Loops

It's Day 2 at #AdobeSummit – Let’s Talk Root Cause Analysis! ?? The energy at the Loops Booth 290B has been incredible! I’ve had the chance to meet so many analytics leaders looking to unlock deep causal insights - and with our seamless integration, Loops is proving to be the "Secret Weapon" for Adobe Analytics users. Today, I’d love for you to stop by and see our Root Cause Analysis in action. Loops doesn’t just surface anomalies -it pinpoints exactly why key metrics are changing, so you can act fast and drive impact. If you’re at #AdobeSummit, swing by Booth 290B - I’d love to chat! #AdobeAnalytics #DataDriven #RootCauseAnalysis #CausalAI #ProductAnalytics

Our team is in #Vegas this week, at #AdobeSummit excited to showcase Loops, one of the newest Adobe Technology Partners, at this exciting event. If you're there, Connect with Tom Laufer, Ellie Lagziel, and Jillian Bucci at Booth 290B!

Relying on Before-and-After analysis (pre post) to evaluate product changes seems simple - but it’s often misleading and can lead to incorrect decisions. Too many teams make decisions based on results that don’t account for seasonality, marketing campaigns, shifting user mix, errors, experiments, multiple releases, and other factors. At Loops, we put Before After analysis to the test. We took real-world A/B test data and ran it through both Before-and-After analyses and our own Loops Release Impact model. We then compared both sets of results to the A/B test. The results confirmed the risks of relying on Before-and-After analysis: ? Before and After Analysis only matched A/B test results 48.25% of the time ? Loops Release Impact Analysis delivered 90.3% average accuracy of A/B tests! ?? Our latest technical white paper, "Avoiding Before and After Analysis Fails - Lessons from Three Real-life Cases and How to Get it Right" by Iyar Lin, Data Science lead at Loops, dives deep into three real-life use cases. It highlights where before and after (pre post) analysis would have resulted in wrong decisions - and how advanced causal impact modeling came out very close to the respective standard A/B test result, every time. 1?? Case 1: A concurrent positive trend led to a false positive with Before and After- but actually, the release was a fail 2?? Case 2: A changing user mix completely skewed results - that would have led to a wrong decision with Before-and-After analysis. 3?? Case 3: Multiple releases created complexity and short test periods - Before and After analysis couldn’t keep up. If you are concerned about the accuracy of your data-driven decisions, product growth, and avoiding analytical blind spots, this white paper is for you. ?? Download now and level up your approach to measuring product impact ?? #DataScience #ProductAnalytics #CausalInference #ABTesting #Loops #CausalAI #BeforeandAfter #AIInsights

Retailers have no shortage of data - but are you surfacing the insights that truly matter? E-commerce leaders track AOV, ROAS, NPS, and churn, but knowing what’s changing isn’t enough—you need to know why. Traditional products analytics often leave teams reacting to trends instead of driving them. That’s where Loops comes in. Our AI-powered analytics platform helps large retailers uncover the real drivers behind KPI shifts and make data-backed decisions with confidence, with: 1?? Root Cause Analysis: Automatically identify the reasons behind fluctuations in key metrics such as Average Order Value (#AOV), Return on Ad Spend (#ROAS), Net Promoter Score (#NPS), and inventory turnover. This proactive approach allows you to address issues before they impact your bottom line. 2?? Real-Time Gen-AI Alerts Insight Summaries: Receive personalized alerts and insight updates on trends, anomalies, forecasts, and the impact of recent initiatives directly through Slack, Microsoft Teams, or email. This ensures your team stays informed and agile in responding to changes in your top KPI. 3?? Product Release Impact Analysis: Measure the effect of every product change on your KPIs with over 90% accuracy of standard A/B testing but with minimal traffic, time, and resources. Loops' causal models account for variables like performance improvements, marketing promotions, seasonality, pricing adjustments, experiments, product errors, and user mix changes, providing a clear view of each change's impact. 4?? User Journey Optimization: Identify and rank user paths that significantly influence your KPIs at every stage of the customer lifecycle. By understanding these journeys, you can optimize marketing strategies, landing pages, and the entire user funnel to drive conversions and retention. Proven Results with Loops: ?? ? 200% Increase in Conversions: Achieved through Loops' "User Journey" insights at Wahi Real Estate. ? $5 Million Revenue Saved: Through causal analysis of a core KPI drop at a major consumer goods retailer, enabling a partial release with a negative impact to be rolled back before it hit all users. ? 15% Increase in Day 2 Retention: Observed at 18Birdies, enhancing customer engagement and loyalty. Move beyond traditional dashboards, uncover hidden growth opportunities, and make data-driven decisions that propel your retail business forward. Discover how Loops can unlock your company's potential. #RetailAnalytics #AI #DataDrivenDecisionMaking #EcommerceGrowth #eCommerce #retail #CausalAI National Retail Federation, Shoptalk

