Signal Integrity Is the Foundation for AI Go-to-Market Agents
Episode Summary
Lara Shackelford, CEO of Hawksmoor.ai, explains how to build an AI go-to-market system, starting with the customer signals it runs on. When those signals do not connect, she says, the cost is growth: her example is a sales executive getting on calls with customers who had a horrible call with customer success two weeks earlier, a call logged in a system that was not connected to the seller. Her signal integrity model has three pillars. Semantic consistency means every system shares the same definitions, temporal coherence means knowing when something happened, and lineage transparency means seeing the path that led to it. Her work starts with an audit of every system that touches a customer, then a dashboard that ranks the signals by their importance to the customer experience. The second half turns to where human judgment stays in the workflow, why an outside party should run the audit, and why the audit comes before the agents.
Key takeaways
Connect the system where customer success logs a problem to the seller who calls that customer next. In Lara Shackelford’s example, the customer success person was not trying to let anyone down and logged the call, but that system was not connected properly to get the information to the seller.
Put a shared semantic layer in place before you look for the payoff from AI. She calls semantic consistency one of the most crucial pillars, and says that without a semantic layer in place, the payoff from AI often does not happen
Begin the audit with every system where you engage prospects and clients. Hers covers the customer relationship management (CRM) system, the marketing automation system, Google Analytics, the advertising platforms and web traffic, and she says it is a lot, but crucial, because it lets you see all the signals related to an individual and a buying committee
Find out whether your teams date the same revenue differently. She says an audit quickly shows one group reporting an opportunity closed on one date while another uses the date the money hit the account, which a smaller company does not always see and which has major implications for the revenue it reports
Put the signals on a dashboard, ranked by their importance to the customer experience. She monitors them on an ongoing basis, starting with the signals she believes matter most to that customer
Keep a human on the communication that goes to your most senior client contacts, for now. She says that today you do not want communication drafted by an AI or a large language model (LLM) going directly to a client’s CEO, though she adds that no one knows what the future holds; you can automate many steps along the way and even recommend content and language
Inventory the agents you already have. She includes it in the audit: which agents you have today, which to keep, where you need more, and the places where, today, a human has to be there for the interaction
Run your main AI initiatives through a filter, and set guardrails for everything else. Her use case workbook looks at time to value, growth or efficiency, and risk, but she says you will not run every agent through it, so you also need a framework for where in the organization it is safe to play and the limit beyond which you talk to the department head and make a case for investment
Bring in an outside party to run the audit. Having done it from inside an organization and externally, she says it is best to bring in an outside party, because people inside bring biased views, with no bad intentions behind them, and make assumptions
About Lara Shackelford
Lara Shackelford is CEO of Hawksmoor.ai, which helps leadership teams build and execute enterprise AI and AI-native go-to-market strategies. She has 30 years in enterprise technology and go-to-market leadership at companies including Microsoft, Oracle, Intel, Marketo, DataStax and Looker, and her roles have included Global Head of AI and Demand Generation at Microsoft and a global leadership role at Intel covering AI strategy, product and industry marketing. She holds a Postgraduate Diploma in AI for Business from the University of Oxford. Hawksmoor.ai launched publicly at the HumanX 2026 conference in San Francisco.
In this episode
| 00:06 | Welcome, and the customer context an AI go-to-market system inherits |
| 02:14 | What disconnected signals cost, and the customer success call that never reached the seller |
| 03:57 | What a signal is: internal and external |
| 05:02 | The three pillars: semantic consistency and the tennis ball, temporal coherence, lineage transparency |
| 07:30 | The signal audit, and ranking signals on a dashboard |
| 10:47 | What an LLM looks at when it represents your company |
| 13:31 | Why brand and demand are one investment in AI search |
| 14:56 | Where a human stays in the loop, and an inventory of agents |
| 16:46 | Where LLMs still fall short on an authentic voice |
| 17:37 | A life sciences company planning for competitors catching up |
| 19:35 | The use case workbook, and guardrails for the rest |
| 20:32 | The first steps: an audit, then management readiness |
| 22:07 | Why an outside party should run the audit |
| 23:24 | Resources, and a community for women in AI |
| 24:39 | The leadership skill: persistence and resilience |
| 25:13 | Remember one thing: know where you are starting before you build agents |
In Lara’s words
“You ask an LLM a question and it’s trying to figure out which version of a tennis ball do you mean, and is that even a tennis ball at all? And it’s just churning through tokens.”
