something that's partially completed.
Keith Mescha (02:07.342)
You can complete it, have an insightful conversation in real time with information that's up to date. The other side of that is, and then we've learned this in our experiences, there's a lot of corporate data in organizations. And when you ask these GPTs and things for information, you're getting stuff that maybe doesn't apply. I know hallucination is a big thing out there, but even more important, it's just bad data or unsecured data that shows up that does not go into the ask, right? And so prompt engineering, those kinds of things become important, but you know,
To think everybody's going to get the greatest prep engineer in the world is I think a fallacy, right? You've got to learn that AI needs to get smart enough so I can ask it in my own words, in my own corporate lingo, and I still get the insights and things that are important for there. So we've had a number of discussions recently around specifically marketing, which is part of sales, is how do we leverage the technology to help us create compelling content faster? It's really about faster and more than it is replacing what we're doing today.
you know, talk about weeks, weeks to content. Let's get to days to content or hours, right? So it's about acceleration more than anything. It's not, it's not going to replace what you're doing today, but it's going to accelerate it. Yeah. And you got to think about also in regulated environments and or large organizations that it's a little slower. Approvals have to happen and there's your chains of command on, is this the right messaging? it on brand? And so by the time you get something through the pipeline and maybe six weeks, that information's at the speed of business today, it might not be relevant anymore.
How do we reduce that cycle, that approval cycle? So things come, you I always think of the lawyer use case, right? Reviewing large documents quickly, reviewing marketing content quickly and serving up where we think there's challenges, you know, for the marketing automation side of the house, which then leads to being able to get those insights in there more quickly. Anybody can go read the documents and go research companies, but if you can bring a synthesized summary and insights like you talk about to drive that, that just, it just makes for a better experience and a more relevant topic points to discuss with folks.
I can go search LinkedIn and see who changed jobs and reach out. But boy, if I got that served up my CRM as an action to go do is every morning, 15 minutes, here's the insights about the people you're working with, the context you've had in the last 90 days and where they're moving and going and new job announcements versus having to go seek it out. It's served up to me. That's where I see this technology really making an impact in a seller's day-to-day life.
Dave Irwin (04:24.822)
I agree completely. Yeah. The sellers are overwhelmed. It's interesting. You bring up these obstacles that can stand in the way. So even if you get that part right, are there other processes embedded in client organizations that actually slow it down from there and prevent it and prevent the efficiencies from being gained?
I think one of the things that I've been doing a lot of reading and I do a lot of audio books around AI is, know, the point systems are great, right? Addressing a pain point, but it really comes down to a systematic change in how work is done across the entire organization is where you actually see movement. So yeah, great. I can provide insights, but if I can't get an email approved to go out for 30 days, what's the point of getting that insight quickly? So we've got to look at the internal systems as well as part of that. So.
And the approach we've taken at EY is we're looking at AI in all areas of our business, not just in key pain points, because we've got to automate the entire front to back system. And it has to be part of that. And where does the right technologies matter? Sometimes we throw a technology at a problem where it's really not a technology problem that needs to be solved. It's maybe a process or something, or we actually need people doing something a little bit differently in this particular scenario. But we got to free them up and we can do that with technology elsewhere to allow them to do the thing we need them to be focused on.
keeps falling off the list. I also go back to deal scoring has been around forever in CRM systems. It's been in Salesforce and Microsoft Dynamics for quite some time. I'm sure the other vendors have it too. that's great. It's a healthy deal. It's a trending up deal. But boy, if you can tie that with some insights and actions to say, hey, this one is flailing, but look at these relationships over here. And this person just got moved into the organization. And they just signed this net new contract with XYZ company. Now you've got something to go maybe make.
change that health score, right? Because it's only as good as the data that's being fed. I think that's one of the other big areas we talk about is data silos. The AI can only act on what it knows. And we've seen these out in the news where the models were trained on two-year-old internet data. So you ask it like simple questions like, what's today's date? It doesn't know it, You certainly in enterprises, there's a lot of data in a lot of places and a lot of separate systems. Do you have a strategy around pulling that together?
