Leysan Gilfanova | CFO Advisory

Saks emerged from bankruptcy this week, five months after it filed. The corporate press releases call it a bright future and a stronger balance sheet under its new corporate name, Exemplar Luxury Group. On paper, the numbers back that up. The company came out owing about $1.2 billion, a massive drop from the $3.4 billion it carried going in. But coming out of bankruptcy doesn't mean a company's problems are solved. It means a judge signed off on a structural reset to keep the entity alive. For Saks, that reset came down to a debt-for-equity swap. The lenders it owed agreed to forgive most of the debt, and in return, they became the new owners. Those lenders aren't retail operators. They're two investment firms, Pentwater Capital and Bracebridge Capital. The previous equity holders lost everything. Even Amazon, which took an estimated 23% stake in the company during the 2024 Neiman deal, saw its massive ownership stake completely wiped out. To survive, the new Saks had to aggressively slash its footprint, cutting thousands of jobs in the process. It emerged with just 49 total stores. Ironically, 33 of those are Neiman Marcus locations, and just 15 are Saks Fifth Avenue doors. It shuttered the vast majority of its discount operations, including off-price brand Saks OFF 5TH. Saks is abandoning the bargain shopper almost entirely to focus its remaining capital on full-price luxury. Then there is Richard Baker, the real estate financier who controlled Saks. Baker's playbook was built on leverage. He bought Saks in 2013 with borrowed money and mortgaged its Fifth Avenue flagship. In 2024, his acquisition of Neiman Marcus stacked billions more in debt onto the balance sheet. He was out when Saks filed for bankruptcy in January, his ownership stake wiped out. The creditors subpoenaed him over his communications and the deals behind the collapse, and a litigation trust can still bring claims against him. The brands that supply Saks were owed about $700 million. During the bankruptcy, the company set aside $600 million specifically to pay those suppliers. It's a simple reality: the store cannot operate if luxury brands refuse to ship them inventory. And underneath it all, the reason Saks ended up here hasn't changed. It lost its core shoppers to competitors like Nordstrom and Bloomingdale's. Wiping out the debt fixes the balance sheet, but it doesn't automatically bring those shoppers back. So that's what coming out of bankruptcy really means. New owners, a much smaller company, thousands of eliminated retail jobs, and the same business problem it started with. That's the lesson: Clearing the debt was the easy part. Fixing the business fundamentals is a completely different story. 🔔 I'm Leysan, ex-CFO who breaks down billion dollar business models every single week. Thank you for reading! #BusinessModel #Bankruptcy #Retail

Most business owners using AI keep getting generic answers. Smart enough, but surface-level, like advice from someone who's never seen their business. The reason is simple: AI doesn't know your business. Every new chat starts from zero, so you're either re-explaining yourself every time or settling for generic. The fix is to build it a brain. One place that holds the core of your business, so it shows up to every conversation already knowing how you operate. That's your business brain. Here's the part that makes it easy. You don't sit down and write an essay. You open the AI you already use, and you talk. Answer these five questions out loud, the way you'd explain your business to a new hire. I dictate mine with Wispr Flow so I can just speak it. When you're done, tell the AI to summarize what you said and save it as a file called business-brain.md. That's your brain, built in one sitting. You're answering five questions, not writing a manual. More isn't better here. Past a point, extra detail makes the AI worse, not sharper. A few pages is plenty. The five questions, most important first: 1. What does your business do, and where are you taking it? Start here, because everything else builds on it. A sentence or two on what you do and the goal you're driving toward this year. Now its answers aim where you're going instead of nowhere in particular. 2. Who are your customers? Who you serve, what they're struggling with, and the words they actually use. The more real this is, the more what it writes sounds made for them instead of for everyone. 3. What do you sell them? Your services or products, and roughly what you charge. So when it writes a proposal or answers a customer, it's working from what you actually offer. 4. How do you talk? Your tone and the words you use, so what it writes comes back sounding like you, not like generic AI. This is the one most people skip, and it's why their output sounds like everyone else's. 5. How do you want the AI to work with you? Your rules for working together. Get to the point, don't make things up, ask when you're not sure, tell me when I'm wrong. That's it. Answer those five and let the AI write them up. One last step, and it's the one that makes the whole thing work: don't paste the file into every chat. Drop it into a Project in ChatGPT or Claude, or a Gem in Gemini. You attach it once, and from then on it shows up in every conversation already knowing your business. This is the foundation. The full system goes deeper, with a detailed file behind each piece: your brand, your ideal customer, your products and services, and more. That's a bigger build, and I'll cover it in a future post. Follow Leysan Gilfanova for practical AI for business owners.

