Season 2, Episode 05 | AI for Revenue Leaders: The AI Hat Podcast
Hosted by Mike Allton

The Problem Every Revenue Leader Knows (But Hates to Admit)

Your sales team is missing quota. You don't know if it's the pipeline, the messaging, or Stage 3 conversion rates. So you do what any smart Revenue Leader does - you call in a consultant and ask for a diagnostic.

And then you wait.

Two weeks. Three weeks. Sometimes a month.

By the time the report lands on your desk, your quarter is over. The deal you lost? Already closed with a competitor. The bottleneck you needed to fix? It just cost you another $200K in slipped pipeline.

Here's the hard truth: The traditional revenue diagnostic process is too slow for the velocity of modern B2B sales. And it's too shallow. You get surface-level insights when what you need is scenario modeling that shows you where the money is hiding and which lever to pull first.

In 2026, you don't have the luxury of waiting three weeks for answers. Your board is demanding ROI on every dollar. Your competitors are moving faster. And your team is drowning in Admin Drag - spending 70% of their time on non-revenue activities while you're paying them to sell.

What if you could get a revenue diagnostic in 24 hours instead of three weeks? And what if that AI-powered analysis found opportunities your manual process would miss?

That's exactly what Tara Kinney built - and in this episode, she walks you through the entire workflow.

Meet Tara Kinney: The RevOps CEO Who Rebuilt the Diagnostic Model

Tara Kinney is the Founder and CEO of Atomic Revenue, a RevOps management consultancy. She's supported 76 companies across 8 industries and led 18 CRM implementations. She knows what "normal" looks like in revenue operations - the slow qualification cycles, the manual pipeline analysis, the weeks-long diagnostic reports.

And she blew it all up.

Her agency used to take 2-3 weeks to deliver a revenue diagnostic. Now? Her Digital Crew does it in 24 to 48 hours. And it's not just faster - it's smarter.

Her AI agent found a $21 million revenue opportunity that she - a seasoned consultant with decades of experience - had missed in her manual analysis.

This isn't a chatbot writing emails. This is an agentic workflow that analyzes your pipeline, models multiple improvement scenarios, ranks them by ROI, and delivers a report that prepares your strategists for high-intent conversations - all while you sleep.

The $21 Million Story: When AI Outperforms the Expert

Let's start with the moment that makes every cynical CRO sit up and pay attention.

Tara ran a manual revenue diagnostic for a client with a long, complex sales cycle (18-24 months, multiple conversion stages). She found a $10 million opportunity by identifying a way to improve one conversion metric by 2%.

Then she ran the same analysis through her AI agent.

The AI found a $21 million opportunity by suggesting a 3% improvement on a different conversion metric - one Tara wasn't even tracking in her manual review.

What did the AI see that she missed?

As Tara explains in the episode:

"It takes a lot of analysis and calculation to figure out what is the value of improving this conversion rate at this stage. When you have an 18 or 24 month sales cycle and you have nine or 12 conversion points over that timeline, it is easy to see what the conversion rate is. It's much more difficult to mathematically model what it would mean to change that conversion point - especially at stages where you don't know price or value of a customer yet."

The AI could consume more data (proposals, financial reports, conversion scorecards) and cross-reference it faster (financials × conversion rates × pricing models) than any human could manually process.

But more importantly, the AI could model multiple scenarios in parallel and rank them by ROI impact - something that would take a human days or weeks to calculate.

The result?

Both the human analysis and the AI agreed: Don't invest in more lead generation (which would've cost a fortune and only yielded 0.25 additional projects over 18 months). Instead, fix the conversion bottlenecks at Stages 5-7 where hundreds of deals were stalled - and unlock $21 million in pipeline that was already there.

The 90% Sales Cycle Reduction: Hard Numbers on Efficiency

The $21M story proves AI agents can see what humans miss. But what about speed and capacity?

Here's where Tara's workflow delivers measurable P&L impact for her own agency:

"Time in the selling process has gone down drastically. I mean, it's less than 10% of what it was before to get to the same level of agreement between someone on our team and a prospect. So we're saving about 90% on that sales process from the time someone finds us and does the Revenue Reality Check until they get through the first meeting and we're talking about the next steps and making a proposal. And that's huge because until you're making a proposal, that's all cost of sales that has no real project tied to it. So we've eliminated one of the more expensive parts of most organizations' selling process."

Translation for Revenue Leaders:

• Before: 2-3 weeks from first contact to qualified proposal conversation
• After: 24-48 hours from website visit to strategist-prepared discovery call
• Efficiency Gain: 90% reduction in cost-of-sale time
• Capacity Impact: Strategists can now handle 3-4 revenue diagnostics per month (vs. 1 diagnostic every 2 months previously)

This isn't just "faster qualification." This is capacity reclamation - freeing your highest-paid talent from low-value busy work so they can focus on strategic client conversations.

The Revenue Reality Check Workflow: How It Actually Works

Okay, let's get tactical. If you're a RevOps Director or VP of Sales and you want to understand how this workflow is architected, here's the step-by-step breakdown Tara shared:

Step 1: The Prospect Interaction (10 Minutes)

A prospect visits atomicrevenue.com and clicks the Revenue Reality Check link. They interact with Revenue Maven - Atomic Revenue's AI personality (with a "charming AI voice," as Tara describes it).

