GrowCFO Show

.entry-img img{
display:none !important;
}
.single .hentry .entry-img{
display:none !important;
}

https://open.spotify.com/episode/0rBRhSM8E5kLRjNRvQlG0w

Reliable forecasting is essential to sound financial planning, but AI cannot compensate for unreliable data or broken workflows. Carolina Lago, Managing Director at Tactic Financial discusses why CFOs should first establish clear data flows, a single source of truth, and dependable processes before layering AI into forecasting. That groundwork helps make forecasts more consistent, verifiable, and useful for decision-making.

The conversation outlines a practical model: use established tools and rules for repeatable calculations, apply AI where analysis and scenario exploration benefit from it, and keep people involved to supply business context and review outputs. With those foundations in place, AI can help finance teams analyze more forecast drivers and test more scenarios—not simply produce results faster.

About Caroline Lago

Carolina Lago has spent more than 25 years building financial models and leading FP&A teams across Latin America, North America, and Europe. She now runs Tactic Financial and The Tactical Room, where she teaches finance professionals how to actually put AI to work: Python in Excel, AI agents, and her Three forces to AI framework for producing repeatable reliable results.

Her focus is the unglamorous middle of AI adoption, the part after the demo and before the rollout, where most finance teams stall. She runs a cohort-based AI Accelerator for FP&A teams, writes The Tactical Room newsletter, and speaks internationally, including at the Global Excel Summit in London and the AFP annual conference. She is a member of the Financial Modeling Global Leaders Council.

Carolina is based in Basel, Switzerland, and teaches from the trenches, not the textbook.

Key topics covered:

CFOs should prioritize process design before selecting or adding AI tools.

Finance teams need to map data flows, clarify ownership, and create a reliable source of truth.

Repeatable calculations and data-cleaning steps should remain governed by dependable rules and tools such as SQL or Power Query.

AI is better suited to analysis and exploring possibilities than to recreating deterministic calculations each time.

A human-in-the-loop workflow can add business context and help identify errors or misleading outputs.

Once the foundations are sound, AI can help teams examine more line items and test a wider range of scenarios.

Links

Carolina Lago on LinkedIn

Kevin Appleby on LinkedIn

GrowCFO Mentoring

Timestamps: 

00:02:55 — AI should build on functioning processes rather than replace them. 

00:06:16 — Establish data flows, a single source of truth, and clear process responsibilities.

00:09:14 — Use structured workflows and rules for data cleaning instead of relying on uncontrolled prompts.

00:15:41 — The three-part workflow: deterministic processes, AI analysis, and human judgment.

00:20:11 — AI can help finance teams investigate forecasts more deeply and explore more possibilities.

00:27:20 — Design the workflow first, then identify where AI can add value.

Find out more about GrowCFO

If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode.

GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here.

You can find out more and join today at growcfo.net

.entry-img img{
display:none !important;
}
.single .hentry .entry-img{
display:none !important;
}

https://open.spotify.com/episode/1akR1CIZLVRQ89FSp1hAZe

As finance teams face mounting pressure to adopt AI, CFOs need a practical starting point – not a rushed, organization-wide rollout. Gavin McGahey, CTO & Co-Founder at AccountsIQ, examines the gap between finance professionals’ interest in AI and their need for confidence in its accuracy, security, and business value. The episode references a survey of more than 1,000 senior finance professionals and reports that 81% mostly or fully trust AI-driven outputs.

McGahey recommends beginning with limited, measurable use cases, such as accounts payable automation and anomaly detection. He also stresses the importance of well-structured data, careful review of vendors’ data-protection practices, and human oversight of consequential actions. A deliberate approach can help finance leaders test AI’s potential while maintaining review and audit controls.

Key topics covered:

Finance leaders see AI’s potential but need confidence in its accuracy, data handling, and verifiability.

Initial AI projects should be small and matched to a finance team’s readiness.

AP automation and anomaly detection offer practical starting points for saving time and flagging unusual transactions.

Well-structured data is essential for useful AI analysis; poor inputs can lead to convincing but incorrect results.

Focused AI agents can assist with finance workflows while people retain control over approvals and posting.

Finance teams should examine vendors’ data practices, security documentation, and audit trails before adopting AI tools.

Links

Gavin McGahey on LinkedIn

Kevin Appleby on LinkedIn

GrowCFO Mentoring

Timestamps: 

00:31 — Finance professionals see AI’s potential but feel pressure to adopt it quickly.

04:11 — Questions to ask about AI solutions, including data use, transparency, autonomy, and traceability. 

05:40 — Why well-structured data matters when using AI tools for finance analysis.

09:26 — Start small, choose carefully, and adopt AI at the organization’s own pace.

11:17 — AP automation and anomaly detection as practical early use cases.