In March I wrote about Micron killing a 29-year-old brand to go all-in on AI. The knock was always that memory is cyclical and the boom never lasts. A few days ago they announced $100 billion in contracts designed to make sure it does. Micron makes memory — the chips that hold data while a computer runs. For 40 years it's been a brutal business. Three big makers sell nearly the same thing, so the price booms and crashes every few years. The last crash, in 2023, cost Micron $5.8 billion. Then AI created a shortage. And last quarter was the biggest in company history. → $41 billion in revenue, up 346% → 85% gross margin, a record → $28 billion in profit That 85% is insane for a memory company. Last quarter was 75%. A year ago, 39%. So where did the record come from? Price. The shortage let Micron raise prices around 60% in a single quarter. That's not steady demand it can count on. It's the top of a cycle, and every cycle before this one ended in a crash. So instead of just taking the windfall, Micron did something a commodity business almost never gets to do. It locked 16 customers into contracts that run as long as five years. They commit to buy set volumes at a floor price, whether they need them or not. In a shortage, guaranteed supply beats a lower price. These contracts will eventually cover up to half of Micron's revenue, at prices that hold even if the market crashes. But a contract only works if the customer can pay. And some of these buyers are AI companies that still lose money. Micron just invested in one of them, Anthropic. So it's now helping fund a customer it depends on for sales. There's a name for that: circular financing. Micron invests in Anthropic. Anthropic buys from Micron. Part of the demand behind this record quarter is Micron's own money coming back as revenue. It's running through the whole AI economy right now, hundreds of billions of dollars of it, and Micron just stepped into the loop. The rest of its revenue still rides the market, where prices stay high only until new supply catches up. Micron doesn't expect that before 2028. Honestly, locking in these contracts is a smart move. The stock price is a different question. It's at an all-time high. At that price, the market is betting these record profits are here to stay. But this business has boomed and crashed every few years for decades, and the shortage behind the record will end like every one before it. That's the lesson: A signed contract looks like guaranteed money, but it's only as secure as the customer standing behind it. 🔔 I'm Leysan, ex-CFO who breaks down the economics behind tech companies every single week. Thank you for reading! #BusinessModel #AI #Semiconductors

@asmartbear Haha. So true. Just because you can build it, doesn’t mean you should 😂.

With AI you can make products no one wants faster than ever!

Humans are good at doing the 80/20 of a task – the 20% of effort that adds 80% of the value AIs are good at doing the 20/80 – the 80% of effort that adds 20% of the value This is why we're such good collaborators