Lara Shackelford
“They still haven’t nailed that completely authentic version of someone’s voice.”
Lara Shackelford
“It’s best to bring in an outside party because ... you’re going to have so many biased views and with no bad intentions behind them that people will make assumptions.”
Lara Shackelford
“The two words that have been most often used to describe me are persistence and resilience, and I feel like those two are at the heart of what’s required.”
Lara Shackelford
“People can create agents today. But you’ve got to know where you’re starting and where those gaps are. Otherwise, you’re going to be building things that don’t make sense, and they’ll break in a bigger way, very quickly.”
Lara Shackelford
Resources
Lara Shackelford and Hawksmoor.ai
Lara Shackelford on LinkedIn: Her professional profile
Hawksmoor.ai: The enterprise AI and AI-native go-to-market firm she leads, and the first place she sends listeners
GTM Assessment: A free 13-question self-assessment of how ready a go-to-market organization appears for AI-native growth, based on what leadership knows today. She describes an assessment as the lighter-touch place to start, ahead of an audit
GTM Signal Report: A free report she recommends alongside the assessment, which combines detailed signals about your company with external signals to give a sense of where you are today
Signal integrity: Hawksmoor.ai’s own write-up of the model she describes in this episode and its three pillars
Women in Enterprise AI: The community she runs for women working on enterprise AI, where members bring the problems they are working on. It meets most Fridays, and Hawksmoor.ai calls it The Signal Room
Ideas and terms discussed
Signal: Anything related to the buyer experience, internal or external. Her external example is someone searching a non-branded term for your business in a large language model (LLM); her internal example is a product marketing asset used by demand generation in a campaign, which turns external when a prospect engages with it
Semantic consistency: Every system and team sharing the same definition of the same business concept, so a product means the same thing everywhere. Her example is a tennis ball that another system calls a fuzzy, yellow, bouncy thing
Temporal coherence: Knowing when something happened, so you can act at the right time based on it
Lineage transparency: Being able to see the path that led to a signal, so a seller alerted to reach out can see what the prospect did to trigger it
Signal dashboard: The dashboard she builds after the audit, which ranks signals by their importance to the customer experience and is monitored on an ongoing basis
AI visibility: How a company shows up when buyers ask an LLM. She describes the dimensions an LLM weighs, among them consistency and credibility
Use case workbook: The part of her audit that looks at current and future use cases for time to value, growth or efficiency, and risk
Automate, augment, recommend or keep human: The four modes for how AI participates in revenue workflows, named by Christina Ellwood when she asks where human judgment should stay in the loop
Also named on air
Marketo and Google Analytics: Named as systems the signals come from: a campaign response in Marketo, and Google Analytics 4 as part of the audit
G2: The software review site, and her example of a citation shift worth watching, when a company stops showing up there in the same way
Reddit: Named by Christina Ellwood as a source the AI engines weight heavily even where a company’s buyers are not
Related AI Realized episodes and events
Agentic AI Business Strategy: Retrofit or Reimagine the Work: Blaine Mathieu of The River Group on choosing between retrofitting the work a company already does with agentic AI and reimagining it in a way competitors find harder to copy.
AI Search Visibility: When AI Says Your Company Is Dead: Curtis Sparrer of Bospar on how AI answer engines decide what a company is, and how to correct the record when one gets it wrong.
AI for Go-To-Market: The New Revenue Team Playbook: Jonathan Kvarfordt on how revenue teams are being rebuilt around AI, and how he measured a generative engine optimization and public relations program.