Revenue Maven walks them through a series of diagnostic questions about their:

• Sales process and conversion stages
• Marketing alignment and lead quality
• Operations handoffs and client onboarding
• Tech stack and data infrastructure

Total time investment for the prospect: 10 minutes.

Step 2: The Backend Orchestration (Automated)

The moment the prospect hits "Submit," here's what happens automatically:

1. HubSpot Integration: Contact information flows directly into HubSpot and the prospect is dropped onto an automated nurturing campaign
2. Slack Notification: A flag goes into a dedicated Slack channel alerting the Atomic team that a new Revenue Reality Check has been completed
3. Human Triage Decision: A team member reviews the submission and decides whether to use HubSpot enrichment credits to pull additional company data (revenue size, employee count, tech stack signals)

Critical Filter: This is where Atomic Revenue screens out prospects that are too small (under $5M annual revenue). The AI still delivers value to those smaller companies, but the follow-up conversation shifts to referrals instead of direct engagement.

Step 3: The AI Analysis (1 Hour)

Once enrichment is complete, the AI agent goes to work:

• Multi-LLM Cross-Verification: The agent bounces between Claude, Perplexity, and ChatGPT to verify findings and reduce hallucination risk
• Conversion Stage Scoring: Each area of revenue production (sales, marketing, operations, legal, client onboarding) is rated on a 1-10 scale based on the prospect's responses
• Gap Identification: The AI flags the biggest pain point or expertise gap within the prospect's team
• Scenario Modeling: The AI models multiple improvement scenarios and ranks them by revenue impact

Output: A draft report with 90-day high-priority recommendations.

Step 4: The Human-in-the-Loop Review (30-60 Minutes)

This is where governance meets quality. A strategist at Atomic Revenue:

• Reviews the AI-drafted report word-for-word
• Verifies the recommendations align with industry best practices
• Adds strategic nuance the AI can't provide (e.g., "Given this company's stage, tackle the legal bottleneck before the sales handoff")
• Finalizes the report for client delivery

Why this step is non-negotiable (in Tara's words):

"I personally would never trust something right now going to my clients that I haven't seen or I haven't edited and reviewed very thoroughly. AI can tell you what it finds, but it doesn't tell you what to do about it. You still need an expert to know - do I need more sales reps? Different enablement tools? Better pricing? A better qualifications process?"

Step 5: The Strategic Assignment (Automated)

Based on the diagnostic scores, the prospect is automatically assigned to the right strategist:

• Skillset Matching: If they scored 2/10 in sales and 7/10 in marketing, they're routed to a sales-focused consultant
• Industry Matching: If HubSpot enrichment reveals they're a FinTech company with a sales gap, they're assigned to a strategist with FinTech experience (e.g., "Simon" in Tara's example)
• Calendar Integration: The prospect receives follow-up emails with the assigned strategist's HubSpot meeting link - they can book directly onto the calendar

Step 6: The Prepared Discovery Call (Strategic, Not Diagnostic)

Timeline: 24-48 hours from initial website visit.

What's Different:

• The strategist has already reviewed the diagnostic report
• The prospect has already received the report and identified their priorities
• The database (HubSpot) is fully populated with enriched company data
• The conversation is about implementation roadmap and next steps, not "tell me about your business"

Result: The first call is a high-intent, strategically aligned conversation - not a 90-minute discovery slog.

The Tech Stack: What You Need to Replicate This

For the RevOps Directors and VPs listening who want to deploy a version of this in their own org, here's what Tara recommends as the minimum viable stack:

Core Integration Layer:

• CRM: HubSpot (or Salesforce with similar enrichment/automation capabilities)
• Notification System: Slack (or Microsoft Teams)
• AI Layer: Multi-LLM orchestration (Claude + ChatGPT + Perplexity for cross-verification)
• Voice AI (Optional but Recommended): Conversational interface for the diagnostic questions

Critical Success Factors:

1. Train the AI on YOUR questions: Don't just plug in a generic "sales diagnostic" prompt. Tara spent significant time training Revenue Maven on the specific questions that reveal revenue impact across functions.
2. Build the Governance Layer First: Before you automate anything, define:
   • Who reviews the AI output?
   • What's the QA checklist?
   • How do you route prospects to the right human?
3. Start Small, Test Ruthlessly: Atomic Revenue ran "lots of testing" before deploying publicly. Don't skip this step.

Tara's Warning:

"There's not a quick win if you're going to custom build something. We sent a team member to get additional AI training, allocated focused energy, and did lots of testing. But we also built it to be white-labeled so we can offer it to other companies - because too many people are reinventing the wheel right now."

Translation: You can build this yourself in 60-90 days if you have the technical capacity. Or you can license a white-labeled version of Atomic Revenue's tool and deploy it in 30 days.

The Three Revenue Killers Hiding in Your Pipeline

One of the most valuable insights from this episode wasn't about AI at all - it was about where revenue leaks in B2B pipelines.