Brunello Cucinelli hit €1.4B in 2025 — while luxury had its worst year in 15 years. LVMH revenue declined 5%. Kering saw drops across most of its portfolio. But this 72-year-old Italian founder, who started dyeing cashmere in a small workshop in 1978, hit record highs both years. His edge? He targets 10% growth per year. In 2012, bankers at his IPO told him to target 30% annual growth. His response: "Forget it. If you want a company that makes profits by going against the principles of humanity, don't buy us." For 14 years, he's held that discipline. But a growth cap alone doesn't explain the results. The model behind it does. Headquarters isn't Milan or Paris. It's a 14th-century castle in Solomeo — a 500-person village he spent 40 years restoring. He calls it the Hamlet of the Spirit. He built a School of High Craftsmanship to train the next generation of artisans. He doesn't compete for talent on the open market. He creates it. They earn 20% above industry norms, work 8:30 to 5:30, and take 90-minute lunches. He built a place where they want to live and work. A culture that makes them stay for decades. So when COVID hit in 2020 and revenue dropped 10%, he didn't cut a single artisan. He kept full salaries and didn't ask suppliers for discounts. The company lost €32M that year. But he grew 31% the next year — while competitors who laid off their craftsmen couldn't find skilled hands to rehire. The result? → FY2019: €608M → FY2025: €1.4B → 15% average annual growth — by not trying to grow faster That's the lesson: The biggest advantage might not be the product or the strategy. It might be the culture that's been building for decades that simply cannot be replicated. 🔔 I'm Leysan, ex-CFO who breaks down iconic luxury brands every single week. Thanks for reading! #BusinessModel #LuxuryBrand #BusinessStrategy

You are the CEO & CFO of your financial life. Act like it.

Warren Buffett on why holding too much cash is just as dangerous as holding too little: Buffett is asked about his view on holding cash as a strategy. He doesn't mince words: "Cash is always a bad investment." But he's quick to draw an important line. The problem isn't cash itself, it's excess cash: "When people said cash is king a year ago, I mean, that's crazy. Cash wasn't producing anything and it was sure to go down in value over time." That said, Buffett isn't arguing for zero liquidity. He frames it as a matter of survival: "You always want to be sure you have enough. It's like oxygen. You want to be sure it's around, you know, but you don't need to have excessive amounts of it around." Too little and you can't operate. Too much and you're quietly bleeding value every single day. He then shows how this plays out in his own decision-making: "We came in with something over 40 billion of cash and we've got about 20 billion now and we've had some earnings. So we put a lot of cash to work and I like that." Deploying over $20 billion wasn't a reluctant move, it was the goal. For @WarrenBuffett, sitting on surplus cash is simply a mistake: "Anytime we have surplus cash around, I'm unhappy." His preference is clear: "I'd much rather own a good business than have cash." Cash belongs in the hierarchy as a tool for survival and opportunity. The moment it becomes excess, it starts costing you.

“Failure is so much more interesting than success. And it always saddens me that school doesn't teach that. At school the thing is to be brilliant and to get the answer right first time. And there are brilliant people who could do that, but for the rest of us who are not brilliant, we have to strive, and we have to go through failure. And we realize that you don't get it right the first time or the second time. In my case and I counted it. It took 5,127 times. With failure you question it “Well why did it go wrong?” And actually the reason it goes wrong is often very, very interesting. Where something works you say “Great that worked.” You don't even stop to wonder why it works. So if you've got to enjoy failure.” —James Dyson

In 2022, FIFA made $7.5 billion from the World Cup. In 2026, FIFA is set to make $13 billion. The biggest cycle in its history — a 73% jump. FIFA is a Swiss non-profit that created the World Cup in 1930. It owns the World Cup brand. It doesn't build the stadiums, broadcast the matches, or sell the tickets directly. It licenses the right to do all of that. Host cities take on the cost of hosting. Broadcasters, sponsors, and fans pay for everything else. → Broadcast rights: ~$3.9 billion → Sponsorship: ~$2.8 billion → Ticketing and hospitality: higher than ever this cycle, with dynamic pricing introduced at a World Cup for the first time In 1974, FIFA started bundling rights globally instead of letting host countries sell their own. The model has scaled since. This time, the tournament expanded: 48 teams playing 104 matches, up from 32 teams playing 64. More inventory to sell. And most of it was contracted years ahead. By end of 2024, FIFA had 62% of cycle revenue locked in — 18 months before the first match. The host cities, meanwhile, were spending. The 16 host cities covered stadium upgrades, security, and transit. They don't share in ticket, concession, parking, or merchandise revenue. They're betting on tourism and brand exposure. FIFA doesn't keep the full $13 billion. About $2.25 billion goes to member associations and $871 million to participating teams this cycle. The model is producing record revenue — and record pushback. In March 2026, fan groups filed an EU complaint over dynamic ticket pricing. ProPublica reported in April on host-city costs. Mark Pieth, a former FIFA adviser, filed a human-rights complaint over the 2034 Saudi award. FIFA's response: a $60 supporter tier covering under 2% of capacity. That's the lesson: Owning what others pay to use is one kind of business. Doing the work is another. The economics work very differently on each side. Whether the pushback changes anything depends on the next cycle. The 2030 and 2034 hosts are already locked in. 🔔 I'm Leysan, ex-CFO who breaks down billion dollar business models every single week. Thank you for reading! #BusinessModel #SportsBusiness #BusinessStrategy