20:21 — AI agents supporting finance workflows while users retain control.

31:14 — Reviewing data-protection documentation and supplier practices.

35:03 — Building on sound finance processes when introducing AI.

Find out more about GrowCFO

If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode.

GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here.

You can find out more and join today at growcfo.net

.entry-img img{
display:none !important;
}
.single .hentry .entry-img{
display:none !important;
}

https://open.spotify.com/episode/6kI5GE0gJB2IBBoEGtAWJm

As organizations scale AI, CFOs must balance faster, more efficient finance operations with cost control, reliable decision-making, data protection, and accountability. In this episode, the discussion frames AI governance not simply as a compliance checkpoint, but as a performance system—one that can help organizations adopt AI with confidence while supporting business value and innovation.

Bruce Mullan, Managing Partner, AI Governance Partners describes how AI adoption in finance can progress from administrative assistance, to process automation, and eventually to automated decisions. Using invoice processing as an example, he explains how governance can be matched to risk: straightforward use cases can receive lighter review, while higher-risk applications warrant more scrutiny. He also emphasizes evaluating whether AI is the right solution in the first place.

For CFOs, scaling AI also means understanding where employees are already using it, choosing tools appropriate to the task, protecting confidential information, and weighing the cost of AI against its value. The conversation highlights AI literacy, human oversight, and post-launch feedback as practical ways to monitor reliability, errors, cost, and unintended consequences – without rushing simply to catch up.

Key topics covered:

AI governance can be designed to enable performance and innovation, rather than operate only as a compliance gate.

Finance teams can progress from AI-assisted administration to automated processes and, in some cases, automated decisions.

Governance can be proportionate to risk, with lighter review for low-risk use cases and fuller assessment for higher-risk ones.

CFOs need visibility into “shadow AI”- tools employees may already be using outside formal processes.

AI literacy, human review, and feedback after deployment help teams assess accuracy, reliability, cost, and unintended effects.

A measured, steady approach can be more effective than rushing into AI adoption out of fear of falling behind.

Links

Bruce Mullan on LinkedIn

Kevin Appleby on LinkedIn

GrowCFO Mentoring

Timestamps: 

00:00 — Kevin Appleby introduces Bruce and frames AI governance as a performance system, not just a compliance system.

12:16 — The discussion turns to AI governance as a way to support value and innovation while avoiding organizational bottlenecks.

18:22 — Bruce outlines stages of AI use in finance, illustrated through invoice processing.

23:43 — Risk-based review and asking whether AI is actually needed for a use case.

27:00–30:20 — AI literacy, human oversight, feedback loops, and monitoring performance after deployment.

34:35 — Advice for CFOs who feel behind: avoid panic and take a measured approach.

37:06 — Shadow AI, employee use of public tools, and the challenge of protecting sensitive information.

45:14–49:21 — AI costs, tool accuracy, and why confidential information should be handled cautiously.

Find out more about GrowCFO

If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode.

GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here.

You can find out more and join today at growcfo.net

.entry-img img{
display:none !important;
}
.single .hentry .entry-img{
display:none !important;
}

https://open.spotify.com/episode/2TUU4Nmtl7w7NGfAz2nzdf

A product’s strong gross margin does not always translate into strong profit. When support, sales, customer service, and other indirect costs are left out of product-level analysis, a high-margin product can consume more resources than it generates. In this episode, Pauline Healey explains why businesses need to look beyond headline margins to understand the true cost of serving each product and customer.

Pauline Healey, founder of Logical BI, discusses how her fractional CFO work combines finance with operational, supply-chain, and inventory expertise, particularly for manufacturing and engineering businesses. She explains that reliable cash-flow forecasting must account for the full operating cycle—from purchasing stock and managing production to delivery and customer payment.

The conversation explores how businesses can improve profitability by investigating operational waste, checking that completed work is invoiced, and examining product costs more closely. Pauline and Kevin discuss how a product with an apparently attractive margin may still reduce overall profit if it drives unallocated costs such as customer support and additional sales effort. They also cover business systems, data consistency, and Pauline’s use of AI and her Profit Harmony Hub membership platform.

Key topics covered:

Pauline’s background in corporate finance and operations led her to build Logical BI around fractional CFO support for manufacturing and engineering businesses.

Supply-chain complexity affects cash flow: forecasts need to account for inventory, supplier payments, production, delivery, and customer payment timing.

Businesses should assess whether existing systems and plug-ins can meet their needs before making a costly move to an ERP system.

A single source of truth helps teams avoid conflicting figures and focus on business decisions rather than reconciling inconsistent data.

Looking for shop-floor waste and process gaps can uncover profit improvements; Pauline described a process loophole that left around £30,000 of completed work uninvoiced.