Keith Mescha (06:30.574)
so that it can actually provide the right value and insights. we're seeing that, I see that today in my world. We've got a wonderful SharePoint with all kinds of knowledge in it, but if it's not tied into our sales co-pilot, it's not helping me, right? It's not going go elsewhere. And I ask a question and I get two different answers depending on which co-pilot I'm in. Like that's not a great user experience and your adoption is going to suffer if you don't fix those problems. And those are not front end user problems. Those are backend system problems where you have to change the entire system front to back on that. And I don't know that
A of leaders and companies go to conferences and here we need to be using AI and they come back and say, go make AI happen, but they don't realize the investments that are involved in that. And the things that have to happen to really enable it efficiently. There's a big problem there, but there's also how do you do something and change your culture to think AI first? And it's going to take time. It's not an overnight sensation, but people are scared of it don't want to use it. You'll never get the ball rolling in the right direction, I think is kind of where I stand.
I think the right way to approach it is you define a use case, but you have to track the workflow all the way through front to back of what the AI is impacting across the whole chain of functions or activities. And there's so many of those in the account planning sector, building account plans, the research side, updating the CRM, all those kinds of things. Where do you guys see some of the initial productivity gains? Are there certain low-hanging fruit that are just..
Easier, let's measure, measure, measure, figure out where the gaps are and then go identify those gaps. And it might not be the most obvious areas. It typically isn't. It's typically somewhere you don't think is that area. So if I think about account selling and, and in large accounts, right, there's, there's probably multiple things happening at that account. And how do you get the information and knowledge to everybody so that they're all on the same, on the same page? So sure. I can go into CRM and I can look at all the closed one ops and I can do this data munging and drilling through and clicking, you know, 7,000 clicks to get to the answer.
Well, what if I just have a account summary that gets automated, generated to, know, AI and it goes out and does that for me and does some analysis, not only with the external data, but then bring in some external data to that and say, okay, we won these, we lost these, here's who was involved, here's what was involved, here's what the scope and maybe it's deal size. And it could start to resonate on some of that information, at least get you, you you're to have a conversation about expanding an account. Like where should we be thinking about this? And then you bring in the external data like.
Keith Mescha (08:52.268)
you guys provide in your platform, right? It gives us some signals and things around that to say, you know, should we even be talking about selling them X because they're so focused on Y or they have these other, you know, motivations or, we understand their, their, their revenue cycles or their seasonality comes into play right now is not a good time to go have this conversation with them. Bring that into your account planning and strategy sessions. I think a lot of people think about the productivity of running an email faster. Well, you you only get so much value out of that, but if you've got 12 people on an RFP call,
for three hours a day for three weeks. Boy, I'd love to gain some efficiency there. $100 sitting on a call and typically two or three people are doing most of the talking and one person's updating the deck and the rest of the people are multitasking. So how do we fix that problem? And you know, here's the RFP, here's our account strategy, shove it into the proposal engine, it kicks out 80 % of what we need and then we spend half hour a day with three people over the course of a week. same cost of sales goes down significantly with that, right?
Dave Irwin (09:50.548)
It's all the downstream activities. you think about, you know, what goes into just a single pursuit, the number of people involved and the amount of assets that are created, the number of meetings and engagement back and forth with the client. during that entire time with all the people involved, I mean, it's hundreds of thousands of dollars of investment. You add that up on big accounts. It's millions of dollars are being invested to pursue things that you really need.
to try to optimize that win rate. to do that, you have to be relevant and you have to be continuously relevant and aligned. So the more you can automate across that entire process, I think from putting together the packaging around what's in the solution, how does it map to the need? Does it get adjusted? Who are the resources that are assigned to work on that particular initiative? Do they match up with the stakeholders and the altitude level all the way to the pitch and the value proposition at which pitches have worked well in that situation before? I AI ultimately
can automate so much of that, but you have to have the data feeding in to train the models to link everything together.
