Native[risk]

I fell victim to a fraudster. She’s 4 years old. We came back home from the park last Sunday. It was starting to get late and I took a peek at her room: Mayhem. “We’re going to wash in five minutes, please put your room in order”, I said and turned to do the same in our living room. Not two minutes had passed when she came in running, announcing: “I cleaned up my room!” So I took another peek, and to my surprise it was indeed clean! “Good job! Now go and wait for me in the shower, I’ll be right there”. I went into her room to get her PJs and then I saw it: She just shoved all the mess under the desk. I’m talking blankets, dolls, half-eaten food, you name it. I got mad, I called her and I lectured her about lying to her dad. She looked abashed. I wasn’t sure whether it was because she understood her wrongdoing, or because she got caught. She cleaned it up while I watched this time, then we were off to the shower. Nothing out of the ordinary for any parent. But here’s the sad part: What’s the difference between how we treat four-year-olds and how we treat fraudsters when they get caught? ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Merchant Payments Ecosystem is just 2 weeks away! Here's when to find me: Day 1 // March 18th 16:45 "Cutting through the hype: How to select the right fraud vendor for you" ? (6 seats left - reserve now before they are gone!) Day 2 // March 19th 11:25 "MPE Fraud Prevention Lab" Bonus track: FAN Network pre-MPE event // March 17th 15:00 "Biggest Mistakes in Fraud Prevention" I'm really looking forward to catching up with so many great folks and dear friends! If you're there and want to chat, or just say hi - let me know! See you all soon. -------------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Fraudsters attack companies that are slow to respond. Here's how I win over them: ?? Detect & Assess ? Industry average: 3-4 weeks ? Main time leak: Waiting for chargebacks ? Battle-tested fix: Real-time monitoring of failed attempts ? New response time: 2-3 days Analyze & Plan ? Industry average: 5 days ? Main time leak: Siloed teams analyzing separately ? Battle-tested fix: Cross-functional rapid investigation protocol ? New response time: 1 day Test & Deploy ? Industry average: 1-2 weeks ? Main time leak: Manual release cycles ? Battle-tested fix: Automated deployment pipeline ? New response time: 1-2 days Reaction time before: 6 weeks Reaction time after: 4 days Key lesson: When you have an adversary, it's not about tech, accuracy, or data. It's about one thing only: Your reaction time. Want the full breakdown? In this week's "The Saturday Fraud Strategist" I'm going to dive deep into: ? How the French army's paralysis in WW2 mirrors modern fraud defense ? Why your fraud team's expertise isn't reflected in results ? The 6 silent bottlenecks slowing down your response If you're not subscribed yet, the link is under my picture. ?? See you on Saturday! ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Which fraud model performs better? A vendor’s or one built in-house? Let’s compare: Vendor pros: ? Little to no dev cost ? Data volume is higher ? 3-6 months for engine tuning ? Large feature store (supposedly) Vendor Cons: ? Black-box ? API limitations ? Higher latency In-house pros: ? You’re always prio 1 ? Optimal customization ? Data relevancy is higher In-house cons: ? You bear all costs ? ~1 year for engine tuning So, what’s the verdict? As always: “It depends”. Strapped for resources and need a solution yesterday? Go with a vendor. Ready to invest for the long term? Start thinking about building your own AI stack. But here’s a question to the audience: When would you use both? ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