Tara has delivered revenue diagnostics across dozens of industries, and she's identified three consistent bottlenecks that kill conversion rates and lifetime value:

Revenue Killer #1: Sales Handoffs

Anytime you hand a deal from one team to another (Sales → Sales Engineer, Sales → Pricing, Sales → Ops), you risk losing momentum and context.

The AI Solution: Track conversion rates at every handoff stage and model the ROI impact of improving each one.

Revenue Killer #2: The Legal Bottleneck

This one surprised me - but Tara says over 50% of enterprise revenue diagnostics flag the legal review process as a major drag on revenue.

The Problem:

• Over-zealous redlining by corporate attorneys that kills deals
• Slow, cumbersome contracting processes that add weeks to close timelines
• Lack of due diligence that causes problems post-sale (impacting account management and lifetime value)

The AI Solution: Build a contract review agent that flags high-risk redlines vs. low-impact legalese, speeding up approvals without sacrificing protection.

Revenue Killer #3: Client Onboarding Gaps

The handoff from Sales to Operations (post-signature) is where customer experience - and future upsell opportunity - lives or dies.

The Problem: Slow onboarding, dropped context from the sales process, or lack of clear success metrics early on.

The AI Solution: Automate onboarding workflows and use AI to summarize sales commitments/context for the ops team.

Why "Human-in-the-Loop" Is Non-Negotiable (The Governance Layer)

Throughout this episode, Tara repeatedly emphasized that AI drafts, but humans finalize - and for good reason.

The Risks of Letting AI Run Unsupervised:

1. Hallucination: Even with multi-LLM cross-verification, AI can still produce incorrect or misleading outputs
2. Brand Risk: An AI-generated report that sounds "off" or makes a bad recommendation damages your credibility
3. Strategic Nuance: AI can tell you what it found, but it can't tell you what to do about it - you still need human expertise to prioritize and sequence the fixes

Tara's Governance Model:

• Multi-LLM Verification: The diagnostic agent bounces between Claude, Perplexity, and ChatGPT - they verify findings against each other and pull forward the best result
• Expert Review: A strategist reads the AI-drafted report word-for-word, edits for accuracy, and adds strategic recommendations
• Client Delivery: Only after human QA does the report go to the prospect

The Payoff:

Quality and speed both win. The AI handles the data processing and draft generation (saving 60-75% of labor). The human adds the strategic layer that turns "interesting findings" into "actionable roadmap."

The Business Model Shift: From Cost Center to Revenue Engine

One of the most brilliant aspects of Tara's Revenue Reality Check workflow is how it repositioned diagnostics within her own business model.

The Old Model (Cost Center):

• Schedule a 90-minute discovery call
• Spend weeks analyzing data manually
• Deliver a one-page recommendation that felt like a sales pitch
• Slow qualification = high cost-of-sale with no project attached yet

The New Model (Revenue Engine):

• Prospects self-serve the diagnostic 24/7 on the website
• AI delivers value in 24 hours (building trust immediately)
• Strategists only engage after the prospect is pre-qualified and pre-educated
• First meeting is about implementation roadmap, not "tell me about your problems"

Result: The Revenue Reality Check became a lead qualification engine that delivers value, builds trust, and eliminates 90% of the expensive "pre-proposal" sales cycle.

And because it's white-labeled, Tara can now offer it to other agencies and RevOps teams - turning a internal efficiency tool into a scalable product offering.

Key Takeaways for Revenue Leaders

✨ The 24-Hour Diagnostic Framework

Tara architected a Revenue Reality Check agent that's white-labeled, HubSpot-integrated, and delivers qualified pipeline in 48 hours (not 3 weeks). The workflow combines AI analysis with human governance, auto-assignment to the right strategist, and direct calendar booking.

✨ When AI Outperforms the Expert

Tara's AI agent found a $21M opportunity (3% conversion lift) that her manual analysis missed ($10M with 2% lift). Why? The AI could consume more data, cross-reference financials with conversion rates and pricing, and model multiple scenarios in parallel - calculations that would take a human days to complete.

✨ The Human-in-the-Loop Model

Atomic Revenue governs the workflow so AI drafts the report, strategists QA/finalize it, and quality + speed both win. Multi-LLM cross-verification (Claude ↔ Perplexity ↔ ChatGPT) reduces hallucination risk, and expert review ensures strategic recommendations are actionable.

The Audit Question: Where Should You Start?

If you're a Revenue Leader listening and you want to audit your own "Admin Drag" and figure out where a diagnostic agent could help, Tara recommends starting with handoffs and legal bottlenecks.

Pull these metrics first:

1. Conversion rates at every sales handoff stage (Sales → Sales Engineer, Sales → Pricing, Sales → Ops)
2. Average time-in-legal-review for enterprise contracts (and win rate before vs. after legal involvement)
3. Client onboarding cycle time (signature → first value delivery)

Then ask: "If we improved conversion at [this stage] by 3%, what would that be worth in annual revenue?"

That's the question that unlocks the $21 million opportunities hiding in your pipeline.

How to Experience the Revenue Reality Check Yourself

Tara built this tool to be experienced, not just explained.

If you want to see how an AI diagnostic agent works (and get a 90-day roadmap for your own revenue operations), go to atomicrevenue.com and click the Revenue Reality Check link.

