ABsmartly

ABsmartly is built API-first. 🔌 It might seem like a developer’s pipe dream to have an experimentation platform with open architecture. (Pun intended!) But with the AI revolution, this is when an integrations mindset really shines. Something that makes ABsmartly unique is that we don’t run to build every new feature that competitors offer. Why? Because it would fracture our focus AWAY from what we’ve learned ACTUALLY matters for teams to make better decisions at scale. (We know because we’ve lived it.) What do we focus on instead of bloating our product with features? 👉 strong technical foundations, 👉 statistical excellence, 👉 a templated approach to keep your experiments and data organized, 👉 decision-making guardrails to help democratize experimentation And most important of all? 👉 TRUE partnership to help our clients use the platform in a way that builds a culture of experimentation AROUND the tool. So, if you’re looking for an experimentation platform that’s future-proof and can evolve with your business—check us out.

With all the hype around AI, it's hard to tell what *is* and *is not* possible. That's why ABsmartly advisor, Erin Weigel, wrote a blog post to introduce the concept of an MCP server with some practical examples of what you can do with it. Useful things she covers in the post are... 👉 What’s an MCP Server? (And Why It Matters for Experimentation) 👉 Four AI Workflows You Can Build with ABsmartly’s MCP Server ... and more! Scroll through the post highlights in the carousel below. And check the comments for a link to read more!

Building a successful experimentation culture means hiring people with the right mindset. 🧠 But what character traits should you look for as you build your team? And how do you know your candidate has them? 🤔 In our most recent blog post our strategic advisor (and former hiring manager at Booking.com), Erin Weigel, shares five traits you should look for to build an experimentation culture like Booking.com's. Erin drops the honest truth that most people don't have the guts to put ALL OF THEIR WORK to a literal test. But to "ship less bad and MORE GOOD" (and important ethos in a healthy culture), two of the five traits to look for are: 1️⃣ Bravery & 2️⃣ Humility What are the other three? 👀 Go read the post to find out! 👇

Building a successful experimentation culture means hiring people with the right mindset. 🧠 But what character traits should you look for as you build your team? And how do you know your candidate has them? 🤔 In our most recent blog post our strategic advisor (and former hiring manager at Booking.com), Erin Weigel, shares five traits you should look for to build an experimentation culture like Booking.com's. Erin drops the honest truth that most people don't have the guts to put ALL OF THEIR WORK to a literal test. But to "ship less bad and MORE GOOD" (and important ethos in a healthy culture), two of the five traits to look for are: 1️⃣ Bravery & 2️⃣ Humility What are the other three? 👀 Go read the post to find out! 👇

Many people think that experimentation is about finding wins and validating great ideas. 💡 But that's only half the story. The real, counter-intuitive impact of a mature experimentation program isn't about WINNING MORE—it’s about MESSING UP LESS. When Booking.com analyzed a full year's worth of experiments, they uncovered a surprising truth: Preventing bugs and bad decisions had 10x more business impact than all of their winning ideas combined! 😲 Experimentation usually hits a wall because it takes commitment and corporate humility to test every code change into release. But treating experimentation as structural risk mitigation is how you protect (and grow!) your profit margins. Here's how: 🛑 You catch silent conversion killers. A simple feature rollout might look awesome, but real-time data monitoring shows you quickly if it spiked page load times or drastically decreased sales. 🎯 You get fast, causal data. Instead of guessing why a metric dropped three weeks after a release, you get immediate feedback tying the problem directly to a specific change. If you want your team to move fast and NOT break things, you need to design your release process for structural risk mitigation. ABsmartly can help you with that. 😄

Cloud-first or warehouse native? ⚖️ This choice often crops up when teams picks an experimentation platform. But there isn’t a single “right” approach. There's only the “right” choice for where your business is today. If you need to figure out which experimentation infrastructure fits your business, check out the carousel below. It breaks down the benefits and tradeoffs of both approaches: ☁️ Cloud-First gives you speed and simplicity. It's perfect for teams that need real-time monitoring and fast answers without managing warehouse infrastructure. 🏢 Warehouse native gives you precision and power. It's built for mature data teams that need absolute data privacy and perfect alignment with their internal business intelligence (BI) tools. The big news? You can now do both with ABsmartly. We just officially launched warehouse native support for BigQuery, Snowflake, Clickhouse, Redshift, and Databricks! 🎉

Less debates. Fewer alignment meetings. Zero time wasted fixing mismatched numbers. ✨ This is what happens when you stop arguing over numbers and move your testing logic into your data warehouse. Here's what you get when you go warehouse native: ⚖️ Metric Alignment Your experiment results neatly align with finance and executive reports because they stem from the same source. 🔒 Privacy & Security Your sensitive user-level data never leaves your systems, which makes SOC2, HIPAA, and GDPR compliance easier. 💰 Mature Overall Evaluation Criteria (OECs) Create and report on your own sophisticated business metrics to track and measure long-term customer and business value. But, is there a catch? Yep. Full warehouse native systems rely on batch processing. This means it can take anywhere from 15 minutes to 24 hours to see data come in. You trade real-time monitoring speed for total data consistency. For mature data teams care about precision, accuracy, and governance, that trade-off is worth it. This is why we just launched Warehouse Native support at ABsmartly. 🎉 You can now compute results inside your own warehouse environment while keeping our experiment management and advanced stats. (Redshift, Snowflake, Clickhouse, BigQuery, and Databricks are the warehouses we currently support.) This all said, we're happy that we now give customers both cloud first and warehouse native. However, we've got a vision for something better. And we're building that now. More to come on experimentation infrastructure with ABsmartly soon!

"One specific bug in Booking’s experimentation tooling accidentally invalidated three whole months of data. Imagine being an experimenter and finding out a quarter’s worth of your hard work was based on a mistake. That’s the kind of nightmare that causes people to completely check out. Learning from this experience, Jonas told Shiva in the podcast that, “It's actually not about the dashboards and the tool itself. It's about building trust. If you lose trust, people will not run experiments." From a great interview of Jonas Alves by Shiva Manjunath at the "From A to B" podcast. Read the rest of the transcript here: https://lnkd.in/d6JrBcC9 or listen to it on: YouTube: https://lnkd.in/d_rcYQVc Spotify: https://lnkd.in/dEmYpN9k Apple Podcasts: https://lnkd.in/dDkHhqdp

“Why am I giving you thousands of dollars just to randomize traffic?” Shiva Manjunath asked this question to our very own Jonas Alves on a recent episode of From A to B. Erin Weigel, our strategic advisor, loved Jonas's answer so much that she just dropped a blog post about their convo here: https://lnkd.in/eGvDw_XV As a bonus, Erin covers why Shiva brought LeBron James into the experimentation conversation. So, first read the blog. Then run—don't walk—to listen to the episode linked at the end of the post. 🏃♀️➡️ 🏃➡️ 🏃♂️➡️

Great conversation with our co-founder and CINO, Jonas Alves, on the experimentation podcast From A to B, and a refreshing take on what A/B testing is really about. It’s not just about splitting traffic. It’s about building trust in your decisions. A big thank you to Shiva Manjunath for the thoughtful questions and for making it such a fun conversation. If you care about experimentation, data credibility, and making better calls under uncertainty, this one’s worth your time 👇 https://lnkd.in/giAsQwTG