Product profitability analysis should include indirect costs: a high-margin product may reduce overall profit if it requires substantial support, customer service, or sales resources.

Links

Pauline Healey on LinkedIn

Kevin Appleby on LinkedIn

GrowCFO Mentoring

Timestamps: 

00:01 — Kevin introduces Pauline Healey and Logical BI.

04:07 — Pauline describes Logical BI’s focus on manufacturing and engineering, and its broader operational support.

06:13 — The discussion turns to supply-chain complexity and its effects on cash flow.

16:29 — Pauline discusses evaluating existing systems and reducing reliance on disconnected spreadsheets.

24:14 — Kevin and Pauline discuss finding waste and improving processes rather than defaulting to overhead cuts.

25:00–26:42 — Pauline describes a missed-invoicing process gap and explains the revenue impact of unbilled work.

28:15–30:42 — They explore how hidden support costs can make a seemingly high-margin product less profitable overall.

31:00 — Pauline shares her perspective on AI use in business and finance.

35:11 — Pauline introduces Profit Harmony Hub, her membership platform for business and finance insights.

Find out more about GrowCFO

If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode.

GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here.

You can find out more and join today at growcfo.net

.entry-img img{
display:none !important;
}
.single .hentry .entry-img{
display:none !important;
}

https://open.spotify.com/episode/3ZNZMVgTlQPe0bhhozHSRb

For fifteen years the software P&L was a settled question: hosting was a rounding error, gross margins held above 80%, and investors underwrote growth against a predictable annuity stream. That assumption is breaking. AI and cloud infrastructure has become one of the largest and least predictable costs on the P&L, moving faster than most finance teams have built the vocabulary or forecasting discipline to handle it. Ed Barrow, Founder and CEO of Cloud Capital, joins Kevin Appleby to unpack research with 100 growth-stage CFOs: 96% blew through their AI budget last year, with overruns driven more by compute and storage than by model costs, and the average software company now spends 15–20% of revenue on compute and AI,  a fourfold jump, while over half of AI-native companies exceed 30%.

Barrow draws a distinction many finance teams miss: workforce AI scales with headcount and belongs in OPEX, while product AI scales with customer usage and belongs in COGS. Getting that split wrong distorts unit economics badly — he has seen entire cloud bills dumped into gross margin, and entire model bills buried in R&D.

He also warns against betting on falling token prices: consumption has outrun per-unit declines, and providers will pivot to profitability much as Uber did. With roughly $800 billion of debt-financed data centre buildout this year, the market will shift from pay-as-you-go to committed multi-year contracts. His prescription: befriend engineering, learn the language of tokens and model selection, and act as a chief investment officer who sets guardrails before the bill arrives.

Key topics covered:

96% of CFOs exceeded their AI budget last year, and the overruns were driven more by compute and data storage than by model and token costs, a blind spot for teams focused solely on token spend

AI spend has reached 15–20% of revenue for the average software company, up fourfold in months, with over half of AI-native companies above 30%, a fundamentally different business model from the 80%-margin SaaS playbook investors built their assumptions on

Workforce AI belongs in OPEX and product AI belongs in COGS, and misallocating between them distorts gross margin in either direction, particularly when tokens arrive through three or four billing routes at once, including hosted models inside AWS, Google Cloud and Azure invoices

Seat-based pricing breaks when costs scale with customer usage, creating the “token treadmill”, loyal, highly engaged customers accumulate data and context and become the least profitable cohort over time unless pricing is aligned to value and outcomes

Betting on falling token prices is a dangerous gamble, consumption growth has outrun per-unit price declines, and heavily funded model providers heading toward IPO will move from subsidised customer acquisition to profitable pricing, following the Uber trajectory

Debt-financed data centre buildouts will push the market from pay-as-you-go to committed contracts, handing CFOs material long-term liabilities to underwrite, with commitments often carrying 30–40% discounts but real balance sheet risk if utilisation falls short

Links

Ed Barrow on LinkedIn

Kevin Appleby on LinkedIn

GrowCFO Mentoring

Timestamps: 

00:32 — From rounding error to runaway bill: how the cloud line item changed

03:32 — Workforce AI vs. product AI: the OPEX/COGS split that changes everything

13:15 — 96% of CFOs blew their AI budget — and compute, not tokens, was the culprit

18:18 — The token treadmill: why your most loyal customers become your least profitable

20:55 — The end of the 80% gross margin playbook, and the Uber analogy for token pricing

26:41 — $800bn of debt-financed data centres and the coming shift to committed contracts

30:53 — What to do tomorrow: befriend engineering and think like a chief investment officer

36:11 — Paying for tokens through four different channels, and the allocation trap

Find out more about GrowCFO

If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode.

GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here.

You can find out more and join today at growcfo.net