Agentic AI and Revenue Work: What Actually Pays Off: Christopher Penn of Trust Insights on where agentic AI has produced revenue, and how to measure it.
Frequently Asked Questions
-
Signal integrity in go-to-market means preserving the meaning, context, timing, identity and lineage of information as it moves across a company’s systems, teams and time. That is how Hawksmoor.ai defines it, and Lara Shackelford, its CEO, describes what happens without it: a sales executive gets on a call with a customer who is angry about a horrible call with customer success two weeks earlier, and learns about it only from the customer, because the customer success person logged the call in a system that was not connected properly to get the information to the seller. She says the cost is growth, customer satisfaction and employee engagement, and that organizations that think they have a growth problem often do not see that the answer is in the system, right in front of them.
-
The three pillars of signal integrity are semantic consistency, temporal coherence and lineage transparency. Semantic consistency means every system shares the same definitions: in Lara Shackelford’s example, a company whose systems call the same product a tennis ball, a fuzzy, yellow, bouncy thing and something used in a sport on a court leaves a large language model (LLM) just churning through tokens trying to work out which one is meant. Temporal coherence means knowing when something happened so you can act at the right time, and lineage transparency means seeing the path that led to it, so a seller alerted to reach out can see that the prospect responded to a campaign in Marketo. She calls semantic consistency one of the most crucial pillars, and says that without a semantic layer in place, the payoff from AI often does not happen.
-
To audit go-to-market signals for AI, start with an inventory of every system that touches a customer, so you can see all the signals related to an individual and a buying committee and find where they are breaking down. Lara Shackelford’s audit covers the customer relationship management (CRM) system, the marketing automation system, Google Analytics, the advertising platforms and web traffic, and she says it quickly reveals problems such as, at a smaller company, one group reporting an opportunity closed on one date while another uses the date the money hit the account. From there she builds a dashboard that ranks the signals by their importance to the customer experience, monitors them on an ongoing basis and starts with the most important. She says there is always an operational layer of cleanup, and that growth and innovation work moves forward while the underlying infrastructure is being fixed.
-
The first steps to building signal integrity are an audit, which can be preceded by a lighter-touch assessment, and a look at management readiness and organizational maturity. Lara Shackelford says the audit shows where the efficiency gains are, where the gaps are, what needs closing urgently, and how to build a strategy for growth and innovation. The readiness work can start with executive interviews scored across the organization, and she says that almost always an organization is not where it thinks it is, and that there are always a few executives who see it completely differently.
-
To make sure AI represents your company accurately, start by auditing your AI search visibility across everything that would show up in a large language model (LLM), including your website, your press and the content you create. Lara Shackelford says LLMs look at several dimensions, among them consistency and credibility, such as whether reputable partners are talking about you, and that even how a website is registered can affect that credibility. From the audit she recommends a content plan that meets the buyer where they are, and watching for shifts such as a change in how the company shows up in G2, the software review site.
-
Today, a human should stay in the loop for communication at the most senior level of a client relationship, such as a message to a client’s CEO or one of the CEO’s direct reports. Lara Shackelford says you do not want communication drafted by an AI or a large language model going directly to that contact, though she adds that no one knows what the future holds. Her reason is authenticity: these models have gotten much better at narrative in someone’s voice, but they still have not nailed a completely authentic version of it. You can automate many things along the way and even recommend content and language, she says, but at that level it still has to be a human being.
-
When AI lets competitors copy a go-to-market system, that system becomes less unique, so the advantage it gives a company today may not last. Lara Shackelford describes a life sciences client whose go-to-market system is one of the things that differentiates it from its peers, and which came to her saying it had no problem today but asking what happens in 18 months, or less, when everyone can catch up. She says telling a large language model to go and build it is not that simple today, but the models are becoming more sophisticated every day, so the company wants to keep that gap as long as it can while it builds its next differentiation.