?#GenAI + #AdobeAnalytics = @Loops.?Want to see how it all works? Our team is dropping major insights for Adobe Analytics users at Booth 290B next week! Don’t miss it. #AdobeSummit

Loops customer and friend, Marc Stone of ClickUp recently wrote a powerful Substack article (“AI Isn’t Lifting All Boats - The Sobering Truth About the Productivity Inequality of AI”) and related Linked post focused on how AI is supposed to empower teams but in reality, is creating a growing gap within productivity them. ?? In Product Analytics, we hypothesize that AI will help make analytics more accessible - allowing anyone to ask questions, generate insights, or even write SQL on demand. What impact will this have on the analysts? Marc refers to a Nov-2024 study (Artificial Intelligence, Scientific Discovery, and Product Innovation, by Aidan Toner-Rodgers, Massachusetts Institute of Technology) that explored how AI helped drive innovation within a team of scientists. They found that scientists using AI discovered more materials and filed more patents - but not everyone on the team benefited equally with AI. ?? ?? ?? The top scientists became twice as productive, while those who typically didn’t innovate as much saw little improvement. Why? Because AI automated much of the idea-finding process, it forced scientists to spend more time evaluating AI-generated suggestions. The best scientists used their expertise to prioritize good ideas, while others wasted time testing bad ones. The takeaway: ?? AI amplifies existing skills. If you’re already great at what you do, AI will make you even better. But analysts who don’t have the soft skills of critical thinking, skepticism, attention to detail, etc., or business context and domain expertise may not enjoy the benefits of AI. AI isn’t a magic wand - it’s an amplifier. ?? The teams who win with AI aren’t just automating work; they’re elevating their expertise. The future of product analytics belongs to those who combine AI’s speed with human intuition and strategic thinking. ? Is your analytics team bridging a business impact gap - or widening it? Let’s discuss. ?? #ProductAnalytics #AIAnalytics #GenAI #GenerativeAI

AI is rapidly transforming product analytics, shifting how insights are generated and how analysts drive impact. Instead of spending hours on manual data work, AI-powered tools are automating analysis and accelerating insights, reshaping the analyst’s role. ?? How AI Is Changing the Role of Analysts AI is already streamlining key areas for analysts. Repetitive and tedious analyst work of querying data, generating reports, and identifying surface-level trends is gradually being automated or made significantly more efficient through AI. For example, AI is now helping write SQL, monitor KPIs, build dashboards, and truly create self-served analytics that empower insights by simply asking questions in free languages to get insights. ?? What Analysts that thrive in the AI Era will focus on: The most critical aspects of analytics will remain human-led. The analysts who shift their focus to the following will thrive : ?? Leveraging domain expertise – The role is no longer 70-80% dashboard-building; it’s about having and leveraging domain expertise to critically review insight opportunities and form opinions about the most impactful. ?? Business context & strategy – The ability to leverage deep business context and form opinions. ?? Skepticism and Critical Thinking – AI can identify and produce a lot - but the best analysts will leverage a healthy level of skepticism and critical thinking because Gen AI is sometimes wrong. ?? Communication – Analysts must not underestimate the importance of effective communication to verify that insights are translated into actions. This skill will become crucial in the AI era. ?? Collaboration across teams – AI helps create insights while the analytics teams will focus on the highest-impact opportunities - where their strategic cross-functional collaboration with product, design, and marketing will be crucial. ?? Delivering strategic insights – AI highlights opportunities, but analysts will require skills in deep research and creativity and must leverage their unique domain expertise to deliver strategic insight that leads to new products/roadmaps. ?? Improving data quality and building semantic layers – AI is only as good as the data it’s trained on - analysts play a crucial role in ensuring accuracy. ?? AI won’t replace analysts - but the role is evolving fast. We're seeing companies changing their hiring process because of this, such as what they actually test for in the interviewing process. Analytics leaders should adjust the process and test other skills. The future belongs to those who embrace AI while sharpening the uniquely human skills that drive real impact. How do you see AI reshaping product analytics? Let’s discuss. ?? #ProductAnalytics #AIAnalytics #GenAI #GenerativeAI