Most companies are pouring money into AI and getting almost nothing back. A small group is seeing 163% productivity growth. PwC just analyzed over a billion job ads and thousands of companies for its 2026 AI Jobs Barometer. What it found splits businesses in two. Almost every business now spends on AI. Only 8% of CEOs say it has brought in real revenue. And 20% of companies are capturing 74% of all the gains. So what are the winners doing differently? Most companies point AI at the work they already do, to get it done a little cheaper. The winners point it at growth. They use AI to take on more work, reach new customers, and build things they couldn't before. They redesign how the work gets done instead of bolting AI onto the old way. They also lean further into it. The companies getting the most from AI are twice as likely to hand real tasks to AI agents, start to finish. And it shows up in the numbers. The companies using AI the most grew their headcount 52% since 2018. The ones barely using it grew 36%. They're growing, so they're hiring. The very best performers go further still. Their productivity, the revenue each employee brings in, has grown 163% since 2018. Their wages have grown 68% to match. AI pays off when you aim it at growing the business. It mostly disappoints when you aim it only at cutting costs. So the real question for your business isn't whether to use AI. Where are you pointing AI right now? ♻️ Repost to help another business owner get more out of AI. 🔔 Follow Leysan Gilfanova for practical AI for business owners. I help business owners use AI to grow revenue, profit, and capacity. DM me if you want to talk it through.

In March 2024, e.l.f. was the best-performing beauty stock in America. Just over two years later, it's down 71%. It's still growing. Still gaining market share. Still the #1 brand with Gen Z. So how does a company keep winning customers — and lose three-quarters of its value? e.l.f. sells cheap versions of expensive makeup. An $11 primer modeled on a $38 one. A $7 concealer modeled on a $32 one. The low price is the whole appeal. And it worked. Sales grew nearly 480% in six years: → FY2020: $283M in sales → FY2024: $1 billion → FY2026: $1.6 billion But that low price drew the most price-sensitive shoppers in beauty. A model like this only works while costs stay low. In 2025, that became more difficult. Most of what e.l.f. sells is made in China, and the tariff rate on those goods doubled to about 55%. To cover it, e.l.f. raised prices a dollar that August. Sales dropped. When it cut one product's price from $18 to $14 instead, sales jumped 40%. e.l.f. got the message. It's rolling the price increases back. The $58 million it's getting back in tariff refunds is going into lower prices, not profit. And longer-term, it's moving production out of China. 45% has already shifted, up from 1% three years ago. Even so, e.l.f. is still growing. Sales jumped 35% last quarter. But almost all of that came from rhode, the brand it bought last year for a billion dollars. Strip rhode out, and the rest grew about 1%. A company still loved by its customers, but worth 71% less than it was two years ago. That's the lesson: Customer love and pricing power are two different things. The customers you win on price are the ones you can't raise prices on. 🔔 I'm Leysan, ex-CFO who breaks down billion dollar business models every single week. Thank you for reading! #BusinessModel #Pricing #BusinessStrategy