-
Yes, an outside party should run an AI go-to-market audit, because people inside an organization bring biased views, with no bad intentions behind them, and make assumptions. Lara Shackelford has run it from inside an organization and externally for companies, and says an internal facilitator just leans into the bias, tempering what they hear from a department because of what they assume it already thinks. She has not yet seen a case where an organization was as aligned as it thought it was, and says the unvarnished truth at the start is what equips it to move together.
-
[00:06] Christina Ellwood: Welcome to AI Realized, the podcast for enterprise executives leading AI adoption from tackling security, data, and operational challenges to navigating organizational transformation, AI deployment offers a unique opportunity to redesign our organizations from the inside out. I’m Christina Ellwood, your host for today’s podcast, and today we’re looking at a problem that affects every enterprise using AI to transform its revenue systems, your AI is only as good as the customer context it inherits. And in most organizations, that context is stale, conflicting across teams, or lost entirely during handoffs. Our guest spent nearly 30 years building revenue engines inside enterprise go-to-market: Oracle, Microsoft, Intel, DataStax, Looker, and she’s here to share what enterprise leaders should build to get their AI go-to-market system right. Lara Shackelford is CEO of Hawksmoor.ai, a consulting firm for enterprise go-to-market teams. Lara, welcome to AI Realized.
[01:38] Lara Shackelford: Thank you so much, Christina.
[01:41] Christina Ellwood: It’s a pleasure to have you. So, Lara, building an effective AI go-to-market system is bringing into focus the data and context gaps in our existing systems. Which, of course, as you and I both know, having done this for many years, they those gaps are are big, are real, are persistent, and require a lot of effort to both close and keep them closed. You’ve said the customer lived one journey, but your systems remembered another, which is a great quote. And an AI amplifies that gap. Where do you see that disconnect, and what’s it costing us?
[02:14] Lara Shackelford: The cost to organizations is tremendous, and most organizations don’t see it. They think that they have a growth problem, and why are we growing at 5% less this year than we did last year? And so many times the system, it is in the system, right? And the answer is right in front of you. And so, what it’s costing us is growth. It’s literally costing us growth. It’s costing us also engagement with employees. When I’ve seen organizations and been a part of organizations where you have systems that are so messy, and the signal integrity does not work between sales and marketing, or marketing and the other groups that they work with, marketing and customer success. A salesperson will get on a call with a customer. Still, I just heard this a couple of weeks ago. I’m then the seller said, "I’m. It was actually the sales executive. He said, "If I get one more call with a customer where I get on and they tell me that they were angry as could be at us, and they just had a horrible call with customer success two weeks ago, and they didn’t tell us, and guess what? No, customer success person wasn’t trying to let anyone down. They logged it in in their system, but that system wasn’t connected properly to get that information to the seller. That’s a great example of it’s costing us revenue, it’s costing us customer satisfaction, and growth.
[03:30] Christina Ellwood: You formalized this into what you call signal integrity, which you describe as having three pillars: semantic consistency, temporal coherence, and lineage transparency. So let’s start with, what do you mean by a signal? What does it mean to have signal integrity? And then, how do these three pillars help us understand whether we have signal integrity?
[03:57] Lara Shackelford: Yeah, a signal would be anything that is related to the buyer experience, that that is internal or external, right? You’re going to have external signals that someone searches on a non-branded term for your business in an LLM. You’re going to have internal signals where, let’s say, that product marketing created an asset related to the to a quarterly theme, and that demand generation or your performance marketing organization use that asset in a campaign that would be an internal signal, and then it turns external, of course, when a prospect or client engages with it. Right, so you’re going to be looking at both the internal and external signals in that example, and that’s there. There’s so many signals, and we’re never. It’s kind of a living, breathing thing. We’re there always going to be new signals.
[04:50] Christina Ellwood: Yeah, so that that can make it very difficult, I imagine, to both audit them and to manage them. So, what do you mean by the three pillars? What is semantic consistency? What is temporal coherence? And what is lineage transparency?