So absolutely that's a goal, right? How do we bring the right relevancy? There's so many times and we have hundreds of teams, channels and chats and all these things, right? When has anybody ever done this or where's the last time we did this? The amount of times I see those a day, it makes my skin crawl, right? Where's our AI agent that is like, go find the right solution that's in my teams that I can just like, hey, my client's interested in these things. Here's kind of what their industry is. Here's what their go-to-market strategy is.
provide me a list of all the top 10 things that EY offers that I could be bringing forth to my client. And I always say like, I'm only one person. I can't do that for everything, right? And there's probably hundreds of me at EY, but I only know who I know in my network. I only know the solutions that I'm aware of. And to be honest with you, I get overwhelmed sometimes. I forget things I've even actually built four or five years ago because it's going so fast. So how do I use this technology to remind me and make things relevant or what's changed or what's new or maybe a solution we were early in.
Keith Mescha (11:48.13)
couple of years back has matured significantly. And so we had some fits and starts with it, but now it's matured and we should be reigniting those conversations. You had a guest on recently about the wind back strategy, right? So maybe two years ago, it wasn't the right time, but now it's the right time to have that conversation again. And the cost to acquire an existing customer relationship with is way lower than net new to conversation. So, you know, I have this thought after watching the prior episode around how do I build an agent to go look through all our closed opportunities and all our closed projects and
generate a list of leads for myself just internally, right? And so we should have that reactivation agent. Like, how do we go build that? I think that's an important piece and every client we talk to, cross-selling is a topic, especially with mergers and acquisitions that happen. where's the, you know, I would love an AI agent that goes through all the webinars I didn't attend last week and give me like the top 10 highlights of things I should be bringing to my clients. That'd be a great agent to go build if I can do something like that, right? So again, is the technology there yet? Partially, I think it's getting closer, but.
it certainly will be in the next couple of years. These things will be more and more a reality. if we're not thinking this way now, we'll be behind the eight ball by the time this stuff comes around and it'll surpass us,
So what's been expected that the sellers will just somehow magically piece all these things together. And they're given assignments like go cross sell everything that we've got over here, but without some kind of agent that can surface, you know, the context of when those other capabilities are relevant. It's a great area though of use case to apply. Who in the organization, a typical client organization would be in charge of creating these agents and or inventing the use cases from which agents would be?
Keith Mescha (13:25.932)
That's an interesting question. And I think that's part of the problem is where do these things land? And I've seen this over the years with just with the low code platforms. I'm very involved in that as well. And it's always trying to have a conversation like it's not really well defined in a lot of organizations who owns that. Is it IT? Is it the business? Because it's not coding. It's not development. It's low code. It's more of the office activation pieces in the Microsoft stack. Right. And so we've had a lot of challenges with getting commitment because there was no clear ownership.
We've all heard about the sales team go off and buying their own CRM system, not using the corporate IT standard. I've been to so many clients where we've to go clean up that mess, right? Because people are off using their Amex to buy 10 licenses of XYZ solution because they didn't like what the IT was providing, Shadow IT. And these new SaaS platforms, these AI tools are spreading like wildfire all over this. And I think we say it in one of our pieces, like, if you don't think your people are using it because they're not using yours, they're certainly using it. They're just not using the one that they're supposed to be using.
Right. I've seen that quite a bit in companies, right? Somebody's out there using it. It's in the news everywhere. So how do you get to where they want to adopt it? I believe it has to be a multifaceted front. We are seeing the decisions move up into the C-suite more and more and out of the lines of business, because as I said, it has to be a complete company initiative. It has to be a system, not a point solution. Otherwise, it will most likely will not be successful. So there has to be buying, there has to be investment in training, adoption.
And that comes in, know, great. create some trainings and put them on a SharePoint site. No, you have to have the office hours. You have to have the touchy feely in person. We do a lot of hackathons. We do a lot of immersion sessions. We bring all our executives in and do a three hour AI workshop where they go build a new campaign for a new product line in groups. Right. And this is out the executive level of our firm just to get them to understand the power of this technology. But then the business needs to be there to innovate. You know, there's this concept.
a fusion team where you've got business IT and working together. I think it's that kind of model where we see success. And if it's an IT led initiative, it's going to have some pushback. If it's a business only, you're going to have risk involved. Right. So it's certainly a combination. But I think if you can have this innovation center, there's a bunch of different names companies use for it. But I think it's important where they can go break things, try things in a confined environment that's controlled.