It takes 10 minutes. You'll get a report in 24 hours. And you'll see firsthand how an AI agent can analyze your pipeline and surface opportunities you might be missing.

Resources & Links

• Try the Revenue Reality Check: atomicrevenue.com
• Connect with Tara Kinney on LinkedIn: linkedin.com/tarajkinney
• Download the Executive Guide to Shadow AI: theaihat.com/shadow-ai

About The AI Hat Podcast

AI for Revenue Leaders: The AI Hat Podcast helps VPs of Sales, CROs, and RevOps Directors stop drowning in Admin Drag and start building the Digital Crews that scale revenue. Hosted by Mike Allton (Director of Partner-led Growth, Agorapulse), each episode deconstructs real workflows from practitioners who've deployed AI agents with measurable P&L impact.

Subscribe: Apple Podcasts | Spotify | YouTube

Full Transcript

(lightly edited)

How a RevOps CEO Identifies a $21M Revenue Opportunity in 24 Hours (Not 3 Weeks)

How a RevOps CEO Identifies a $21M Revenue Opportunity in 24 Hours (Not 3 Weeks)

[00:00:00] Tara Kinney: Time in the selling process has gone down drastically. I mean, it's less than 10% of what it was before to get to the same level of agreement between someone on our team and a prospect. So we're saving about 90% on that sales process from the time someone finds us and does the revenue reality check until they get through the first meeting, and we're talking about the next steps and making a proposal. And that's huge because until you're making a proposal, that's all cost of sales that has no real project tied to it. So we've eliminated one of the more expensive parts of most organizations selling process.

[00:00:51] Mike Allton: Mike's got The Hat, the AI kind. Unlocking power in business, you'll find. Cut through the hype, explore the tech. Every week, new ideas to check.

[00:01:15] Allton leads us on a digital quest. AI secrets, he'll always invest. Tune in tight, don't miss a beat. Knowledge, groove, keep you on your feet. Listen up! To the AI flow.

[00:01:38] Business growth, here we go. Cuttin' edge insights, that's the plan. With The AI Hat you'll understand.

[00:02:00] Greetings program. Welcome to AI for Revenue Leaders, The AI Hat Podcast. And whether you lead sales, marketing, or partnerships, your team was hired to drive growth, not to drown in busy work. And each week we help you fire the admin drag that kills momentum and build the digital crews that scale your revenue.

[00:02:18] I'm your host, Mike Allton. Let's get to work. And look, let's get real. Your sales team is missing quota and you've got no idea. Why is it the pipeline? The messaging? Stage three conversion rates? So you call in a consultant, you ask for a revenue diagnostic, and then you wait two weeks, three weeks. By the time the report lands, your quarter is over, and that deal you lost, already closed with a competitor.

[00:02:44] Here's the hard truth. The traditional diagnostic process is too slow for modern B2B velocity and too shallow. You get surface level insights when what you need is scenario modeling that shows you where the money is hiding. My guest today blew that model up. Tara Kinney is the founder and CEO of Atomic Revenue. A Rev Ops management consultancy and she supported 76 companies across eight industries and LED 18 CRM implementations. She knows what normal looks like in revenue ops, and her agency used to take two to three weeks to deliver diagnostics Now. Her digital crew does it in 24 to 48 hours, and it's not just faster, it's smarter.

[00:03:28] Her AI agent found 21 million in revenue opportunity that she, a seasoned consultant had missed in her manual analysis. And this isn't just a chat bot writing emails. This is an agent. Workflow that analyzes your pipeline. Models improvement scenarios, ranks them by ROI and delivers a report while you sleep.

[00:03:48] Hey Tara, welcome to the show!

[00:03:50] Tara Kinney: Hi. Thanks for having me, Mike. I'm excited to talk about AI with you.

[00:03:55] Mike Allton: I am so thrilled to have you on here and learn from you and, and figure out what it is that exactly you did. But before we get into that, before you built that agent, what did your revenue diagnostic process look like?

[00:04:06] You know, what was it like before and why was it a problem for both you and your team and your prospects?

[00:04:15] Tara Kinney: For, for about eight years, revenue diagnostics has been the primary tool that atomic revenue uses to analyze. All of the revenue opportunities and impacts within an organization. It's very cross-functional and it would take probably an equal amount of time to analyze the information. And we had to limit the amount of information we requested from clients because it was all human digging through documents and tech stacks, and scorecards trying to find golden nuggets. Um, and then after we find the things that we want to present and the opportunities we discover, we spend the other half of the time we have to build a client figuring out how to put it into some kind of a report package that makes sense, that can be consumed and implemented by our client. And those two pieces, as more technology has come to play and the selling process has gotten more complex or with larger organizations, there's a lot more people involved and a lot more steps to the process.

[00:05:31] It became longer, but also nearly impossible with a human eye to dig through that much information and find the high value, high priority opportunities. So AI really came in and allows us to save labor on the two pieces that don't actually need a human. And the human in the loop then gets to apply their expertise to what AI discovers and modify deliverables that AI has drafted.

[00:06:07] Mike Allton: All right. Well, I love hearing that, but for some of the revenue team members and leaders, maybe even agency owners that, that are listening and maybe have not gone through this, walk us through like a specific example. Prospect reaches out, asks for revenue health check.