[05:02] Lara Shackelford: Yeah, so the semantic is one of the most crucial, and it’s where your payoff real often doesn’t happen with AI if you don’t have your semantic layer in place. And here’s an example: something as simple as a B2C company that they sell tennis balls, and in one system, it’s a tennis ball, and one system, it’s a fuzzy, yellow, bouncy thing, and one system, it’s used in a sport on a court. Of all used in a sport on a court, I’m not trying to sound like duck. Anyway, so semantic is how do you make sure that you have you realize those definitions all mean the same thing. They all mean tennis ball, right? So, when you’re doing promotions and you’re focusing on how can I get more revenue and sell more tennis balls this quarter, you’re across all of your systems. You’re you have the same interpretation of it, and that semantic the definitions are so important, and if you you could imagine if your definitions aren’t aligned, you end up in a situation where an LLM is just burning through tokens because there you’ve got you ask an LLM a question and it’s trying to figure out which version of a tennis ball do you mean, and is that even a tennis ball at all? And it’s just churning through tokens. So that’s a huge area. Temporal is time, right? And I always just remember it with a T for time. But temporal is that you know when something happened and that you’re able to act at the right time based on when that happened. And then lineage is what what’s the path to that happening, right? A seller gets a an alert that it’s time to reach out to someone. How do they know that’s actually something that they feel comfortable reaching out to? It’s because your lineage transparency is in place, right? And they can see that someone responded to a campaign in Marketo, and they did this. They can actually see it, right? So the lineage piece is so important, also.
[07:01] Christina Ellwood: So that’s a sequence of things that the buyer went through up to that point, that indicates it’s a signal to talk to them.
[07:01] Lara Shackelford: Exactly, a signal along the path.
[07:01] Christina Ellwood: Okay, gotcha. Okay, so for a CRO or CMO who’s never actually audited their signal layer, what you call the signal layer before, walk us through what that looks like. What are you looking at? What are you looking for? And then, what are some of the fixes for the gaps?
[07:30] Lara Shackelford: Yeah, when we go into clients, we do this audit initially, and we’ll look at their CRM system. We look at whatever marketing automation system that they’re using, and we look at all systems where they’re engaging with prospects and clients, and we look at Google Analytics, GA4 today, and other advertising platforms that they’re advertising on, web traffic. We look at every system that touches a customer, and that sounds like a lot, but it is a lot, but it’s just crucial that we have that information because then you can start to paint the picture and say these are all the signals related to this individual into this buying committee, and and so that’s really where we start when we’re doing a signal audit is just getting an understanding of everything that’s in the systems, and when we do that, it’s amazing because you start to reveal really quickly where one group is reporting one date for when they close an opportunity, another group is using a different date because they’re using when did the money hit the account, and you’re a small company. You’re not a big private company or public because that would be in place. But if you’re a small company, you don’t always see those things, but they have major implications for the revenue that you’re reporting. Those are some examples of where we start, but we really do an inventory first in that audit to look at all the signals that exist and identify where signals are breaking down.
[07:30] Christina Ellwood: Okay, and then when you go to fix them, how are you identifying the place to start, the priority or the flow?
[07:30] Lara Shackelford: So we do we create a signal dashboard and actually rank the signals in terms of their importance to the customer experience with the company, and and then we monitor those signals on an ongoing basis, and when we do that, we literally start the ones that are most that we that we believe are the most important to that customer, and the way that we come up with it is looking at every touch point that individual had with us to that point that we know of, right, and and then look at how do we remove friction from that experience with us, and then how do we accelerate? What what things can we do to make that opportunity that experience they had with us even better? So we start with always there. There’s always an operational layer of cleanup always, and we identify prioritize those because there’s never just one thing. But then we look at the growth opportunities, the opportunities for innovation, and then we start pilot. I don’t even want to say pilot because the AI pilot has gone its own way, thankfully. But then we start moving forward on areas of innovation and growth where we can do more while we’re fixing the underlying infrastructure.
[10:14] Christina Ellwood: Yeah, because it’s obviously very important that the signals that you’re feeding into your AI system are trustworthy, and if they’re not, you’re just accelerating a bad situation.