The best way to learn AI is to jump right in. Not ready for that? Here are nine free courses you can take today. Every one clears the same bar: it helps a business owner actually put AI to work, not just understand it. Start simple → learn to use it well → turn it into a strategy → apply it to your own business START WITH THE BASICS 1. Google — Introduction to Generative AI. A 45-minute intro to what generative AI is and how it works. 🔗 coursera.org/learn/introduc… 2. DeepLearning.AI — Generative AI for Everyone. Andrew Ng on how generative AI works and where it fits in a business. 🔗 coursera.org/learn/generati… 3. Microsoft — Career Essentials in Generative AI. A structured intro to using AI tools in everyday work. 🔗 linkedin.com/learning/paths… LEARN TO ACTUALLY USE IT 4. Anthropic — AI Fluency: Framework & Foundations. A framework for delegating to AI, directing it, and checking its work. 🔗 anthropic.com/ai-fluency 5. Vanderbilt — Prompt Engineering for ChatGPT. How to write prompts that get better answers from AI. 🔗 coursera.org/learn/prompt-e… TURN IT INTO A STRATEGY 6. Amazon — Generative AI Learning Plan for Decision Makers. How to plan an AI project and prepare your organization. 90 minutes. 🔗 skillbuilder.aws/learning-plan/… 7. IBM — Generative AI for Executives and Business Leaders. Where AI pays off across marketing, service, finance, and HR. 🔗 coursera.org/specialization… 8. Wharton — AI Fundamentals for Non-Data Scientists. AI fundamentals taught for business leaders. 🔗 coursera.org/learn/wharton-… APPLY IT TO YOUR BUSINESS 9. University of Maryland — AI Empowerment for Small Businesses. Built for business owners bringing AI into a small company. 🔗 coursera.org/learn/ai-empow… You don't need all nine. Pick one and actually do it this week. Short on time? The Google intro and Anthropic's AI Fluency are the quickest. Which one are you starting with? 🔖 Save this so the whole list is here when you need it. 🔔 Follow Leysan Gilfanova — practical AI for business owners.

Most business owners I talk to think AI is too technical for them. It's actually the opposite. AI makes technical work accessible to non-technical people. Work that used to require an analyst, a developer, or a team of specialists — AI does it now. Here are 14 ways to start, without being technical. 1. Focus on the Problem, Not the Tech — You don't need to know how AI works. Find the bottlenecks in your workflows (data entry, emails, scheduling) and let AI handle them. 2. Dedicate Time to AI Exploration — AI is moving incredibly fast. Block 30-60 minutes a week to read newsletters, test platforms, or try new workflows. 3. Start Small and Scale Smart — Don't overhaul at once. Start small: AI summarizing meeting notes, repurposing a blog post for a week of social. Measure, then expand. 4. Protect Data Privacy — Free-tier AI may train on your inputs, meaning your client data could surface in someone else's output. For sensitive work, use paid plans like Claude Team. 5. Make AI Obvious — Pin AI tools to your browser, set ChatGPT or Claude as your homepage, or put the app on your phone's dock to stay top-of-mind. 6. Habit-Stack AI — Integrate AI into what you already do. Use voice tools like Wispr Flow or Otter.ai to brainstorm or dictate emails while walking or commuting. 7. Embrace the Art of Iteration — The first AI response is rarely final. The skill is asking again: "make this punchier" or "cut in half." Treat the first output as a starting point. 8. Pick a Master Tool — Pick one AI as your primary (Claude, ChatGPT, Gemini), then bring it into platforms you already use. Slack, Notion, and Gmail have built-in AI features. 9. Build an AI Business Brain — AI doesn't know your business unless you tell it. Build a doc with your context (what you do, customers, goals, voice) and upload to Claude Projects. 10. Make AI Your Business Collaborator — Treat AI as a true collaborator. AI can analyze sales trends, draft follow-ups, research competitors. Workflows change with a teammate that fast. 11. You're Still the Decision-Maker — AI generates drafts fast, but can't tell what's right for your business. Review every important output and don't outsource your judgment. 12. Build Trust Before You Step Away — AI is most useful once you trust it. Start side-by-side: prompt, edit, send back. As you learn what it does well, hand off larger pieces. 13. Share Your Learnings With Others — You'll learn AI faster trading notes. Set up a chat group with your team or peers figuring out AI. Share what's working. 14. Measure Economic Value — Did AI cut response time, shorten cycles, free up capacity? Tie AI to outcomes you can see in revenue, profit, or capacity. Which one are you starting with? Tell me in the comments. I help business owners use AI to grow revenue, profit, and capacity. DM me if you want help.

