Private Equity Data Guy

Phil Murphy and I cut through the buzz and give you a practical, no-nonsense playbook for actually getting AI working in a PE-backed services business. You won’t get a vendor pitch here — we mandated adoption rather than piloted it, and that forced reality to surface fast. The mandate was simple: every functional leader pick one real, recurring workflow, baseline it, and run it with AI. I let them pick the how; I removed the option not to try. What broke wasn’t the models — it was inconsistent processes, scattered data, and tribal knowledge hiding in inboxes and spreadsheets. So you and I learn a crucial rule: one process, one data structure per function. Don’t start with a data lake; start with one authoritative place for each kind of data, permissions you can enforce, and your top five processes written down. If you’re stepping into a portfolio company next week, my hands-on advice is the same thing we used: pick a high-volume, repeatable, reviewable task (invoicing, AR reconciliation, timesheet/ex travel approvals), run it with AI, and measure before-and-after. Couple adoption with governance from day one: inventory approved tools, draw a clear line for data that never goes into models, and create a sanctioned path that records activity. Remember my quip — you don’t need to add “AI” to your company name to get value — but you do need structure to see it. I promise you, adoption plus visibility is the symbiotic engine that turns table-stakes efficiency into a real, defensible advantage.Chapters:00:11 - Where the Next Dollar of AI Value Is06:26 - Mandating AI: How Synergy Transformed Back-Office Operations16:53 - Turning Internal Visibility into an Enterprise AI Governance Product21:33 - Defining Full Visibility: The Four Pillars of AI Governance31:40 - Advice for a new PE‑backed CEO: where to start with AI adoptionCompanies mentioned in this episode:Synergy ECPValfurious Capital PartnersCACIBooz Allen HamiltonStrata SolutionsAnthropicOpenAIAmazonAmazon Web Services (AWS)MetaGoogleCapital OneACG National Capital

I spoke with Sal Tirabassi about the role a fractional CFO can play as a company grows and prepares for a possible sale. He described how his team helped a business consolidate reporting across two related entities, correct system errors, and clarify the performance of its manufacturing and rental operations.Our conversation covered investor grade financial data, tracing margin changes to specific customers and products, measuring marketing channels through retention, and how AI may affect early career learning. I asked what smaller businesses gain from senior fractional CFO support and how experience across different clients shapes the advice a finance leader can give.Chapters:00:15 - The value of 1099 work: flexibility and direct access to decision makers01:17 - Guest Intro — Sal Tarabassi: From Growth Investor to Fractional CFO Founder12:26 - Preparing a Business for a $20M Sale18:18 - Tax Savings and R&D Credits for Industrial Equipment Builders24:10 - Investor-Grade Data: What Investors Expect in Diligence31:29 - Inside a NASCAR Racing Team35:44 - Client segments and the value offered to smaller clients44:24 - Closing: Guest Farewell and Contact InformationCompanies MentionedCFO Pro+Analytics: Sal’s firm provides fractional and interim CFO support.MC Partners: Sal spent nearly a decade there as a partner in growth and venture investing.IBM and Accenture: I referenced my early career at both companies and the client work I learned from there.Capital One: I mentioned leading the company’s marketing technology stack. [1]Websites MentionedCFO Pro+Analytics: https://www.cfoproanalytics.com/Sal Tirabassi on LinkedIn: https://www.linkedin.com/in/stirabassi/Guest InformationGuest’s Full Name: Sal TirabassiCompany / Organization: CFO Pro+AnalyticsSalvatore “Sal” Tirabassi founded CFO Pro+Analytics in 2021 after 15 years investing in growth and venture businesses, including nearly a decade as a partner at MC Partners. He leads a team of senior fractional CFOs supporting founder owned and family owned companies.

In this episode, I talk with Carson Marr, founder of Buyable, about where real money hides inside private equity portfolio companies. Carson spent years advising executives at firms like Blackstone, Vista, and Thoma Bravo on buying and building enterprise technology before he built Buyable, a company that reviews contracts and billing records and only gets paid when it finds missed revenue. His team found millions of dollars sitting in the gap between a billion dollar company's contracts and its billing system, work completed in sixty days.Carson explains the two conditions an outcome based deal needs, a buyer who is hungry for a specific result and a project narrow enough to prove direct cause and effect. We talk through why contract and billing gaps open up during mergers, why escalator clauses get missed for years, and why executives choose Salesforce or IBM because the downside of a safe choice protects their career. Carson also shares a story about turning a research tool into a pipeline engine that created one hundred and four million dollars in new pipeline within sixty days for a large AI company.Chapters:00:09 - Turning Research into Revenue: $104M Pipeline in 60 Days01:10 - Introducing Carlson Ma — From Deal Advisor to Founder of Buyable09:06 - Proving outcomes before engagement — using niche case studies18:13 - Case Study: Turning Research into $104M Pipeline in 60 Days24:06 - The CRM Dilemma: When Implementation Isn't an Outcome-Based Play31:02 - AI Disruption: Industries at RiskCompanies Mentioned Buyable, RSM, Blackstone, Vista Equity Partners, Thoma Bravo, Salesforce, Sierra, IBM, ZoomInfo.Websites Mentioned Carson's personal site is carsonmarr.com. He also writes on Substack and can be found on LinkedIn. His company site is Buyable.Guest Information Carson Marr is the founder of Buyable, a company that reviews contracts and billing systems inside private equity portfolio companies and only gets paid when it finds missed revenue. He spent years in management consulting at RSM, where he worked with firms including Blackstone, Vista, and Thoma Bravo on enterprise technology decisions inside their portfolio companies. He writes about deal making on Substack and can be found on LinkedIn.