[10:26] Lara Shackelford: Exactly.
[10:28] Christina Ellwood: Okay, for an enterprise leader who’s running a traditional SEO playbook, what should they build to make sure that AI represents their company accurately in the LLMs? Because of course, today a buyer is more likely actually to go to an LLM rather than to a traditional website.
[10:47] Lara Shackelford: Yeah, they one they need to look across everything that that would show up in an LLM and have a good understanding, and they need someone who really under you know digs deep on understanding AI visibility, and there are a lot of dimensions that you look at. Like a an LLM will look at date. They love data. They look at consistency. They look at what’s the word I’m looking for. They look at credibility. Right? Are, for example, partners talking about you, and are those partners who are talking about your business are they reputable? So you have all these different dimensions, and what they need to do is initially they do your own audit, your AI search visibility. You look at your website, you look to see are there things there that are causing any friction and any chance that someone’s not going to find you on an LLM, even as something as where the what the how the website is registered will impact that credibility, right? So it’s really they would want to dig in and look at the different dimensions. And we’ve got a an approach that we start looking at when it comes to AI search visibility. We have an application, so we run the application, identify the areas from their website, press content they create different touch points with a customer and then and then from there they really want to look at making sure they’ve got a plan for content to meet the individual where they are and I can give you a specific example of this where where where a citation shift will will be a major kind of area focus for someone like where they show up in G2. If they stop showing up in the same way, that’s a sign, right? And so, what does that mean, and what do they do to address that?
[12:34] Christina Ellwood: Yeah, these third party signals are also complicated because a lot of the third party signals the AI engines are looking at are not even places where it’s natural for us to show up as a brand. Reddit is very popular. Not everybody needs to be in the Reddit ecosystem. That’s not where their buyers are. That’s not where their customers go. Yet it’s still very biased to that particular source of information. But your point about the structure is also really important. The LLMs are very structured and use consistency as a proxy for credibility if they don’t have anything else. So if you’ve got date inconsistencies or you’ve got even something as simple as capitalization inconsistencies, you will down it’ll downgrade the credibility of all your content because of that. So you really need to be very mindful to be consistent in the way you’re talking about the business. That means brand and messaging and positioning need a structured way to evolve.
[13:31] Lara Shackelford: Exactly, and this is where the whole brand and demand for marketing leaders has been such a debate for so long. And it’s, I recently worked with someone who runs inbound marketing for a very large tech technology enterprise company, and and they said, "Can you help me? How do I how do I figure out how to tell my leadership where I need to invest in brand and where versus where I invest in demand? And it was like, it is one and the same. Your brand, you have to invest in brand to be able to show up in the right way in LLMs and meet your buyers where they are.
[14:04] Christina Ellwood: So you lay out four models for how AI should participate in revenue workflows. I think these are also quite useful for people who are trying to think through their own systems. You talk about four modes: automate, augment, recommend, or keep human, and those are popular labels in other agentic system evaluation as well. Do we do we fully automate? Do we just augment what we already have? Do we allow it to recommend, or do we keep it human and use agents to help the human rather than the other way around, so you’ve been through automation waves, and you know every vendor says automate everything, but specifically in the go-to-market system, should a leader keep human judgment in the loop? And if so, what’s the test for making that call that you apply?
[14:56] Lara Shackelford: Yeah, I certainly you want to automate where you can and where it makes sense, but there are certain points. Like, I’ll give you an example of you have one of the largest. Let’s say you have a management consultancy that is, you have a partner who runs an industry for one of the top three management consultancies. That partner is so focused on their communications with their customers. They’ve been in that industry, that that particular industry in management consulting for probably 30 years, 35 or longer, and if you if you had communication that you developed and it was going to go to one of their biggest clients, the CEO of Coca-Cola, let’s say, or their one of the CEO’s direct reports, you’re never go. I shouldn’t say never. We don’t know what the future holds, but in today, you do not want communication drafted by an AI or an LLM going directly to that contact. So you can automate so many things along the way to creating content for that individual, to the kinds of super executive exclusive events that you invite them to, but when you’re communicating with them, that it has to be human today. And you can even recommend content that you might use and language you could use in your communications, but it’s still got to be a human being at that level. That’s just one example where, and I do, by the way, an inventory of agents as part of the audit. What agents do you have today? What should you keep? Where do you need more agents as we start to build this strategy? And but where are the places where there’s today no exception? A human has got to be there for that interaction.