[00:06:20] In the old world, what happened next and, and how long was it before you actually delivered value?

[00:06:25] Tara Kinney: Yeah, so we have two different AI tools that we use. Uh, the revenue reality check on the website is our current quick AI tool with a 24 hour turnaround. It actually is about an hour turnaround, and it provides 90 day high priority recommendations.

[00:06:46] It only takes someone about 10 minutes to complete that process. Historically, we used to do a revenue operations assessment, so you would reach out to atomic revenue, we would schedule a meeting with you. We'd spend 90 minutes talking through lots of information and digging through all of these details.

[00:07:06] And at the end of it, you might get a one page notes worksheet that's been completed with a recommendation of how we could sell you services to fix your problems, which was not the intent, the purpose of businesses to help other businesses grow. That was why Atomic Revenue was founded, but that's way too long for a company to wait and it feels like a sales pitch, not like help. So with, uh, the Revenue Reality Check, now people can just go to our homepage. Anytime of day 24 7, they go talk to Revenue Maven, our AI personality, who also has a very charming AI voice. And she can walk them through a series of questions or they can type in the answers to the questions.

[00:07:59] They hit submit and then within 24 hours they have a report in their inbox of the high priority opportunities for the next 90 days. Um, and then with that, they can decide how they want to proceed. Now that you have a map, you can implement it with your team and talk about how to get resources, or there's a link in there to schedule a call with someone from Atomic Revenue if you don't have the resources in house to do what's recommended, and you need a little additional support.

[00:08:36] Feels much less like selling and a lot more like helping.

[00:08:40] Mike Allton: Right, right. And, and we're gonna get back to that process in a moment. 'cause I, I wanna learn, you know what then happens next, but going back to, like that moment when you decided that, oh. I don't know. I dunno what the decision was if we need to implement AI or what, but I I want to bring us to that moment.

[00:09:00] When you realized you did need to architect it differently, did you just think, wow, we should use AI agents? Or was there a specific deal that you lost or what was it that caused you to make that shift and something you'd been doing a certain way for a long time?

[00:09:16] Tara Kinney: Yeah, the big part is scalability of humans.

[00:09:19] We only have so much human capacity, and one of the things that's been very difficult at Atomic Revenue is that I am not scalable. And so we might have a lot of leaders in a lot of technical discipline areas, but they don't necessarily know the questions to ask in a different part of the business that is detrimental to revenue. And training people to ask all the right questions and to be able to confidently navigate those conversations is very challenging. So by bringing in the AI agent, we are able to do two things. One is. AI supplements the areas of expertise in revenue impact that a particular senior consultant might not have.

[00:10:09] If they're a sales leader, they might not be able to ask marketing questions, and vice versa. So this sets a nice baseline, and the consultants then get the report. To prepare for a conversation with a client and the report goes to the consultant that's most aligned with the needs of the prospect.

[00:10:32] So if a prospect scores a two out of 10 in sales and a seven out of 10 in marketing, they're going to be assigned to a sales leader to review their report and talk about what some of the gaps are. That has been hugely helpful in the sense that when you're going to meet with a prospect for the first time, you don't know what resources they have in house and where the problems might lie, where the weaknesses are.

[00:11:05] Um, so it really helps us match them appropriately with a good strategy consultant at that early stage. We still do revenue diagnostics separately, which is a much deeper dive than the revenue reality check. But it gives you a good starting point of understanding, you know, where the business is recognizing it's failure already.

[00:11:31] Mike Allton: That is fascinating to me. 'cause you're right, that's that the layer of. Information and then strategy that I, I don't often see in any kind of agency, in any kind of world, you know, where you're, you're mapping the client's strengths and weaknesses against what you can offer them. Let's talk about that.

[00:11:51] $21 million story. You ran a manual diagnostic. The thing that, again, that you've been doing for years, you found a $10 million opportunity, which sounds fantastic, 2% improvement, but then your AI, I guess, ran the same analysis and found over twice that 3% lift on a different metric that you weren't even tracking.

[00:12:10] What did the A IC that you had missed

[00:12:14] Tara Kinney: so. It's interesting how much more information you can put into an AI digester for it to look at than what a human could possibly review. We can consume twice as many proposals, much more in-depth reports, and AI does a better job of cross-referencing financials with conversion.

[00:12:40] Rate analysis with pricing. So in the AI tool, we were able to dump in a lot more information. That's one reason why it was able to find something that we didn't see. But even if we didn't do that, it takes a lot of analysis and calculation to figure out what is the value of improving this. Conversion rate at this stage.

[00:13:10] Mm-hmm. And so when you have an 18 or 24 month sales cycle and you have nine or 12 conversion points over that timeline, and each one of those is a loss of a portion of your deals. It is easy to see what the conversion rate is. It's much more difficult to mathematically model what it would mean to change that conversion point.

[00:13:41] Because at some of those stages you don't know price or value of a customer yet. And so there's a lot of math involved. I wasn't terrible at math, but that's not why I do what I do. Uh, so to help the management team and really provide some better insights, it's nice to have that additional calculating capacity of ai and that's really where the secret was.