Here's the scariest sentence in private equity: every KPI on the value creation plan is green and enterprise value is exactly where it started. I paint that horror show because it's exactly what I've seen: dashboards full of check marks, a room of smart people, and nobody who can explain why the company isn't worth more. You get the quick wins in year one—pricing tweaks, a bit of cost takeout—and everyone celebrates. Then year two arrives, the heavy-lift initiatives—cross-sell, add-ons, new segments—take longer, definitions diverge, and the board deck looks prettier than reality. What breaks the thesis isn't bad strategy; it's broken data. I walk you through how every lever—pricing, procurement, cross-sell, M&A—relies on numbers you can trace. Most portfolio companies run on a patchwork of systems, one heroic finance person, and no agreed definitions. So the dashboard turns green because we're measuring activity, not value. The net result: enterprise value flatlines and buyers discount everything because they can't trust the baseline. If you want to find the gap, try my one-meeting test: pick the single number the plan depends on and ask the room who can trace it to source, define it, and reproduce it. If the room goes quiet, you found the problem. I also flag a free scorecard at the end to see where you stand and offer a chat to show you where the plan and the data disagree while you still have time to fix it.Takeaways:You should stop trusting green KPI dashboards — they often measure activity, not the actual value increments buyers will pay for.I explain why value creation is the whole game now: cheap leverage and multiple expansion no longer carry deals.If you buy a new BI tool or hire a fancy consultant without fixing definitions, you're just decorating broken data — don't order the pizza yet.We learned the hard way: merging companies means merging messy datasets first, otherwise any post-acquisition claims won't survive diligence.You can test your portfolio quickly — pick the one number the plan depends on and ask who can trace it back to source.Foundations matter: build data definitions and ownership early, because they compound value each quarter and speed up provable initiatives.I even put a free scorecard at the bottom — use it now so we can find where your numbers and the plan disagree before it's too late.

In this episode I talk with Afrozy Ara, founder and CEO of LuminaData, an AI transformation platform for finance operations backed by Techstars and Berkeley SkyDeck. Afrozy started her career as an electrical engineer, then spent over a decade in enterprise consulting at firms including Mu Sigma, where she worked directly with Fortune 500 finance teams and built out diagnostic practices that benchmark how finance work actually happens inside large organizations. She started LuminaData a year and a half ago, beginning with a single problem inside reconciliations, and her founding thesis is that AI adoption is not the same as AI transformation.We talk about where the money actually hides inside a finance team, from duplicate invoices and unpaid remittances to the nonstandard contracts that large customers like Amazon force onto their vendors. Afrozy explains why giving a team of ten people the same AI tool can produce ten different answers without a shared process behind them, and why she still keeps a human in the loop for every exception rather than letting an agent make deterministic decisions on its own. She also walks through the three step method her team uses inside a client: look across the whole business, pick one area and go deep, then show a small result fast to build trust before scaling the work across the order to cash process.Chapters:00:19 - The Agent: Reimagining Your Process04:11 - Thesis: AI adoption is not AI transformation08:15 - AI in the Enterprise — From Copilots to Capturing Value20:56 - Humans in the Loop: Designing AI Agents with Human Oversight25:05 - Connecting platforms to reveal business performance and KPIs33:21 - Personal Writing, Unfinished Books, and AICompanies MentionedLuminaDataTechstarsBerkeley SkyDeckMu SigmaCapital OneAmazonRampBrexQuickBooksAnthropicWebsites Mentionedlinkedin.com/in/afrozy-ara-57527612luminadata.aiGuest InformationAfrozy Ara is the founder and CEO of LuminaData, an AI platform for finance operations backed by Techstars and Berkeley SkyDeck. She began her career as an electrical engineer before moving into enterprise consulting, spending over a decade at firms including Mu Sigma, where she partnered with Fortune 500 finance teams on data and process problems. She started LuminaData a year and a half ago, beginning with a single problem inside reconciliations, and her work today centers on her belief that AI adoption is not the same as AI transformation. You can find Afrozy on LinkedIn.Guest's Full Name: Afrozy AraCompany / Organization: LuminaDataWebsite or Main Page: Not providedGuest Social Media Links: https://www.linkedin.com/in/afrozy-ara-57527612/