[16:33] Christina Ellwood: And am I right in thinking that you say that’s not a case for an agent because the fidelity is so important in that communication, or is there some other standards you’re applying?
[16:46] Lara Shackelford: The fidelity, it’s in just the authenticity, and LLMs have gotten so much better today at language, and language isn’t the term, but narrative, right? In creating messages that are in someone’s voice, but they’re still not. They still haven’t nailed that completely authentic version of someone’s voice. And no matter how much they’ve changed it, in a case like that where you’ve got CEO to CEO level communication, I would never trust that today to a to an LLM to hope that it gets the not hope trust that what you’ve done means that the voice will be correct.
[17:24] Christina Ellwood: So, for companies that are doing well with traditional go-to-market today, maybe they’ve got nice, healthy pipeline. Their revenue is growing. What’s the case for rebuilding their go-to-market around AI?
[17:37] Lara Shackelford: Yeah, there’s one organization that I work with a life sciences company, and they have the sharpest go-to-market system. And they are, it is one of the things that differentiates them from their peers. And they really own how they show up in LLMs. They’re innovative. They’re fast. They’re signal. They’re, they’ve got. They’re just on top of making sure that they understand their signals and they’ve got the signal integrity, but their worry is what happens in three years. That is literally one of two of their biggest competitive differentiators today, and they’re they came to me and said we don’t have a problem today, but what happens in 18 months from now, even or less, when everyone can catch up to where we are in go-to-market today, and that doesn’t become as unique because people can tell an LLM to go and build it, right? And and it’s not that simple today, but they’re obviously every day becoming more and more sophisticated. That’s an example where that company was like, I want to focus on what is my moat in three years’ time, and how do I make sure I continue to be as great as I can be on go-to-market, and keep that gap as long as I can while I build for the future differentiation to make sure the company has has a as good a future potential as they have to date.
[18:30] Christina Ellwood: I think that’s a really important point, and it’s not just at the company level that one must make a decision about whether you are using AI for competitive advantage or you’re using AI to improve your operational efficiency or to address micro opportunities. There’s many ways to use it, and I think we could do a little better with our fidelity in talking about the why and why we made the decision for this case, and what’s the timeline for reconsidering it, or better than the timeline, what are the triggers for reconsidering?
[19:35] Lara Shackelford: Exactly, I part of the audit that I use is this use case workbook, and it’s part of the let’s look at what we have today, but it’s also imagining the use cases of the future, and it looks at time to value. Looks at is this a growth opportunity or efficiency? What are the risks involved? And and there’s a filter like that you should put through your main initiatives. But I’m pausing there because you’re not going to go and use that for every single agent that you build, right? So you also have to have some framework of this is where in the organization it’s safe to play. These are your guardrails, and you can create up to this limit. But at this point, that’s where you need to look at the departmental, talk to the department head, look at exceptions, make a case for investment, whatever that may be.
[20:22] Christina Ellwood: So, for a revenue leader who’s listening, who wants to start to building their signal integrity, what are the first three things they should do and put them in order for us?