[00:14:08] And this was low hanging fruit. In that particular analysis, AI and I agreed that if you invested in lead generation and increased leads by 30% at the end of the 18 month sales cycle, you would have 0.25 additional projects. Well, what's 0.25 projects worth to you in ROI On. How much does it cost to increase leads by 30%?

[00:14:39] So we immediately started looking for more cost effective and lower hanging fruit opportunities. And over 18 months, this company had hundreds of deals in its pipeline that were all stalled out at various stages or not advancing for different reasons, and both the human analysis and ai. Identified that the middle of the pipeline around conversion points five through seven is where the real revenue opportunity was.

[00:15:15] And you really shouldn't even invest in additional lead generation until you fix those conversion points. Otherwise, you're kind of burning cash.

[00:15:27] Mike Allton: Wow, that is fascinating. And it also makes my head hurt because mathematics is not my strong suit at all. I actually went to college for computer science and had to stop and transition to history 'cause I just couldn't get through statistics.

[00:15:40] Folks were talking with Tara Kenny about how AI is being deployed for agencies and clients to not just save time, but actually identify revenue opportunities. And I've got several more questions you're not gonna wanna miss, but first. A quick word about how to talk to your team about Shadow AI.

[00:15:56] Right now, your best people are drowning. They're buried in data entry, meeting prep, and internal reporting, the admin drag that kills momentum. To keep their heads above water. They're turning to AI tools you haven't approved. They aren't trying to be reckless. They are trying to be efficient. They are trying to hit the goals you set for them, but this Shadow AI creates massive blind spots in your data security.

[00:16:24] You can't afford to ignore it, but you also can't afford to ban it. You need to architect it. I've created the executive guide to shadow AI to help you bridge this gap. It's a blueprint for turning a security risk into a competitive advantage through proper governance. Don't fight the shadow, sanction it.

[00:16:47] Download the free guide today theaihat.com/shadow-ai. That's theaihat.com/shadow-ai.

[00:17:00] Okay, Tara, let's get tactical. When a prospects, they're on your site, they request the free Revenue Reality Check and, and they put in all that information that you ask for, what happens?

[00:17:09] Next, walk me through the the agent's job step by step.

[00:17:15] Tara Kinney: So like any Ai, you have to have good data and you have to train it for what you want it to look for and what you want it to produce. That takes time. So it did take us a while to get. The whole system to work, but it has to be integrated within your business.

[00:17:36] So for us, when someone goes and completes the revenue reality check, their contact information goes directly into HubSpot. They're dropped onto an automated nurturing campaign. A flag goes into Slack for us to know that a revenue reality check has been completed in a dedicated Slack channel. Um, there's a human element opportunity there where you get to review the submission and determine whether or not you want to use your.

[00:18:11] HubSpot enrich credits to enrich that contact and get more information about that company. Um, and that lets you know whether or not they're qualified for your services. So this tool helps us screen out companies that are really too small to work with Atomic Revenue, and it tells us that when we talk to them or when we follow up, we're gonna be providing them with a referral to someone who serves businesses their size.

[00:18:41] And when I say too small, I'm saying under 5 million annual revenue. Um, so if companies are really small this tool still gives them stuff they can work with. It delivers value, but our follow-up conversations are go in a different direction. So when we enrich the contacts, then our HubSpot is fully populated with all of the data that we would need and leverage in Ai then to look at how their performance might relate to their peers as well as it tells us who we should assign them to. So I talked before about how we rate on a scale of one to 10 different areas of revenue production based on. Their interaction with our AI agent, and when that comes back, it gives us an idea of where their biggest pain point is or where their biggest gap is within their team as far as expertise.

[00:19:49] So we can pair them with a skillset, but then once we enrich the contact, we can also pair them with a sales rep. And when I say sales rep, I actually mean a fractional executive consultant, so mm-hmm. We don't have salespeople. So that's one area where the AI agent allows us to save money, and it also gives the prospect a better experience 'cause they're talking to someone now who's not only filling an expertise gap within their business, but also might have industry experience. So if they come in and they're from FinTech and they've got a sales need. It has automatically routed them to Simon and they get their report, they get their follow-up emails to book time with Simon, and then they're using Simon's HubSpot meeting link so they can put time right on his calendar.

[00:20:44] So within 24 hours they've gotten a report. They've been assigned to someone who's qualified not only by skillset, but by industry. And Simon's been notified. The database has been populated. They have a report, Simon has a copy, and whenever they get to meet, everyone is prepared and well aligned for what that conversation is going to be about.

[00:21:11] That should in most cases shorten the sales cycle fairly significantly, so long as the person who completed the revenue reality check is a decision maker. But even if it's not. You're building trust within the organization and you can get to the decision maker faster, um, with all of these details in place.

[00:21:33] And with that, it only took the person 10 minutes to fill out the, or to engage with the AI agent. And Simon just got a full report on a prospect with more information than you could get as a sales rep. Uh, really any other way. As a sales rep, you just spend all your time researching and trying to figure out, how can I talk to this person?

[00:21:58] What are they going to need? You spend half your call with them trying to diagnose the opportunity, and in this case, you have the diagnosis of the opportunity and a lot more information before you ever even get on a call.