Adobe's revenue is at a record. AI revenue tripled. Forecast raised. The stock dropped 45% over the past year. For over a year, Wall Street has been worried AI is going to kill the legacy software business. Photoshop and Illustrator now compete with Midjourney and ChatGPT — outputs that used to take hours now take seconds. But generating an image is only the start. Turning it into something a designer can actually use still takes Adobe. That gap is what Adobe is fighting to protect. That's why Adobe has been responding on every front. → Built their own AI tool, Firefly, inside Photoshop, Illustrator, and Premiere → Paid $1.9 billion for Semrush, which helps brands show up in answers from ChatGPT and Perplexity, not just Google → Launched freemium tiers to bring in new users They had also planned to raise Creative Cloud prices later this year. They held off, worried customers would leave. Either way, the slowdown is showing up. Underlying business growth slowed from 10.2% to 8.2% this year — the Semrush acquisition covered the difference. Meanwhile, the team making these bets is also changing. Both top executive seats are in transition — the CEO after 18 years, the CFO just gone. The 45% decline has been investors pricing in AI risk for 12 months. Adobe holding prices flat and losing both top executives just confirmed what investors had been worried about. But Adobe has made hard calls and been doubted before. In 2013, Adobe killed their perpetual license business. Customers used to pay once and own Photoshop forever. Now they had to subscribe monthly. Customers revolted — petitions, social media outrage, threats to switch. Wall Street caught the signal. The stock dropped 12% in the months that followed. But by 2015 it had hit new highs. And revenue followed — from $4.4 billion in 2012 to $24 billion in 2025. Whether this is 2013 again, or the first time the bet doesn't work, nobody knows yet. That's the lesson: A comeback is never guaranteed, but adapting to disruption always starts with a step backward. 🔔 I'm Leysan, ex-CFO who breaks down the economics behind tech companies every single week. Thank you for reading! #BusinessModel #AI #BusinessStrategy

Most business owners are trying to figure out which AI tool is best for them. Claude, ChatGPT, or Gemini? The truth is there's no one answer. All three have strengths and weaknesses. It depends on what your business actually needs day to day. Here's what I personally do, and what I recommend to business owners. Pick one as your master — the tool that fits the most of your work. Feed it your business context so it shows up to every session already knowing how you operate. That's your business brain (this is where it gets powerful!!!). Use the other two for the specific jobs they handle better. My master is Claude. 90% of my work runs through Claude Code, and I switched to it before it became popular. Gemini handles fact-checking, cross-referencing, and second opinions. ChatGPT for quick ideation and one-off image work. Here's what each one is best at. 1️⃣ Pick Claude for long documents and deep analysis. ↳ Contracts, financial analysis, long-form writing, code review. ↳ Handles up to 1M tokens with sustained reasoning. 2️⃣ Pick ChatGPT for everyday speed and visual work. ↳ Drafting, brainstorming, summarizing, image generation. ↳ Largest plugin ecosystem and custom GPTs for recurring tasks. 3️⃣ Pick Gemini for work inside Google Workspace. ↳ Gmail, Docs, Sheets, Drive. ↳ Native image and video generation (Imagen + Veo), plus Deep Research across web, Gmail, Drive, and Meet. Pick a master, give it a business brain, and use the other two for what they're best at. That's how you actually compound value from AI. Which one is your main tool? ♻️ Share if you found this useful. 🔔 Follow Leysan Gilfanova, CPA — practical AI for business owners.