[20:32] Lara Shackelford: First thing is an audit. It is get get a sense of where you are today, and you can even start with an assessment, which is just a lighter touch, and and there’s an assessment on our website. Lots of people offer these assessments that you can do that give you a sense of where you’re starting. But then, if are going to you know you’re going to commit to this and you don’t know how an audit is the next step. It gives you the sense of where are the efficiency gains, where are the gaps, what do I need to close urgently, and then how do I build my strategy for growth and innovation, based on what where we see the market going, the opportunity, but also understanding where we are today. So it is that an assessment, an audit. We also do, and I recommend this for leaders. It’s a looking at your management readiness, right, and your organizational maturity, and how ready you are for this AI transformation, and that can be as simple as you start with executive interviews, and you score the responses, and you look across the organization. And almost always, I haven’t had a case yet where an organization thinks that they’re in one place than they actually are. In terms of, there are always a few execs that think that it’s completely different, so I think that’s an important step too. Those two places are where I’d start.
[21:47] Christina Ellwood: That actually provokes another question, which is this: Are these three steps, the assessment, the audit, and the readiness evaluation, are those steps best completed internally under internal leadership, or is it best to think about bringing in someone from the outside who maybe has no no vested stake in the outcome.
[22:07] Lara Shackelford: I’ve done this from the inside of an organ organization, and I’ve done it externally for companies. And I can tell you my experience and what is that. It’s best to bring in an outside party because you’re just you’re going to have so many biased views and with no bad intentions behind them that people will make assumptions. I think we have plenty of people who can go and help us with AI because we’ve all these people who are really good at AI technology, and then the person who those reports to is actually head of consulting services. And those consultants are not there to do AI for the company. They’re there to to service the customers. And so, anyway, I think I’ve not yet seen a case where people are and organizations are as aligned as aligned as they thought they were. And I think if you have someone internally facilitating it, you just lean into the bias, and they think, "Oh, I know so and so is this way. That department leans in this direction, so we’ll just maybe temper that a little, and they need that really unvarnished truth when they start because that’s what equips the organization to move together.
[23:11] Christina Ellwood: Gotcha. Okay, so that’s good advice. Before we wrap up, let me just ask a few closing questions. What resources would you recommend for listeners who want to learn more about you, your company, and the model that you’ve described today.
[23:24] Lara Shackelford: Yeah, certainly. On my website, you can find the go-to-market assessment. I also have a signal report, which will give you a sense of some of the signals that are interesting and different ways to engage by combining alpha really detailed signals about your company with external signals that come in, so I would take one of those two or both, get a sense of where you are today. I also do a shorter version of the audit, a more in-depth assessment to help a company get started, and so that’s the place that I would start. Also, if for any women who are listening that want a place that is, they just want to be able to talk it through with other women. I run a women in AI community. We, starting next week. We meet every Friday, and the women in the group will just bring problems that they’re working on, and it’s a really safe way for them to say, "Hey, we’re starting down this journey, or we’re midway and we’re here. What are you all facing? How are you addressing situations like this? So that’s another good place for women to be able to start.
[24:25] Christina Ellwood: I like it. Okay, and what in the AI revolution here? You’ve been a leader for a long time. What are you finding is your sort of superpower as a leader? What’s the skill most valuable for leaders in the AI era?
[24:39] Lara Shackelford: The two words that have been most often used to describe me are persistence and resilience, and I feel like those two are at the heart of what’s required. You have to be willing to persist, yeah, and you have to, and you have to, and it’s not just you as an individual or your team, but you need to build a resilient system, right? And that means signal integrity, that means governance. So those two terms really ring true to me in this era we’re in.
[25:06] Christina Ellwood: If listeners remember just one thing from our chat today, what would you want them to remember and why?
[25:13] Lara Shackelford: Remember to start with and start and audit your situation, and you can do an audit quickly. Doesn’t have to be long, but start to understand where you are and where you want to go. That’s I. And whoever you work with, however you do it, yes, everyone wants to jump in, and people can create agents today. But you’ve got to know where you’re starting and where those gaps are. Otherwise, you’re going to be building things that don’t make sense, and they’ll break in a bigger way, very quickly.
[25:13] Christina Ellwood: Okay, Lara Shackelford, CEO of Hawksmoor.ai. Thank you for sharing your experience with us today.
[25:13] Lara Shackelford: Thank you so much, Christina.