You automated fraud prevention but still struggle with scaling? Do this to grow without more hires: ?? Level 1: Foundation ? What: Model Participation Rate measurement ? Why: You can't improve what you don't track ? Result: Clear scaling baseline Level 2: Acceleration ? What: Systematically increase AI-driven decisions ? Why: AI scales better than rules or headcount ? Result: Growth-ready operations Level 3: Optimization ? What: Master the performance-scale balance ? Why: Conscious resource allocation ? Result: Sustainable scaling Build in this order. Each level enables the next. Want the full breakdown? In this week's "The Saturday Fraud Strategist" I'm going to dive deep into: ? A simple calculation to test if you're ready to scale ? Why most fraud automation actually hurts scalability ? The counterintuitive way AI reduces operational complexity If you're not subscribed yet, the link is under my picture. ?? See you on Saturday! ---------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Early stage founder in fraud? These are the top fallacies I’m currently seeing: ?? ?? Mismatch between ICP and user persona Building a cutting edge AI application? Don’t sell it to SMBs that won’t know how to use it. Building a modern real-time decision engine? Don’t waste time on incumbent FIs who manage everything with batches. ?? Too much focus on technology The only people that care about it are investors. (And it’s still secondary to growth). Stop self-obsessing on the cool things your team can build. Start obsessing on what pains your customer. Don’t get disappointed when the solution is boring. Making money is often boring. ?? Too little focus on delivery Technology < Product < Delivery It’s so rare to meet founders that truly get it. If your product doesn’t solve a real pain, your tech is meaningless. And if delivering it to the customer is a pain, your traction will stall. Remember: integration can easily cost more than your annual fees. Make it easy to start using your product. Don't ask me how I know. ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Anti-fraud systems can destroy a business. Too many Fintechs: ? Focus on acquiring “bad signals” ? Use AI as a single line of defense ? Obsess on studying fraud rings In the meantime, their conversion rates plummet. Here’s what I’ve learned managing fraud over 15 years: ? “Good signals” are not less important ? AI works well only when reinforced with rules ? Good customers are easy to spot The most expensive mistake? Trying to do everything at once! I’ve watched Fintechs feeding “good signals” to their fraud detection models, only to get no meaningful results. Let your AI figure out what’s fraud. Let your rules override it by looking for specific behaviors. Remember: Good customers follow patterns exactly like fraudsters do. Study and target them. You’ll see your conversion improve in no time. Do ?????? use designated rules to override declines? ------------ Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "Subscribe to newsletter" under my name ??

Unhappy with your fraud team’s accountability? Use my team performance framework: ?? ??♀? Team 1: Fraud Operations ? Primary KPI: Segment-based conversion rates ? Measurement method: Per flow/geography ? Success signals: Weekly improvement trends ? Team ownership: Rule optimization & case investigation ??♀? Team 2: Feature Engineering ? Primary KPI: Feature Impact Score ? Measurement method: Same model, new features ? Success signals: 3% AUC lift per quarter ? Team ownership: Data infrastructure & quality ??♀? Team 3: Modeling ? Primary KPI: Model AUC ? Measurement method: Same dataset, new algorithm ? Success signals: 5% AUC improvement YoY ? Team ownership: Algorithm design & training In my previous fraud org: Accountability score before: 40% Accountability score after: 95% Want the full breakdown? In this week's "The Saturday Fraud Strategist" I'm going to dive deep into: ? How to break down KPIs without breaking team spirit ? Why letting teams choose their approach drives better results ? The secret to measuring ML model improvements accurately If you're not subscribed yet, the link is under my picture. ?? See you on Saturday! ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

I've spent 15+ years in fraud prevention. My secret sauce? Less data science. I don't: ? Rely solely on ML models ? Trust raw chargeback data ? Let PhDs work in isolation ? Chase perfect fraud scores Instead: ? I combine domain expertise with data ? I question every fraud label ? I build cross-functional fraud teams ? I focus on understanding the 'why' behind the numbers A lack of chargebacks doesn't mean you're winning - it might just mean you're blind to the problem. I've seen clients celebrate 'zero fraud' – not realizing their 3DS was masking massive revenue loss. 9 times out of 10 it’s the result of: ? Copying someone else's fraud strategy ? Using out-of-the-box settings ? Letting someone else make decisions for you Which of these mistakes is costing you revenue right now? ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??

Every fraud team shares this uncomfortable truth: They might be costing you more than fraud itself. Here's what nobody talks about: False Positives run deeper than blocked payments ? The problem: 25% of declines are good customers ? The hidden cost: Lost LTV, not just transactions ? The real numbers: Equal to or higher than fraud losses The monitoring trap ? What teams try: Test rules before go-live ? Why it fails: Controls get stale but stay active ? The blind spot: No one measures compound impact The actual solution? Monitor False Positives across ALL controls holistically, not just individual rules. But that's just the start. I have a systematic framework that cuts False Positives by up to half. Want the full breakdown? In this week's "The Saturday Fraud Strategist" I'm going to dive deep into: ? Why False Positives cost more than your fraud losses ? The 4 hidden drivers behind most False Positive spikes ? 6 practical solutions you can implement this week ?? If you're not subscribed yet, the link is under my picture. ?? See you on Saturday! ----------- Enjoy this? ?? Repost it to your network and follow Chen Zamir for more. Want to dive deeper? Join "The Saturday Fraud Strategist". Each week I'll be sharing detailed advice on how to grow Fintechs safely. ?? Click "View my newsletter" under my name ??