The Boring Pre-Work That Makes AI Actually Work for CRO

in this episode…Lucia chats with Divya Isaiah, head of experimentation at the BBC, a 100-year-old business with a weekly audience of half a billion people. Before that, she built a global CRO function covering 50 markets at Accenture Song and ran experimentation at Channel 4 and BA. She started in plant biology and marine biology, never planned a career in experimentation, and now leads a 14-person center of excellence supporting every BBC product, from iPlayer to News, Sport, Weather and Sounds.it’s about:→ how to build a center of excellence from a team of 1 to 14 inside a massive, decentralized organization→ why you start with your audience (the team), not the data, the tooling or the process→ speed vs. velocity: why the real shift is from chasing uplifts to making better decisions→ practicing what you preach: treating “we tried and it went nowhere” as a win→ what a business-wide experimentation summit with 400+ attendees taught her about engagement

in this episode…Margus Veeber is head of web at Pipedrive - the CRM used by over 100,000 businesses, and the host of the Web Unpacked Podcast. he leads global website strategy and has pushed his team from 3–5 tests a month to 10+ with the same headcount.his secret? treating AI as a teammate in every step of the process, with a human always leading.it’s about:→ how a team of 6 scaled from 3–5 tests a month to 10+ without adding people→ launching A/B tests straight from Claude Code, connected to the testing tool through MCPs→ the boring pre-work (your test repository) that decides whether AI gives you winners or slop→ why ugly websites can win, and how cool designs quietly tank conversion→ small, fast tests vs. big bets: how to keep velocity up while still swinging bigCheck out his Podcast page: ⁨ https://www.linkedin.com/company/webunpacked/home/​

in this episode…Muqtadaa Miandara was the senior manager of experimentation at Upbound Group, the company behind Rent-A-Center, Acima, and Bridget. he went from not knowing what tagging was on a Friday to acing a technical interview on Monday. and he's been figuring things out at speed ever since.this year? his team was targeting 300 tests, a massive valuation goal, and a leaderboard that makes ops teams compete for experimentation glory.it's about:→ how to go from 12 tests a month to 12 tests a week, without burning your team out→ the good, better, best investment pitch that got his CGO to say "I want the biggest one"→ why win rate is a smoke screen (and what to measure instead)→ the ideation workshop format that turns any team into experiment generators→ empathy, ethics, and context switching: the 3 traits every great experimenter needsthis episode was recorded while Muqtadaa was at Upbound Group. the views expressed are his own and do not represent those of Upbound Group.If you enjoy learning about these topics, follow our newsletter to get all the frameworks and insights: ⁠https://increaseconversionrate.substack.com/⁠

Get Simon's framework for scaling experimentation across product teams in this month's newsletter: ⁠https://increaseconversionrate.substack.com/⁠ in this episode…Simon Born is head of analytics and experimentation at Kaufland — Europe's fastest growing marketplace. he's been there six years, executed over 1,500 experiments himself, and right now has 80 experiments running simultaneously.he didn't build this from zero — but he scaled it to something most companies can only dream of.it's about:→ how to go from 2 experiments in parallel to 80 running at the same time→ the C-level move that got every team to start experimenting (without forcing it)→ why trustworthiness is the most underrated metric in any experimentation program→ build vs. buy: the tooling decision that shapes everything

Get Stuart's framework for building an experimentation process from scratch in this month's newsletter: ⁠https://increaseconversionrate.substack.com/⁠in this episode…Stuart Clare is the senior manager of digital trading, optimization, and personalization at the AA — one of the UK's most iconic brands. he's spent four years building an experimentation program from scratch, across five different product journeys, with a team of four.calm under pressure, process-obsessed, and still checking dashboards first thing every morning, just like he did when he was A/B testing email subject lines.it's about:→ how to build an experimentation program from nothing — and get leadership to fund it→ the loss prevention argument that makes any business case impossible to ignore→ why a 25% win rate is healthy (and why chasing a higher one is actually dangerous)→ app testing: why it's a completely different beast and what nobody tells you about it→ the 1 thing that has stayed consistent across every company, every product, every journey