[00:22:15] Mike Allton: That's exactly right. We just spent, uh, our entire, you know, third episode of this season talking about, you know, creating an AI agent for that kind of purpose where, you know, you wanna try to get on a call with somebody, or you've got a cold call scheduled, you never talked to 'em before.

[00:22:29] So you're having the agent do some initial research and discovery for you, which is nice. But obviously this is 10 times better because the customer, the prospect, is already providing you with a lot of that information willingly, uh, along with the diagnosis. And, and we're gonna break down for those listening, how this is built 'cause I know that's gonna be your next question, but before we get to that, I, I kind of wanna underline something 'cause I love how you've got this human in the loop aspect to the workflow. The AI drafts the, the report, but you guys, you're strategist, review it, you finalize it. I, I want to hear from you why you think that step is so critical and what would break, do you think, if you just let the AI send the report directly?

[00:23:11] Tara Kinney: Okay. Well, we all have heard of AI hallucination and from our experience, it takes a lot of work, a lot of repeat runs of tests in order to try to eliminate the hallucination. So on the revenue reality check, it's a much lighter tool and doesn't leverage as much AI complexity on the backend. However, it's also a much shorter report and it's looking at much less information, so we do appreciate that it's easier to consume and review, so you can still have that quick turnaround time.

[00:23:53] On revenue diagnostics, we actually built the tool so that different AI tools are checking against each other, so it'll bounce between Claude to Perplexity to Chat GPT. So they are verifying the trust in the information off of each other and then pulling forward the best result. And with that, AI can tell you what it finds, but it doesn't tell you what to do about it.

[00:24:30] You still need an expert to know what do I do now that I've identified a $21 million opportunity with in my pipeline? Do I need more sales reps? Do I need different sales enablement tools? Do I need a better qualifications process? Do I need better pricing? You, there's so many things within that conversion point to look at.

[00:24:58] You still have to have the experience and the expertise of having gone through these situations dozens of times to help isolate within this particular company. What is the order in which you should start to address and fix the opportunity that AI discovered? I personally would never trust something at right now.

[00:25:24] I might in the future, but right now I don't trust things going to my clients. That I haven't seen or I haven't edited and reviewed very thoroughly. Even when I'm using Chat, GPT or you know, a custom chat type of tool, I'm reviewing it word for word to make sure that it hasn't put something in there or changed the language in a way that doesn't reflect my expertise or is going to guide the recipient in a direction that I wouldn't recommend. So I don't think we're there yet with AI to trust it, to just run on its own.

[00:26:11] Mike Allton: Yeah, no, that's absolutely right. That is essential advice. Folks. Go back and listen to that one more time and, and kind of delve it because it's, it's not just the possibility that the AI can make a mistake, which it can, because remember folks, the AI was trained on things like Reddit and YouTube, which are not necessarily valid sources of information. But more importantly than that. It could decide to respond or go in a particular direction, in a way that doesn't align with how you would go, or the choices that you would make as the business owner or the sales rep, whoever it is that use it, is using the tool at that particular moment.

[00:26:49] So that's essential. And that's a great segue 'cause I, I wanna know about the rest of the tool stack. For Rev Ops, leader directors, they're listening, they want to kind of replicate this themselves. What do you think is the, the MVP that they could deploy in 30 days?

[00:27:06] Tara Kinney: That's a great question, Mike.

[00:27:07] And one of the things about being a good leader or a good CEO is that you don't build something that you don't have the expertise to build And I am not a technical expert. I am not a software developer, and it's one of the things that held me back from being able to solve some of these challenges previously.

[00:27:29] You know, I'm much more on the human side and on the analysis side, so. We have a team member and as an organization, we sent that team member to get some additional AI training and we had very specific things that we wanted to accomplish as an organization and we allocated, you know, a focused amount of energy to be able to accomplish the goals.

[00:27:58] And then we did lots of testing before we deployed it out publicly. So I do wanna say there's not a quick win if you're going to custom build something, which is what we did. But I can tell you that the way that we built the Revenue Reality Check, we built it to be a white labeled solution so that now we can offer that to companies and they can put their own questions in and connect it to their own tech stack, and it can be redeployed because there's too many people out right now.

[00:28:33] Reinventing the wheel, and it's causing a lot of turnover in AI tech stack. So people are adopting tools that in three months go away, or they change so drastically that the reason why you chose it to start with is irrelevant now with where the tool is today. So in a really fast paced environment like AI right now, we do have to be careful that we don't overinvest in developing things that are going to become obsolete and that we don't deploy AI tools within our tech stack that don't have.

[00:29:17] Proven longevity and the don't have the backing to last. Some of them are very light on the technical build and heavy on the marketing and sales promotion. So we're still looking for those vaporware elements within the AI world right now, and they're really hard to find.

[00:29:40] Mike Allton: Yeah, that is so true. You have to be careful about which tools you're investing in.

[00:29:45] Even the, you know, the big frontier models, you never know where they're going to go and you don't have to look too far back in history. Look at Google plus some of you maybe haven't even heard of Google Plus listening. Tara, I know you have. That was a huge, massive social network that was run by Google that they just decided in 2015, eh, this isn't working too well for us.