@SJosephBurns Simplicity is the key. But it’s so much easier to overcomplicate things rather than keep them simple!

Warren Buffett: "We haven’t succeeded because we have some great, complicated systems or magic formulas we apply or anything of the sort. What we have is just simplicity itself."

@rosenmaya1991 Well-said Maya. Every new tool adds complexity and overhead.

Unpopular opinion: your team doesn't need another AI tool. It needs someone to kill three tools you're already paying for and replace them with one that actually connects to your stack. Consolidation > accumulation. Every time.

@russellbrunson Very powerful reminder.

"Stop negotiating with your potential. You either step into the fire and become who you’re meant to be, or you spend your life explaining why you didn’t."

@rosenmaya1991 Such an important point! Process is the foundation. Can't build something without knowing what you are actually building!

@CFOAnalysis Every team I talk to wants an AI agent. Almost none of them have written down what step 3 of their process actually is. You can't automate ambiguity — you just get faster chaos. Documentation is the unsexy prerequisite nobody wants to do first.

Nearly every business is using AI now. Almost none are making real money from it. That's not a hot take. It's what the research keeps finding, report after report: historic spending, and almost no measurable return. So the real question has quietly changed. It's not "should we use AI?" anymore. It's "why does everything look like progress while nothing actually changes?" I went through the major AI reports from the past year and pulled the 12 that answer it. The order matters more than the list: Where we are → why nothing's changing → where it's actually paying off → what's coming next → how to win → where it's heading WHERE WE ARE 1. McKinsey — The State of AI 2025. 88% of companies use AI. Only 39% see it reach the bottom line. 🔗 tinyurl.com/28aknkep 2. Stanford HAI — AI Index 2026. The full year in AI. The hype phase is ending; the "prove it works" phase is starting. 🔗 tinyurl.com/2c9h79ef WHY NOTHING'S CHANGING 3. MIT — The GenAI Divide. 95% of companies get zero return on AI. The 5% who don't are doing something specific. 🔗 tinyurl.com/2yla9zzq 4. PwC — Global CEO Survey. Only 12% are getting both more revenue and lower costs from AI. More than half get neither. 🔗 tinyurl.com/yvclfskz WHERE IT'S ACTUALLY PAYING OFF 5. Capgemini — AI in Action. Where returns show up first: 26–31% cost cuts in finance, operations, and customer service. 🔗 tinyurl.com/26d5ytwh 6. Goldman Sachs — AI Productivity. Almost nothing economy-wide, but a clear 30% jump in two jobs: software and customer service. 🔗 tinyurl.com/29owcn78 7. Salesforce — State of Sales 2026. Agents are already cutting sales busywork: 34% less research, 36% less drafting. 🔗 tinyurl.com/2a6p9spc WHAT'S COMING NEXT 8. Gartner — Hype Cycle for Agentic AI. 17% use AI agents today. 60% plan to within two years. 🔗 tinyurl.com/2238zwpr 9. Deloitte — State of AI in the Enterprise 2026. AI agents go from 23% of companies to a likely 74% in two years, but only 21% can govern them. 🔗 tinyurl.com/28tcyjak HOW TO WIN 10. McKinsey — Superagency. Only 1% of companies are AI-mature. What separates them is training, not better tech. 🔗 tinyurl.com/22zjqoak 11. IBM — How to Maximize AI ROI. Teams that follow a few specific practices hit a median 55% ROI. The closest thing to a playbook. 🔗 tinyurl.com/24eey4b3 WHERE IT'S HEADING 12. BCG — AI Will Reshape More Jobs Than It Replaces. 50–55% of jobs reshaped in two to three years, far more than get eliminated. 🔗 tinyurl.com/28yw58zb Most people want to skip to the end and just "do AI." The ones actually getting returns work through it in order. What step is your business stuck on?