Daniël Granja Baltazar⁠ is a CRO Specialist at one of the Netherlands' biggest telecoms VodafoneZiggo, and proving that CRO goes way beyond A/B testing.If you enjoy learning about these topics, follow our newsletter to get all the frameworks and insights: ⁠https://increaseconversionrate.substack.com/⁠in this episode…→ what a 300+ experiments per year experimentation program looks like→ why B2B experimentation can't just copy B2C playbooks→ the MVP mindset: stop falling in love with solutions, start defining the problem→ how qualitative research fills the gap when traffic is too low to test→ using sales knowledge to rewrite product pages (and 10x your leads)→ building an AI-powered experiment library so no learning ever gets lostFind Daniël on Linkedin: ⁠https://www.linkedin.com/in/dgbaltazar/

Gabriel Marçal⁠ is a growth PM at AIOS Medical, and one of the quietest, sharpest minds in experimentation. Solving bigger problems than just running A/B tests for checkouts.follow our newsletter to learn all his frameworks and insights: ⁠⁠https://increaseconversionrate.substack.com/⁠⁠in this episode…from his early days at a CRO agency in Brazil to running experiments on sales call scripts, physical medication boxes, and mobile apps, Gabriel refuses to stay in a box.it's about:→ why your website isn't the only surface worth testing (and what to do when you're blocked)→ how a lean 2-person team scaled to 10x growth without scaling headcount→ the no-show rate problem: how experimenting with sales reminders moved the needle from 70% to 80%+→ subscriptions as a retention game-changer — and why 90%+ adoption doesn't happen by accident→ Bayesian vs. frequentist? the answer that will surprise youFind Gabriel on Linkedin: https://www.linkedin.com/in/gabmarcal/

Get David's prioritization framework in this month's newsletter: ⁠https://increaseconversionrate.substack.com/⁠David Swinstead spent 10 years inside the Booking.com bubble, and came out with some of the sharpest, most contrarian takes in the experimentation space.in this episode…he is now running experimentation programs at e-clerks, and doing stand-up comedy on the side, he treats both the same way: test everything, surrender to the data, and never assume you know the answer.it's about:→ the "no-brainer" test on his first day at Booking.com that completely tanked→ why PIE, ICE, RICE and every off-the-shelf framework is setting your team up to fail→ the 5-criteria prioritization framework he's been refining for 15+ years→ 1 pixel. 1 underline. and why everything on your site can make a difference→ CRO is easy, people are hard. how to actually build a culture that lastsFind David on Linkedin: https://www.linkedin.com/in/david-swinstead-09267030/

Proven at brands like Kia, Montblanc, Blinkist, and Blacklane, Ioana’s framework shows how to implement experimentation that actually scales. Find it in this month’s newsletter: ⁠https://increaseconversionrate.substack.com/⁠in this episode...Ioana has built experimentation programs across industries (from automotive to luxury to tech) and she’s not just running tests, she’s building systems that scale learning. currently working as a product growth consultant at Blinkist, she turns experimentation into a true growth engine. this conversation isn’t about surface-level A/B tests. it’s about:→ the 5-step framework to build an experimentation program from scratch→ why “time to learn” beats perfection in high-performing teams→ how to avoid false positives that can cost companies millions→ why common “best practices” in CRO often fail in reality→ using AI to generate hypotheses and accelerate experimentationFind Ioana & her cohort for women on Linkedin: https://www.linkedin.com/in/ioanaiordache/

How do you pitch 50 experiments a month? ...by making them impossible to refuse. Steal Grace Miller’s framework for "tiny loss" proposals that guarantee a "Yes" from even the most risk-averse clients, it's in this month's newsletter: https://increaseconversionrate.substack.com/in this episode…Grace Miller is building one of the most advanced experimentation cultures in media.she doesn’t just run tests; she turns entire teams into experimenters.as head of experimentation and failure at Flight Story, the company behind Diary of a CEO, she’s scaling experimentation across 15+ teams, from thumbnails to full customer journeys.it’s about:how to build a culture where everyone runs experiments (not just one team)“kill the guesswork”: using data to drive decisions across the entire businesswhy failure, rewards, and rituals (like experimenter of the week) accelerate growthhow to prioritize experiments that actually impact company KPIsscaling learnings across brands, platforms, and audiences without forcing what doesn’t fitFind Grace on Linkedin: https://www.linkedin.com/in/grace-miller-linked-in/