Keith Mescha (15:49.334)
and get to that level of comfort, think the success rate goes up significantly. That's not always easy for all organizations, right? So that's why relying on third party vendors, like companies like ours and yours to help do some of this stuff and accelerate it is why we exist, frankly, right? To help our clients do these things quicker than they could do internally. I was recently involved with a client that, you they brought, it was a manufacturing customer, so they have plants all over the globe and they have plant managers and they kind of run independently.
And they brought their business leaders in to a session and said, what ideas do you have? What problems are you trying to solve? What could make your operations run better? And it was, it was very interesting to hear. were none of the use cases that came out or anything I've ever heard of or thought of. And it was actually, they had really simple problems that were easily solved with just basic automation, but they're just so far behind. Like it wasn't even like the AI part of it. We didn't even get to that frankly, because they just had some basic stuff.
And just the fact that we had an AI hackathon and they had the attention brought these people together, should have heard the collaboration sharing across the different centers because there's really not a mechanism for that. So now we're to take that and say, okay, let's, let's do a, a rating. do a business value assessment of which ones are going to drive the most impact. Let's standardize. again, I think you've touched on this a few times, but didn't say it quite exactly, but don't try to do them all. Pick the most impactful couple one, frankly, two, and go do that at scale and get it out to lots of people. So there's.
common shared benefit from that and show the impact to your leadership, to your investment committees. So that you start to bring and show that and then you iterate on it and you start to get that culture of creative thinking brought forth. And it also demystifies the, this is gonna replace my job, which I think a lot of people are scared to bring ideas forth because they think that's gonna eliminate them. Well, by doing one small piece and showing the value and now you're freeing time up for other activities or other pieces of work. And I think it kind of shows that it's a slow.
Inoculation is the word I use quite a bit. Like you're gonna inoculate people into this stuff. It just doesn't happen overnight. I think it's showing and sharing and being a little vulnerable as a leader is that you're doing it. Our leaders do really good job of that. They're using AI, they're leveraging it, showing our people live in sessions, sharing stories of client interactions they've had around these topics. I think that's, again, it's that culture thing. You can't just say, hey, everybody go do this and you're not using it yourself. And so I think back to the old sales management,
Keith Mescha (18:10.978)
Right? It's like, they want everybody to put data in the CRM, but they run their sales pipeline meetings off of spreadsheets. Right. And if it's not in there, it doesn't exist. Right? So again, I think we're going to get to the point where some of that is automated because nobody wants to put stuff in the systems. And, you know, a of the vendors are talking about how do we have conversations with the, with the opportunity pipeline health bot where it tells me, here's my overdue entries. Here's things that haven't been touched in a while. What do you me to do with these? And we're starting to converse with it. So.
We're taking away the friction of just putting the data in the system, but management's got to run the meetings out of the system. Otherwise, nobody's going to put data in the system, right? So I think, again, it's a culture thing. I think leadership plays a big role in it.
It can change the roles people play in that workflow. So instead of just doing more of the same thing faster, because you have AI, you might do it differently and you might shift, you know, what one role does and another role emerges to do something else in that workflow. You know, and you think about like lead management and marketing plays a role and sales plays a role, but should marketing be instigating that particular agent? Should it be prioritizing and qualifying things differently using research? How does it feed that into sales differently?
in a more automated way to both come together in a collaborative setting.
Actually, I think a really good app to use cases, you marketing successful because they sent a hundred leads into the CRM. Then sales is calling those hundred leads, right? But we get a 1 % hit rate. do make that better? Like how do we prioritize those? How do we have marketing do some research and only provide the ones that are high likelihood to have a meeting or something like that in transactional situation? Relationship selling and large account selling is probably a little bit different, right? It's about who do know?
Dave Irwin (19:35.564)
Yeah, exactly. It's very nice.
Keith Mescha (19:53.906)
We do a lot of mapping of who the key decision makers are, who's in what roles. We do a lot of LinkedIn mining of information and looking at who knows who and all this kind of stuff. And you've got all this data in our email systems and our appointments and things like that. How do we leverage that information to tie together? Instead of people deciding where they should be, let the data tell us what the relationship should be based on the interactions. What we say and what the data is showing are two different things. How do we leverage that and have AI provide us some insights into exactly should be different?