[00:30:05] We're gonna stop it. That was a major issue. Uh, and, and so these things can happen. These big companies can make decisions on a dime it feels like. So we do have to be very careful about that. But what changed after you deployed it? Can you share some hard numbers? How much faster, uh, is qualification or how much like capacity did your team reclaim?

[00:30:25] What's the conversion rate? Maybe from the free reality check to qualified opportunity? How did that help you?

[00:30:31] Tara Kinney: Yeah, so time in the selling process has gone down drastically. I mean, it's less than 10% of what it was before to get to the same level of agreement between someone on our team and a prospect.

[00:30:47] So we're saving about 90% on that sales process from the time someone finds us and does the Revenue Reality Check until they get through the first meeting, and we're talking about the next steps and making a proposal. And that's huge because until you're making a proposal, that's all cost of sales that has no real project tied to it.

[00:31:15] So we've eliminated one of the more expensive parts of most organizations selling process and, and then the other piece is that as business becomes more complex and as we wanna analyze more, businesses don't wanna pay more hours for a human to do more things. So on the delivery side of services, with revenue diagnostics, we're able to keep our pricing close to where it's been for years, but elevate the quality of the product that we're producing, as well as the volume of information that's going into making much higher valued recommendations. So we see that as a big win that, our clients don't have to pay more. Our consultants don't have to spend as much time doing the low value work of formatting presentations and, you know, digging through a stack of documents in a drive somewhere trying to find the golden nuggets.

[00:32:25] So that particular piece, it's cut at least 60, if not 75%. It's still really new. You know, we are just now launching the revenue diagnostics fully in this AI environment. Um, so that piece, we still need to do some more measuring to figure out exactly how much labor it's saving, but we're expecting by the end of Q1 we'll be able to have some firm metrics around how profit margins for us are impacted, how pricing for clients.

[00:33:01] Is impacted so we can keep prices low as well as capacity of consultants. And that I can tell you right now, we are expecting a single strategist to be able to do three or four revenue diagnostics in a month instead of previously it would take us two months and a consultant couldn't have more than two revenue diagnostics going in a single month.

[00:33:29] Mike Allton: That is amazing. My last question for you, Tara. I'm just wondering, for the revenue leader, they're listening right now, if they wanna audit their own admin, drag within their own organization, figure out what their diagnostic agent could actually help with, what do you think is the first question they should ask or the first metric that they should try to pull?

[00:33:51] Tara Kinney: That is a loaded question, Mike. Having done so many of these over the years there are some consistent themes that we find, but it's still not a particular metric or a particular place, I would say. There are two areas that we have identified fairly consistently across all industries and companies where revenue is severely impacted.

[00:34:22] One of them is at client onboarding or any handoff within the sales process. So if sales is handing off to a sales engineer or someone technical to do pricing and bidding. We see a big conversion rate loss there that companies can work on. We see big conversion rate opportunities at the handoff from sales to operations when it, you know, they've signed the contract and they're actually going into the process.

[00:34:50] We see issues that impact account management and lifetime value of customers in a really significant way, um, that are solvable at that stage. And the other one that a lot of people don't think about, but we see a lot in enterprise sales is legal's involvement or engagement in the sales process. And there's been more than 50% of the revenue diagnostics that we've delivered where we have had to flag the legal process as a major impact on revenue production, either over zealous redlining by corporate attorneys that causes, you know, lost sales or a contracting process that takes too long and is too cumbersome or lacks due diligence value that ends up causing problems later, and again, if it doesn't impact conversion rate.

[00:35:53] For new clients, it 100% impacts account management and lifetime value. So that legal involvement in the revenue process and then any handoffs that people have, especially sales to operations for client onboarding. Those are the big ones.

[00:36:13] Mike Allton: That is fascinating. I know everyone listening, their minds are buzzing with, oh my, we, there's a lot that we need to look at.

[00:36:21] You've been terrific, Tara. If they've got questions and they wanna reach out to you, or they wanna try this diagnostic, where should they go?

[00:36:27] Tara Kinney: Yep. If you go to atomic revenue.com you can find the diagnostic on the homepage. You can watch a little video about it. But mostly you can just click on it and go into the diagnostic process.

[00:36:42] You can find me on LinkedIn if you just wanna follow the story and see some of the conversations that we're having. Um, but if you want to engage, go to atomic revenue.com and just click on the revenue reality check link and jump into it. Experience it.

[00:37:03] Mike Allton: Amazing. Thank you so much, Jared. Thank all of you for listening. We will of course, have all those things in the show notes below, and if you haven't yet talked to your team about how they're really. Using AI shadow ai, it's time. Learn what shadow AI is, why you should lean into it, and how to proceed for free theaihat.com/shadow-ai.

[00:37:23] Check it out and I will see you in our next episode. Welcome to the grid.

[00:37:28] That's a wrap on another episode of The AI Hat Podcast. I hope you've gained some valuable insights and actionable strategies to boost your business with AI. If you enjoyed this show, please subscribe, leave a review, and share it with your network. And don't forget to visit the ai hat.com for more resources and to connect with me.

[00:37:46] Until next time, keep innovating and remember, the future of business is human and it's powered by AI.