Generative AI is reshaping analytics, making insights more accessible, and automating complex tasks. But while it’s unlocking new efficiencies, it also comes with challenges. Here’s where Gen-AI is making an impact - and where human expertise remains essential. Leveraging Gen-AI in Product Analytics: ?? Write SQL/Python for me – A popular use case of Gen AI today. Some companies report that AI-generated SQL reduces analyst time spent on SQL by 40%. Debugging can take time, and the business context of the output needs to be validated. ?? Ask Questions & Get Graphs – Business users can query data in plain language, getting visual insights instantly. Gen AI works well in 80+% of cases if you have a strong semantic layer. ?? Ask Questions in Free Language and Get Insights – A powerful use case, helping teams gain insights - such as why KPIs are going down, how many users utilize a particular feature, or why users are dropping in a certain step. ?? Surface Insights Proactively – Leverage Gen AI to automatically surface and deliver insights, proactively revealing high-impact opportunities you didn’t know, nor asked about. ?? Explain the Analysis – AI-generated narratives bridge the gap between data teams and stakeholders. ?? But...Context is King The key to success in leveraging AI in analytics is context. You should treat AI like a new team member - you need to train it using all your documentation, the KPI catalog, code comments, decks, etc. In many cases, AI is a mirror of the health of your analytics setup. If you do not have clear definitions, a solid semantic layer, and tracking you can trust, then the output of Gen AI will reflect poor context and not make sense. ?? What Gen-AI Can’t Replace At Loops, we see that Gen-AI is making a valuable contribution, helping automate repetitive, often tedious tasks like querying data, generating reports, and identifying surface-level trends - but the most critical aspects of analytics, those that require deep research and context, remain human-led. ?? A Pivotal Year Ahead In 2024, the retention for AI analytics features was very low. User expectations ran super high, but the underlying LLM technology was still evolving. With the current progress and given the right context, I believe 2025 will be a pivotal year in terms of both the use and value of leveraging AI analytics applications. ?? Stay tuned for my next post in this mini-series on Gen-AI, where I’ll cover what top analysts will focus on to thrive in this new era. #GenerativeAI #Analytics #CausalAI #DataDriven #ProductAnalytics #CausalML #AI #GenAI

One thing the fastest-growing companies have in common is that they run weekly and monthly business reviews. They stay ahead by NOT waiting for quarterly business reviews. They go deep, weekly and monthly, to understand why KPIs are changing and, as a result, what they need to do to meet their goals. ?? Instead of scrambling to calculate the impact of each factor on the change in the KPI, Loops customers leverage our platform to automatically look at their KPI relationships based on a KPI Tree and segments, marketing sources, experiments, releases and other factors that might have impacted the KPI, to deliver real answers - not just data points. ?? They’re saving a ton of time preparing for business reviews. We’ve heard that previously, they, and other companies, put 40% to 60% of their analytics resources into preparing for these reviews - digging into KPI shifts and trying to figure out why things are going up or down. The problem is that so many things are happening at the same time within big enterprises - there are so many options, and it can be hard for analytics teams to know why KPIs move. Also, much of that work is repetitive. Gen AI and causal analytics can help teams focus their time, allowing them to conduct deep research that delivers game-changing insights that affect the bottom line. ?? What if AI could also handle your Business Review research and preparation for you? You can with Loops' Gen AI-powered insights. You get a full summary - delivered as a PDF or deck - that breaks down: ? Your key metrics - what’s up, what’s down, and why ? Causal analysis - understanding what’s driving KPI changes ? Segment insights - which user groups or experiments are impacting performance ? Recommendations - that provide clear guidance on next steps ? Your next weekly review could be done for you before your coffee even cools. #AI #Analytics #ProductGrowth #LoopsAI #KPIInsights #CausalML #ProductAnalytics

When product KPIs shift - whether up or down - finding the real reason is often more complex than it seems. It’s rarely just one thing. Instead, it’s a web of interconnected factors that each have a different level of impact on the KPIs, and each requires deep analysis. Here are some of the questions I often see teams struggle with: ?? Experiment overload – You just ran 15 different A/B tests. Did one of them cause the shift, or is it something else entirely? ?? Seasonality & external factors – A holiday happened that week. Should you normalize for it, or is there an underlying trend you should act on? ?? KPI relationships – One metric dropped, but so did others. Is this a correlation, or is there an actual causal link? ?? Marketing impact – A campaign is driving a KPI shift, but why? What’s different this time? Most teams rely on dashboards and surface-level correlations to answer these questions, but they’re losing money while searching for the reason and wasting time on these repetitive processes. The real challenge is identifying the true cause - quantified, ranked, and backed by deep AI-powered causal analysis. Without this level of insight, you’re burning resources and making poor product and business decisions - when you could be focusing on the real drivers of change and knowing exactly what to do next to drive the business forward. How do you approach KPI shifts in your product? Let’s discuss. ?? #ProductAnalytics #DataDriven #AI #Growth #CausalML Loops